refactor: embedded ibex (#2136)

README.md

@@ -88,13 +88,12 @@
 - 报告Bug，优先推荐提交[夜莺GitHub Issue](https://github.com/ccfos/nightingale/issues/new?assignees=&labels=kind%2Fbug&projects=&template=bug_report.yml)
 - 推荐完整浏览[夜莺文档站点](https://flashcat.cloud/docs/content/flashcat-monitor/nightingale-v7/introduction/)，了解更多信息
 - 推荐搜索关注夜莺公众号，第一时间获取社区动态：`夜莺监控Nightingale`
-- 日常问题交流：
-  - QQ群：730841964
-  - [加入微信群](https://download.flashcat.cloud/ulric/20241008153952.png)，如果二维码过期了，可以联系我（我的微信：`picobyte`）拉群，备注： `夜莺互助群`
+- 日常问题交流推荐加入[知识星球](https://download.flashcat.cloud/ulric/20240319095409.png)，也可以加我微信 `picobyte`，备注：`夜莺加群-<公司>-<姓名>` 拉入微信群，不过研发人员主要是关注 github issue 和星球，微信群关注较少
 
 ## 广受关注
 [![Stargazers over time](https://api.star-history.com/svg?repos=ccfos/nightingale&type=Date)](https://star-history.com/#ccfos/nightingale&Date)
 
+
 ## 社区共建
 - ❇️ 请阅读浏览[夜莺开源项目和社区治理架构草案](./doc/community-governance.md)，真诚欢迎每一位用户、开发者、公司以及组织，使用夜莺监控、积极反馈 Bug、提交功能需求、分享最佳实践，共建专业、活跃的夜莺开源社区。
 - ❤️ 夜莺贡献者

alert/aconf/conf.go

@@ -60,6 +60,10 @@ func (a *Alert) PreCheck(configDir string) {
 		a.Heartbeat.Interval = 1000
 	}
 
+	if a.Heartbeat.EngineName == "" {
+		a.Heartbeat.EngineName = "default"
+	}
+
 	if a.EngineDelay == 0 {
 		a.EngineDelay = 30
 	}

alert/alert.go

@@ -16,6 +16,7 @@ import (
 	"github.com/ccfos/nightingale/v6/alert/sender"
 	"github.com/ccfos/nightingale/v6/conf"
 	"github.com/ccfos/nightingale/v6/dumper"
+	"github.com/ccfos/nightingale/v6/ibex"
 	"github.com/ccfos/nightingale/v6/memsto"
 	"github.com/ccfos/nightingale/v6/models"
 	"github.com/ccfos/nightingale/v6/pkg/ctx"
@@ -26,8 +27,6 @@ import (
 	"github.com/ccfos/nightingale/v6/pushgw/writer"
 	"github.com/ccfos/nightingale/v6/storage"
 	"github.com/ccfos/nightingale/v6/tdengine"
-
-	"github.com/flashcatcloud/ibex/src/cmd/ibex"
 )
 
 func Initialize(configDir string, cryptoKey string) (func(), error) {
@@ -41,14 +40,14 @@ func Initialize(configDir string, cryptoKey string) (func(), error) {
 		return nil, err
 	}
 
-	ctx := ctx.NewContext(context.Background(), nil, false, config.CenterApi)
-
 	var redis storage.Redis
 	redis, err = storage.NewRedis(config.Redis)
 	if err != nil {
 		return nil, err
 	}
 
+	ctx := ctx.NewContext(context.Background(), nil, redis, false, config.CenterApi)
+
 	syncStats := memsto.NewSyncStats()
 	alertStats := astats.NewSyncStats()
 
@@ -74,13 +73,13 @@ func Initialize(configDir string, cryptoKey string) (func(), error) {
 	rt := router.New(config.HTTP, config.Alert, alertMuteCache, targetCache, busiGroupCache, alertStats, ctx, externalProcessors)
 
 	if config.Ibex.Enable {
-		ibex.ServerStart(false, nil, redis, config.HTTP.APIForService.BasicAuth, config.Alert.Heartbeat, &config.CenterApi, r, nil, config.Ibex, config.HTTP.Port)
+		ibex.ServerStart(ctx, false, nil, redis, config.HTTP.APIForService.BasicAuth, config.Alert.Heartbeat, &config.CenterApi, r, nil, config.Ibex, config.HTTP.Port)
 	}
 
 	rt.Config(r)
 	dumper.ConfigRouter(r)
 
-	httpClean := httpx.Init(config.HTTP, r)
+	httpClean := httpx.Init(config.HTTP, context.Background(), r)
 
 	return func() {
 		logxClean()

alert/dispatch/dispatch.go

@@ -286,7 +286,6 @@ func (e *Dispatch) SendCallbacks(rule *models.AlertRule, notifyTarget *NotifyTar
 
 	uids := notifyTarget.ToUidList()
 	urls := notifyTarget.ToCallbackList()
-	whMap := notifyTarget.ToWebhookMap()
 	for _, urlStr := range urls {
 		if len(urlStr) == 0 {
 			continue
@@ -294,11 +293,6 @@ func (e *Dispatch) SendCallbacks(rule *models.AlertRule, notifyTarget *NotifyTar
 
 		cbCtx := sender.BuildCallBackContext(e.ctx, urlStr, rule, []*models.AlertCurEvent{event}, uids, e.userCache, e.alerting.WebhookBatchSend, e.Astats)
 
-		if wh, ok := whMap[cbCtx.CallBackURL]; ok && wh.Enable {
-			logger.Debugf("SendCallbacks: webhook[%s] is in global conf.", cbCtx.CallBackURL)
-			continue
-		}
-
 		if strings.HasPrefix(urlStr, "${ibex}") {
 			e.CallBacks[models.IbexDomain].CallBack(cbCtx)
 			continue

alert/dispatch/notify_target.go

@@ -100,32 +100,8 @@ func (s *NotifyTarget) ToWebhookList() []*models.Webhook {
 	return webhooks
 }
 
-func (s *NotifyTarget) ToWebhookMap() map[string]*models.Webhook {
-	webhookMap := make(map[string]*models.Webhook, len(s.webhooks))
-	for _, wh := range s.webhooks {
-		if wh.Batch == 0 {
-			wh.Batch = 1000
-		}
-
-		if wh.Timeout == 0 {
-			wh.Timeout = 10
-		}
-
-		if wh.RetryCount == 0 {
-			wh.RetryCount = 10
-		}
-
-		if wh.RetryInterval == 0 {
-			wh.RetryInterval = 10
-		}
-
-		webhookMap[wh.Url] = wh
-	}
-	return webhookMap
-}
-
 func (s *NotifyTarget) ToUidList() []int64 {
-	uids := make([]int64, 0, len(s.userMap))
+	uids := make([]int64, len(s.userMap))
 	for uid, _ := range s.userMap {
 		uids = append(uids, uid)
 	}

alert/eval/eval.go

@@ -5,7 +5,6 @@ import (
 	"encoding/json"
 	"fmt"
 	"math"
-	"reflect"
 	"sort"
 	"strings"
 	"time"
@@ -268,7 +267,7 @@ func (arw *AlertRuleWorker) GetTdengineAnomalyPoint(rule *models.AlertRule, dsId
 	if len(ruleQuery.Queries) > 0 {
 		seriesStore := make(map[uint64]models.DataResp)
 		// 将不同查询的 hash 索引分组存放
-		seriesTagIndexes := make(map[string]map[uint64][]uint64)
+		seriesTagIndexes := make([]map[uint64][]uint64, 0)
 
 		for _, query := range ruleQuery.Queries {
 			seriesTagIndex := make(map[uint64][]uint64)
@@ -293,13 +292,7 @@ func (arw *AlertRuleWorker) GetTdengineAnomalyPoint(rule *models.AlertRule, dsId
 			//  此条日志很重要，是告警判断的现场值
 			logger.Debugf("rule_eval rid:%d req:%+v resp:%+v", rule.Id, query, series)
 			MakeSeriesMap(series, seriesTagIndex, seriesStore)
-			ref, err := GetQueryRef(query)
-			if err != nil {
-				logger.Warningf("rule_eval rid:%d query ref error: %v query:%+v", rule.Id, err, query)
-				arw.processor.Stats.CounterRuleEvalErrorTotal.WithLabelValues(fmt.Sprintf("%v", arw.processor.DatasourceId()), GET_RULE_CONFIG).Inc()
-				continue
-			}
-			seriesTagIndexes[ref] = seriesTagIndex
+			seriesTagIndexes = append(seriesTagIndexes, seriesTagIndex)
 		}
 
 		points, recoverPoints = GetAnomalyPoint(rule.Id, ruleQuery, seriesTagIndexes, seriesStore)
@@ -376,6 +369,11 @@ func (arw *AlertRuleWorker) GetHostAnomalyPoint(ruleConfig string) []common.Anom
 				}
 				m["ident"] = target.Ident
 
+				bg := arw.processor.BusiGroupCache.GetByBusiGroupId(target.GroupId)
+				if bg != nil && bg.LabelEnable == 1 {
+					m["busigroup"] = bg.LabelValue
+				}
+
 				lst = append(lst, common.NewAnomalyPoint(trigger.Type, m, now, float64(now-target.UpdateAt), trigger.Severity))
 			}
 		case "offset":
@@ -424,6 +422,11 @@ func (arw *AlertRuleWorker) GetHostAnomalyPoint(ruleConfig string) []common.Anom
 				}
 				m["ident"] = host
 
+				bg := arw.processor.BusiGroupCache.GetByBusiGroupId(target.GroupId)
+				if bg != nil && bg.LabelEnable == 1 {
+					m["busigroup"] = bg.LabelValue
+				}
+
 				lst = append(lst, common.NewAnomalyPoint(trigger.Type, m, now, float64(offset), trigger.Severity))
 			}
 		case "pct_target_miss":
@@ -452,7 +455,7 @@ func (arw *AlertRuleWorker) GetHostAnomalyPoint(ruleConfig string) []common.Anom
 	return lst
 }
 
-func GetAnomalyPoint(ruleId int64, ruleQuery models.RuleQuery, seriesTagIndexes map[string]map[uint64][]uint64, seriesStore map[uint64]models.DataResp) ([]common.AnomalyPoint, []common.AnomalyPoint) {
+func GetAnomalyPoint(ruleId int64, ruleQuery models.RuleQuery, seriesTagIndexes []map[uint64][]uint64, seriesStore map[uint64]models.DataResp) ([]common.AnomalyPoint, []common.AnomalyPoint) {
 	points := []common.AnomalyPoint{}
 	recoverPoints := []common.AnomalyPoint{}
 
@@ -460,13 +463,61 @@ func GetAnomalyPoint(ruleId int64, ruleQuery models.RuleQuery, seriesTagIndexes
 		return points, recoverPoints
 	}
 
-	if len(seriesTagIndexes) == 0 {
-		return points, recoverPoints
-	}
-
 	for _, trigger := range ruleQuery.Triggers {
 		// seriesTagIndex 的 key 仅做分组使用，value 为每组 series 的 hash
-		seriesTagIndex := ProcessJoins(ruleId, trigger, seriesTagIndexes, seriesStore)
+		seriesTagIndex := make(map[uint64][]uint64)
+
+		if len(trigger.Joins) == 0 {
+			// 没有 join 条件，走原逻辑
+			last := seriesTagIndexes[0]
+			for i := 1; i < len(seriesTagIndexes); i++ {
+				last = originalJoin(last, seriesTagIndexes[i])
+			}
+			seriesTagIndex = last
+		} else {
+			// 有 join 条件，按条件依次合并
+			if len(seriesTagIndexes) != len(trigger.Joins)+1 {
+				logger.Errorf("rule_eval rid:%d queries' count: %d not match join condition's count: %d", ruleId, len(seriesTagIndexes), len(trigger.Joins))
+				continue
+			}
+
+			last := seriesTagIndexes[0]
+			lastRehashed := rehashSet(last, seriesStore, trigger.Joins[0].On)
+			for i := range trigger.Joins {
+				cur := seriesTagIndexes[i+1]
+				switch trigger.Joins[i].JoinType {
+				case "original":
+					last = originalJoin(last, cur)
+				case "none":
+					last = noneJoin(last, cur)
+				case "cartesian":
+					last = cartesianJoin(last, cur)
+				case "inner_join":
+					curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
+					lastRehashed = onJoin(lastRehashed, curRehashed, Inner)
+					last = flatten(lastRehashed)
+				case "left_join":
+					curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
+					lastRehashed = onJoin(lastRehashed, curRehashed, Left)
+					last = flatten(lastRehashed)
+				case "right_join":
+					curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
+					lastRehashed = onJoin(curRehashed, lastRehashed, Right)
+					last = flatten(lastRehashed)
+				case "left_exclude":
+					curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
+					lastRehashed = exclude(lastRehashed, curRehashed)
+					last = flatten(lastRehashed)
+				case "right_exclude":
+					curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
+					lastRehashed = exclude(curRehashed, lastRehashed)
+					last = flatten(lastRehashed)
+				default:
+					logger.Warningf("rule_eval rid:%d join type:%s not support", ruleId, trigger.Joins[i].JoinType)
+				}
+			}
+			seriesTagIndex = last
+		}
 
 		for _, seriesHash := range seriesTagIndex {
 			sort.Slice(seriesHash, func(i, j int) bool {
@@ -562,7 +613,7 @@ func flatten(rehashed map[uint64][][]uint64) map[uint64][]uint64 {
 // [[A3{data_base=2, table=board}，B2{data_base=2, table=alert}]，[A4{data_base=2, table=alert}，B2{data_base=2, table=alert}]]
 func onJoin(reHashTagIndex1 map[uint64][][]uint64, reHashTagIndex2 map[uint64][][]uint64, joinType JoinType) map[uint64][][]uint64 {
 	reHashTagIndex := make(map[uint64][][]uint64)
-	for rehash := range reHashTagIndex1 {
+	for rehash, _ := range reHashTagIndex1 {
 		if _, ok := reHashTagIndex2[rehash]; ok {
 			// 若有 rehash 相同的记录，两两合并
 			for i1 := range reHashTagIndex1[rehash] {
@@ -605,7 +656,6 @@ func rehashSet(seriesTagIndex1 map[uint64][]uint64, seriesStore map[uint64]model
 		if !exists {
 			continue
 		}
-
 		rehash := hash.GetTargetTagHash(series.Metric, on)
 		if _, ok := reHashTagIndex[rehash]; !ok {
 			reHashTagIndex[rehash] = make([][]uint64, 0)
@@ -697,100 +747,3 @@ func mergeNewArray(arg ...[]uint64) []uint64 {
 	}
 	return res
 }
-
-func ProcessJoins(ruleId int64, trigger models.Trigger, seriesTagIndexes map[string]map[uint64][]uint64, seriesStore map[uint64]models.DataResp) map[uint64][]uint64 {
-	last := make(map[uint64][]uint64)
-	if len(seriesTagIndexes) == 0 {
-		return last
-	}
-
-	if len(trigger.Joins) == 0 {
-		idx := 0
-		for _, seriesTagIndex := range seriesTagIndexes {
-			if idx == 0 {
-				last = seriesTagIndex
-			} else {
-				last = originalJoin(last, seriesTagIndex)
-			}
-			idx++
-		}
-		return last
-	}
-
-	// 有 join 条件，按条件依次合并
-	if len(seriesTagIndexes) < len(trigger.Joins)+1 {
-		logger.Errorf("rule_eval rid:%d queries' count: %d not match join condition's count: %d", ruleId, len(seriesTagIndexes), len(trigger.Joins))
-		return nil
-	}
-
-	last = seriesTagIndexes[trigger.JoinRef]
-	lastRehashed := rehashSet(last, seriesStore, trigger.Joins[0].On)
-	for i := range trigger.Joins {
-		cur := seriesTagIndexes[trigger.Joins[i].Ref]
-		switch trigger.Joins[i].JoinType {
-		case "original":
-			last = originalJoin(last, cur)
-		case "none":
-			last = noneJoin(last, cur)
-		case "cartesian":
-			last = cartesianJoin(last, cur)
-		case "inner_join":
-			curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
-			lastRehashed = onJoin(lastRehashed, curRehashed, Inner)
-			last = flatten(lastRehashed)
-		case "left_join":
-			curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
-			lastRehashed = onJoin(lastRehashed, curRehashed, Left)
-			last = flatten(lastRehashed)
-		case "right_join":
-			curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
-			lastRehashed = onJoin(curRehashed, lastRehashed, Right)
-			last = flatten(lastRehashed)
-		case "left_exclude":
-			curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
-			lastRehashed = exclude(lastRehashed, curRehashed)
-			last = flatten(lastRehashed)
-		case "right_exclude":
-			curRehashed := rehashSet(cur, seriesStore, trigger.Joins[i].On)
-			lastRehashed = exclude(curRehashed, lastRehashed)
-			last = flatten(lastRehashed)
-		default:
-			logger.Warningf("rule_eval rid:%d join type:%s not support", ruleId, trigger.Joins[i].JoinType)
-		}
-	}
-	return last
-}
-
-func GetQueryRef(query interface{}) (string, error) {
-	// 首先检查是否为 map
-	if m, ok := query.(map[string]interface{}); ok {
-		if ref, exists := m["ref"]; exists {
-			if refStr, ok := ref.(string); ok {
-				return refStr, nil
-			}
-			return "", fmt.Errorf("ref 字段不是字符串类型")
-		}
-		return "", fmt.Errorf("query 中没有找到 ref 字段")
-	}
-
-	// 如果不是 map，则按原来的方式处理结构体
-	v := reflect.ValueOf(query)
-	if v.Kind() == reflect.Ptr {
-		v = v.Elem()
-	}
-
-	if v.Kind() != reflect.Struct {
-		return "", fmt.Errorf("query not a struct or map")
-	}
-
-	refField := v.FieldByName("Ref")
-	if !refField.IsValid() {
-		return "", fmt.Errorf("not find ref field")
-	}
-
-	if refField.Kind() != reflect.String {
-		return "", fmt.Errorf("ref not a string")
-	}
-
-	return refField.String(), nil
-}

alert/mute/mute.go

@@ -114,7 +114,7 @@ func BgNotMatchMuteStrategy(rule *models.AlertRule, event *models.AlertCurEvent,
 	target, exists := targetCache.Get(ident)
 	// 对于包含ident的告警事件，check一下ident所属bg和rule所属bg是否相同
 	// 如果告警规则选择了只在本BG生效，那其他BG的机器就不能因此规则产生告警
-	if exists && !target.MatchGroupId(rule.GroupId) {
+	if exists && target.GroupId != rule.GroupId {
 		logger.Debugf("[%s] mute: rule_eval:%d cluster:%s", "BgNotMatchMuteStrategy", rule.Id, event.Cluster)
 		return true
 	}

alert/process/process.go

@@ -212,14 +212,6 @@ func (p *Processor) BuildEvent(anomalyPoint common.AnomalyPoint, from string, no
 	event.ExtraConfig = p.rule.ExtraConfigJSON
 	event.PromQl = anomalyPoint.Query
 
-	if p.target != "" {
-		if pt, exist := p.TargetCache.Get(p.target); exist {
-			event.Target = pt
-		} else {
-			logger.Infof("Target[ident: %s] doesn't exist in cache.", p.target)
-		}
-	}
-
 	if event.TriggerValues != "" && strings.Count(event.TriggerValues, "$") > 1 {
 		// TriggerValues 有多个变量，将多个变量都放到 TriggerValue 中
 		event.TriggerValue = event.TriggerValues
@@ -505,7 +497,6 @@ func (p *Processor) pushEventToQueue(e *models.AlertCurEvent) {
 
 func (p *Processor) RecoverAlertCurEventFromDb() {
 	p.pendings = NewAlertCurEventMap(nil)
-	p.pendingsUseByRecover = NewAlertCurEventMap(nil)
 
 	curEvents, err := models.AlertCurEventGetByRuleIdAndDsId(p.ctx, p.rule.Id, p.datasourceId)
 	if err != nil {
@@ -527,11 +518,6 @@ func (p *Processor) RecoverAlertCurEventFromDb() {
 		}
 
 		event.DB2Mem()
-		target, exists := p.TargetCache.Get(event.TargetIdent)
-		if exists {
-			event.Target = target
-		}
-
 		fireMap[event.Hash] = event
 		e := *event
 		pendingsUseByRecoverMap[event.Hash] = &e

alert/record/prom_rule.go

@@ -56,12 +56,11 @@ func (rrc *RecordRuleContext) Key() string {
 }
 
 func (rrc *RecordRuleContext) Hash() string {
-	return str.MD5(fmt.Sprintf("%d_%s_%s_%d_%s",
+	return str.MD5(fmt.Sprintf("%d_%s_%s_%d",
 		rrc.rule.Id,
 		rrc.rule.CronPattern,
 		rrc.rule.PromQl,
 		rrc.datasourceId,
-		rrc.rule.AppendTags,
 	))
 }
 

alert/router/router_event.go

@@ -34,7 +34,7 @@ func (rt *Router) pushEventToQueue(c *gin.Context) {
 			continue
 		}
 
-		arr := strings.SplitN(pair, "=", 2)
+		arr := strings.Split(pair, "=")
 		if len(arr) != 2 {
 			continue
 		}

alert/sender/feishucard.go

@@ -56,8 +56,8 @@ const (
 	Triggered = "triggered"
 )
 
-func createFeishuCardBody() feishuCard {
-	return feishuCard{
+var (
+	body = feishuCard{
 		feishu: feishu{Msgtype: "interactive"},
 		Card: Cards{
 			Config: Conf{
@@ -90,7 +90,7 @@ func createFeishuCardBody() feishuCard {
 			},
 		},
 	}
-}
+)
 
 func (fs *FeishuCardSender) CallBack(ctx CallBackContext) {
 	if len(ctx.Events) == 0 || len(ctx.CallBackURL) == 0 {
@@ -121,7 +121,6 @@ func (fs *FeishuCardSender) CallBack(ctx CallBackContext) {
 	}
 
 	SendTitle := fmt.Sprintf("🔔 %s", ctx.Events[0].RuleName)
-	body := createFeishuCardBody()
 	body.Card.Header.Title.Content = SendTitle
 	body.Card.Header.Template = color
 	body.Card.Elements[0].Text.Content = message
@@ -154,7 +153,6 @@ func (fs *FeishuCardSender) Send(ctx MessageContext) {
 	}
 
 	SendTitle := fmt.Sprintf("🔔 %s", ctx.Events[0].RuleName)
-	body := createFeishuCardBody()
 	body.Card.Header.Title.Content = SendTitle
 	body.Card.Header.Template = color
 	body.Card.Elements[0].Text.Content = message

alert/sender/ibex.go

@@ -12,8 +12,7 @@ import (
 	"github.com/ccfos/nightingale/v6/memsto"
 	"github.com/ccfos/nightingale/v6/models"
 	"github.com/ccfos/nightingale/v6/pkg/ctx"
-	imodels "github.com/flashcatcloud/ibex/src/models"
-	"github.com/flashcatcloud/ibex/src/storage"
+	"github.com/ccfos/nightingale/v6/storage"
 
 	"github.com/toolkits/pkg/logger"
 )
@@ -43,7 +42,7 @@ func (c *IbexCallBacker) CallBack(ctx CallBackContext) {
 }
 
 func (c *IbexCallBacker) handleIbex(ctx *ctx.Context, url string, event *models.AlertCurEvent) {
-	if imodels.DB() == nil && ctx.IsCenter {
+	if models.DB(ctx) == nil && ctx.IsCenter {
 		logger.Warning("event_callback_ibex: db is nil")
 		return
 	}
@@ -108,7 +107,7 @@ func CallIbex(ctx *ctx.Context, id int64, host string,
 			continue
 		}
 
-		arr := strings.SplitN(pair, "=", 2)
+		arr := strings.Split(pair, "=")
 		if len(arr) != 2 {
 			continue
 		}
@@ -142,7 +141,7 @@ func CallIbex(ctx *ctx.Context, id int64, host string,
 		AlertTriggered: true,
 	}
 
-	id, err = TaskAdd(in, tpl.UpdateBy, ctx.IsCenter)
+	id, err = TaskAdd(ctx, in, tpl.UpdateBy, ctx.IsCenter)
 	if err != nil {
 		logger.Errorf("event_callback_ibex: call ibex fail: %v", err)
 		return
@@ -181,16 +180,16 @@ func canDoIbex(username string, tpl *models.TaskTpl, host string, targetCache *m
 		return false, nil
 	}
 
-	return target.MatchGroupId(tpl.GroupId), nil
+	return target.GroupId == tpl.GroupId, nil
 }
 
-func TaskAdd(f models.TaskForm, authUser string, isCenter bool) (int64, error) {
+func TaskAdd(ctx *ctx.Context, f models.TaskForm, authUser string, isCenter bool) (int64, error) {
 	hosts := cleanHosts(f.Hosts)
 	if len(hosts) == 0 {
 		return 0, fmt.Errorf("arg(hosts) empty")
 	}
 
-	taskMeta := &imodels.TaskMeta{
+	taskMeta := &models.TaskMeta{
 		Title:     f.Title,
 		Account:   f.Account,
 		Batch:     f.Batch,
@@ -213,34 +212,34 @@ func TaskAdd(f models.TaskForm, authUser string, isCenter bool) (int64, error) {
 	// 任务类型分为"告警规则触发"和"n9e center用户下发"两种；
 	// 边缘机房"告警规则触发"的任务不需要规划，并且它可能是失联的，无法使用db资源，所以放入redis缓存中，直接下发给agentd执行
 	if !isCenter && f.AlertTriggered {
-		if err := taskMeta.Create(); err != nil {
+		if err := taskMeta.Create(ctx); err != nil {
 			// 当网络不连通时，生成唯一的id，防止边缘机房中不同任务的id相同；
 			// 方法是，redis自增id去防止同一个机房的不同n9e edge生成的id相同；
 			// 但没法防止不同边缘机房生成同样的id，所以，生成id的数据不会上报存入数据库，只用于闭环执行。
-			taskMeta.Id, err = storage.IdGet()
+			taskMeta.Id, err = storage.IdGet(ctx.Redis)
 			if err != nil {
 				return 0, err
 			}
 		}
 
-		taskHost := imodels.TaskHost{
+		taskHost := models.TaskHost{
 			Id:     taskMeta.Id,
 			Host:   hosts[0],
 			Status: "running",
 		}
-		if err = taskHost.Create(); err != nil {
+		if err = taskHost.Create(ctx); err != nil {
 			logger.Warningf("task_add_fail: authUser=%s title=%s err=%s", authUser, taskMeta.Title, err.Error())
 		}
 
 		// 缓存任务元信息和待下发的任务
-		err = taskMeta.Cache(hosts[0])
+		err = taskMeta.Cache(ctx, hosts[0])
 		if err != nil {
 			return 0, err
 		}
 
 	} else {
 		// 如果是中心机房，还是保持之前的逻辑
-		err = taskMeta.Save(hosts, f.Action)
+		err = taskMeta.Save(ctx, hosts, f.Action)
 		if err != nil {
 			return 0, err
 		}

alert/sender/larkcard.go

@@ -42,7 +42,6 @@ func (fs *LarkCardSender) CallBack(ctx CallBackContext) {
 	}
 
 	SendTitle := fmt.Sprintf("🔔 %s", ctx.Events[0].RuleName)
-	body := createFeishuCardBody()
 	body.Card.Header.Title.Content = SendTitle
 	body.Card.Header.Template = color
 	body.Card.Elements[0].Text.Content = message
@@ -75,7 +74,6 @@ func (fs *LarkCardSender) Send(ctx MessageContext) {
 	}
 
 	SendTitle := fmt.Sprintf("🔔 %s", ctx.Events[0].RuleName)
-	body := createFeishuCardBody()
 	body.Card.Header.Title.Content = SendTitle
 	body.Card.Header.Template = color
 	body.Card.Elements[0].Text.Content = message

alert/sender/webhook.go

@@ -166,10 +166,10 @@ func StartConsumer(ctx *ctx.Context, queue *WebhookQueue, popSize int, webhook *
 			retryCount := 0
 			for retryCount < webhook.RetryCount {
 				needRetry, res, err := sendWebhook(webhook, events, stats)
-				go RecordEvents(ctx, webhook, events, stats, res, err)
 				if !needRetry {
 					break
 				}
+				go RecordEvents(ctx, webhook, events, stats, res, err)
 				retryCount++
 				time.Sleep(time.Second * time.Duration(webhook.RetryInterval) * time.Duration(retryCount))
 			}

center/cconf/conf.go

@@ -14,7 +14,6 @@ type Center struct {
 	FlashDuty              FlashDuty
 	EventHistoryGroupView  bool
 	CleanNotifyRecordDay   int
-	MigrateBusiGroupLabel  bool
 }
 
 type Plugin struct {

center/center.go

@@ -18,6 +18,7 @@ import (
 	"github.com/ccfos/nightingale/v6/conf"
 	"github.com/ccfos/nightingale/v6/cron"
 	"github.com/ccfos/nightingale/v6/dumper"
+	"github.com/ccfos/nightingale/v6/ibex"
 	"github.com/ccfos/nightingale/v6/memsto"
 	"github.com/ccfos/nightingale/v6/models"
 	"github.com/ccfos/nightingale/v6/models/migrate"
@@ -33,8 +34,6 @@ import (
 	"github.com/ccfos/nightingale/v6/pushgw/writer"
 	"github.com/ccfos/nightingale/v6/storage"
 	"github.com/ccfos/nightingale/v6/tdengine"
-
-	"github.com/flashcatcloud/ibex/src/cmd/ibex"
 )
 
 func Initialize(configDir string, cryptoKey string) (func(), error) {
@@ -61,7 +60,14 @@ func Initialize(configDir string, cryptoKey string) (func(), error) {
 	if err != nil {
 		return nil, err
 	}
-	ctx := ctx.NewContext(context.Background(), db, true)
+
+	var redis storage.Redis
+	redis, err = storage.NewRedis(config.Redis)
+	if err != nil {
+		return nil, err
+	}
+
+	ctx := ctx.NewContext(context.Background(), db, redis, true)
 	migrate.Migrate(db)
 	models.InitRoot(ctx)
 
@@ -72,12 +78,7 @@ func Initialize(configDir string, cryptoKey string) (func(), error) {
 		return nil, err
 	}
 
-	go integration.Init(ctx, config.Center.BuiltinIntegrationsDir)
-	var redis storage.Redis
-	redis, err = storage.NewRedis(config.Redis)
-	if err != nil {
-		return nil, err
-	}
+	integration.Init(ctx, config.Center.BuiltinIntegrationsDir)
 
 	metas := metas.New(redis)
 	idents := idents.New(ctx, redis)
@@ -110,13 +111,10 @@ func Initialize(configDir string, cryptoKey string) (func(), error) {
 	go cron.CleanNotifyRecord(ctx, config.Center.CleanNotifyRecordDay)
 
 	alertrtRouter := alertrt.New(config.HTTP, config.Alert, alertMuteCache, targetCache, busiGroupCache, alertStats, ctx, externalProcessors)
-	centerRouter := centerrt.New(config.HTTP, config.Center, config.Alert, config.Ibex,
-		cconf.Operations, dsCache, notifyConfigCache, promClients, tdengineClients,
+	centerRouter := centerrt.New(config.HTTP, config.Center, config.Alert, config.Ibex, cconf.Operations, dsCache, notifyConfigCache, promClients, tdengineClients,
 		redis, sso, ctx, metas, idents, targetCache, userCache, userGroupCache)
 	pushgwRouter := pushgwrt.New(config.HTTP, config.Pushgw, config.Alert, targetCache, busiGroupCache, idents, metas, writers, ctx)
 
-	go models.MigrateBg(ctx, pushgwRouter.Pushgw.BusiGroupLabelKey)
-
 	r := httpx.GinEngine(config.Global.RunMode, config.HTTP)
 
 	centerRouter.Config(r)
@@ -126,10 +124,10 @@ func Initialize(configDir string, cryptoKey string) (func(), error) {
 
 	if config.Ibex.Enable {
 		migrate.MigrateIbexTables(db)
-		ibex.ServerStart(true, db, redis, config.HTTP.APIForService.BasicAuth, config.Alert.Heartbeat, &config.CenterApi, r, centerRouter, config.Ibex, config.HTTP.Port)
+		ibex.ServerStart(ctx, true, db, redis, config.HTTP.APIForService.BasicAuth, config.Alert.Heartbeat, &config.CenterApi, r, centerRouter, config.Ibex, config.HTTP.Port)
 	}
 
-	httpClean := httpx.Init(config.HTTP, r)
+	httpClean := httpx.Init(config.HTTP, context.Background(), r)
 
 	return func() {
 		logxClean()

center/integration/init.go

@@ -16,12 +16,6 @@ import (
 const SYSTEM = "system"
 
 func Init(ctx *ctx.Context, builtinIntegrationsDir string) {
-	err := models.InitBuiltinPayloads(ctx)
-	if err != nil {
-		logger.Warning("init old builtinPayloads fail ", err)
-		return
-	}
-
 	fp := builtinIntegrationsDir
 	if fp == "" {
 		fp = path.Join(runner.Cwd, "integrations")
@@ -98,7 +92,6 @@ func Init(ctx *ctx.Context, builtinIntegrationsDir string) {
 					logger.Warning("update builtin component fail ", old, err)
 				}
 			}
-			component.ID = old.ID
 		}
 
 		// delete uuid is emtpy
@@ -148,13 +141,13 @@ func Init(ctx *ctx.Context, builtinIntegrationsDir string) {
 
 					cate := strings.Replace(f, ".json", "", -1)
 					builtinAlert := models.BuiltinPayload{
-						ComponentID: component.ID,
-						Type:        "alert",
-						Cate:        cate,
-						Name:        alert.Name,
-						Tags:        alert.AppendTags,
-						Content:     string(content),
-						UUID:        alert.UUID,
+						Component: component.Ident,
+						Type:      "alert",
+						Cate:      cate,
+						Name:      alert.Name,
+						Tags:      alert.AppendTags,
+						Content:   string(content),
+						UUID:      alert.UUID,
 					}
 
 					old, err := models.BuiltinPayloadGet(ctx, "uuid = ?", alert.UUID)
@@ -172,7 +165,6 @@ func Init(ctx *ctx.Context, builtinIntegrationsDir string) {
 					}
 
 					if old.UpdatedBy == SYSTEM {
-						old.ComponentID = component.ID
 						old.Content = string(content)
 						old.Name = alert.Name
 						old.Tags = alert.AppendTags
@@ -239,13 +231,13 @@ func Init(ctx *ctx.Context, builtinIntegrationsDir string) {
 				}
 
 				builtinDashboard := models.BuiltinPayload{
-					ComponentID: component.ID,
-					Type:        "dashboard",
-					Cate:        "",
-					Name:        dashboard.Name,
-					Tags:        dashboard.Tags,
-					Content:     string(content),
-					UUID:        dashboard.UUID,
+					Component: component.Ident,
+					Type:      "dashboard",
+					Cate:      "",
+					Name:      dashboard.Name,
+					Tags:      dashboard.Tags,
+					Content:   string(content),
+					UUID:      dashboard.UUID,
 				}
 
 				old, err := models.BuiltinPayloadGet(ctx, "uuid = ?", dashboard.UUID)
@@ -263,7 +255,6 @@ func Init(ctx *ctx.Context, builtinIntegrationsDir string) {
 				}
 
 				if old.UpdatedBy == SYSTEM {
-					old.ComponentID = component.ID
 					old.Content = string(content)
 					old.Name = dashboard.Name
 					old.Tags = dashboard.Tags

center/router/router.go

@@ -16,7 +16,6 @@ import (
 	"github.com/ccfos/nightingale/v6/conf"
 	_ "github.com/ccfos/nightingale/v6/front/statik"
 	"github.com/ccfos/nightingale/v6/memsto"
-	"github.com/ccfos/nightingale/v6/models"
 	"github.com/ccfos/nightingale/v6/pkg/aop"
 	"github.com/ccfos/nightingale/v6/pkg/ctx"
 	"github.com/ccfos/nightingale/v6/pkg/httpx"
@@ -52,14 +51,9 @@ type Router struct {
 	UserGroupCache    *memsto.UserGroupCacheType
 	Ctx               *ctx.Context
 	HeartbeatHook     HeartbeatHookFunc
-	TargetDeleteHook  models.TargetDeleteHookFunc
 }
 
-func New(httpConfig httpx.Config, center cconf.Center, alert aconf.Alert, ibex conf.Ibex,
-	operations cconf.Operation, ds *memsto.DatasourceCacheType, ncc *memsto.NotifyConfigCacheType,
-	pc *prom.PromClientMap, tdendgineClients *tdengine.TdengineClientMap, redis storage.Redis,
-	sso *sso.SsoClient, ctx *ctx.Context, metaSet *metas.Set, idents *idents.Set,
-	tc *memsto.TargetCacheType, uc *memsto.UserCacheType, ugc *memsto.UserGroupCacheType) *Router {
+func New(httpConfig httpx.Config, center cconf.Center, alert aconf.Alert, ibex conf.Ibex, operations cconf.Operation, ds *memsto.DatasourceCacheType, ncc *memsto.NotifyConfigCacheType, pc *prom.PromClientMap, tdendgineClients *tdengine.TdengineClientMap, redis storage.Redis, sso *sso.SsoClient, ctx *ctx.Context, metaSet *metas.Set, idents *idents.Set, tc *memsto.TargetCacheType, uc *memsto.UserCacheType, ugc *memsto.UserGroupCacheType) *Router {
 	return &Router{
 		HTTP:              httpConfig,
 		Center:            center,
@@ -79,14 +73,9 @@ func New(httpConfig httpx.Config, center cconf.Center, alert aconf.Alert, ibex c
 		UserGroupCache:    ugc,
 		Ctx:               ctx,
 		HeartbeatHook:     func(ident string) map[string]interface{} { return nil },
-		TargetDeleteHook:  emptyDeleteHook,
 	}
 }
 
-func emptyDeleteHook(ctx *ctx.Context, idents []string) error {
-	return nil
-}
-
 func stat() gin.HandlerFunc {
 	return func(c *gin.Context) {
 		start := time.Now()
@@ -287,7 +276,7 @@ func (rt *Router) Config(r *gin.Engine) {
 		pages.POST("/targets/tags", rt.auth(), rt.user(), rt.perm("/targets/put"), rt.targetBindTagsByFE)
 		pages.DELETE("/targets/tags", rt.auth(), rt.user(), rt.perm("/targets/put"), rt.targetUnbindTagsByFE)
 		pages.PUT("/targets/note", rt.auth(), rt.user(), rt.perm("/targets/put"), rt.targetUpdateNote)
-		pages.PUT("/targets/bgids", rt.auth(), rt.user(), rt.perm("/targets/put"), rt.targetBindBgids)
+		pages.PUT("/targets/bgid", rt.auth(), rt.user(), rt.perm("/targets/put"), rt.targetUpdateBgid)
 
 		pages.POST("/builtin-cate-favorite", rt.auth(), rt.user(), rt.builtinCateFavoriteAdd)
 		pages.DELETE("/builtin-cate-favorite/:name", rt.auth(), rt.user(), rt.builtinCateFavoriteDel)
@@ -308,7 +297,6 @@ func (rt *Router) Config(r *gin.Engine) {
 		pages.POST("/busi-group/:id/board/:bid/clone", rt.auth(), rt.user(), rt.perm("/dashboards/add"), rt.bgrw(), rt.boardClone)
 		pages.POST("/busi-groups/boards/clones", rt.auth(), rt.user(), rt.perm("/dashboards/add"), rt.boardBatchClone)
 
-		pages.GET("/boards", rt.auth(), rt.user(), rt.boardGetsByBids)
 		pages.GET("/board/:bid", rt.boardGet)
 		pages.GET("/board/:bid/pure", rt.boardPureGet)
 		pages.PUT("/board/:bid", rt.auth(), rt.user(), rt.perm("/dashboards/put"), rt.boardPut)
@@ -336,7 +324,7 @@ func (rt *Router) Config(r *gin.Engine) {
 		pages.GET("/alert-rule/:arid/pure", rt.auth(), rt.user(), rt.perm("/alert-rules"), rt.alertRulePureGet)
 		pages.PUT("/busi-group/alert-rule/validate", rt.auth(), rt.user(), rt.perm("/alert-rules/put"), rt.alertRuleValidation)
 		pages.POST("/relabel-test", rt.auth(), rt.user(), rt.relabelTest)
-		pages.POST("/busi-group/:id/alert-rules/clone", rt.auth(), rt.user(), rt.perm("/alert-rules/add"), rt.bgrw(), rt.cloneToMachine)
+		pages.POST("/busi-group/:id/alert-rules/clone", rt.auth(), rt.user(), rt.perm("/alert-rules/post"), rt.bgrw(), rt.cloneToMachine)
 
 		pages.GET("/busi-groups/recording-rules", rt.auth(), rt.user(), rt.perm("/recording-rules"), rt.recordingRuleGetsByGids)
 		pages.GET("/busi-group/:id/recording-rules", rt.auth(), rt.user(), rt.perm("/recording-rules"), rt.recordingRuleGets)
@@ -475,7 +463,6 @@ func (rt *Router) Config(r *gin.Engine) {
 		pages.GET("/builtin-payload/:id", rt.auth(), rt.user(), rt.perm("/built-in-components"), rt.builtinPayloadGet)
 		pages.PUT("/builtin-payloads", rt.auth(), rt.user(), rt.perm("/built-in-components/put"), rt.builtinPayloadsPut)
 		pages.DELETE("/builtin-payloads", rt.auth(), rt.user(), rt.perm("/built-in-components/del"), rt.builtinPayloadsDel)
-		pages.GET("/builtin-payload", rt.auth(), rt.user(), rt.builtinPayloadsGetByUUIDOrID)
 	}
 
 	r.GET("/api/n9e/versions", func(c *gin.Context) {
@@ -572,8 +559,6 @@ func (rt *Router) Config(r *gin.Engine) {
 			service.POST("/notify-record", rt.notificationRecordAdd)
 
 			service.GET("/alert-cur-events-del-by-hash", rt.alertCurEventDelByHash)
-
-			service.POST("/center/heartbeat", rt.heartbeat)
 		}
 	}
 

center/router/router_alert_cur_event.go

@@ -65,8 +65,7 @@ func (rt *Router) alertCurEventsCard(c *gin.Context) {
 	ginx.Dangerous(err)
 
 	// 最多获取50000个，获取太多也没啥意义
-	list, err := models.AlertCurEventsGet(rt.Ctx, prods, bgids, stime, etime, severity, dsIds,
-		cates, 0, query, 50000, 0)
+	list, err := models.AlertCurEventGets(rt.Ctx, prods, bgids, stime, etime, severity, dsIds, cates, query, 50000, 0)
 	ginx.Dangerous(err)
 
 	cardmap := make(map[string]*AlertCard)
@@ -163,17 +162,13 @@ func (rt *Router) alertCurEventsList(c *gin.Context) {
 		cates = strings.Split(cate, ",")
 	}
 
-	ruleId := ginx.QueryInt64(c, "rid", 0)
-
 	bgids, err := GetBusinessGroupIds(c, rt.Ctx, rt.Center.EventHistoryGroupView)
 	ginx.Dangerous(err)
 
-	total, err := models.AlertCurEventTotal(rt.Ctx, prods, bgids, stime, etime, severity, dsIds,
-		cates, ruleId, query)
+	total, err := models.AlertCurEventTotal(rt.Ctx, prods, bgids, stime, etime, severity, dsIds, cates, query)
 	ginx.Dangerous(err)
 
-	list, err := models.AlertCurEventsGet(rt.Ctx, prods, bgids, stime, etime, severity, dsIds,
-		cates, ruleId, query, limit, ginx.Offset(c, limit))
+	list, err := models.AlertCurEventGets(rt.Ctx, prods, bgids, stime, etime, severity, dsIds, cates, query, limit, ginx.Offset(c, limit))
 	ginx.Dangerous(err)
 
 	cache := make(map[int64]*models.UserGroup)
@@ -206,9 +201,7 @@ func (rt *Router) checkCurEventBusiGroupRWPermission(c *gin.Context, ids []int64
 	for i := 0; i < len(ids); i++ {
 		event, err := models.AlertCurEventGetById(rt.Ctx, ids[i])
 		ginx.Dangerous(err)
-		if event == nil {
-			continue
-		}
+
 		if _, has := set[event.GroupId]; !has {
 			rt.bgrwCheck(c, event.GroupId)
 			set[event.GroupId] = struct{}{}
@@ -234,7 +227,6 @@ func (rt *Router) alertCurEventGet(c *gin.Context) {
 		event.RuleConfigJson = ruleConfig
 	}
 
-	event.LastEvalTime = event.TriggerTime
 	ginx.NewRender(c).Data(event, nil)
 }
 

center/router/router_alert_his_event.go

@@ -54,17 +54,13 @@ func (rt *Router) alertHisEventsList(c *gin.Context) {
 		cates = strings.Split(cate, ",")
 	}
 
-	ruleId := ginx.QueryInt64(c, "rid", 0)
-
 	bgids, err := GetBusinessGroupIds(c, rt.Ctx, rt.Center.EventHistoryGroupView)
 	ginx.Dangerous(err)
 
-	total, err := models.AlertHisEventTotal(rt.Ctx, prods, bgids, stime, etime, severity,
-		recovered, dsIds, cates, ruleId, query)
+	total, err := models.AlertHisEventTotal(rt.Ctx, prods, bgids, stime, etime, severity, recovered, dsIds, cates, query)
 	ginx.Dangerous(err)
 
-	list, err := models.AlertHisEventGets(rt.Ctx, prods, bgids, stime, etime, severity, recovered,
-		dsIds, cates, ruleId, query, limit, ginx.Offset(c, limit))
+	list, err := models.AlertHisEventGets(rt.Ctx, prods, bgids, stime, etime, severity, recovered, dsIds, cates, query, limit, ginx.Offset(c, limit))
 	ginx.Dangerous(err)
 
 	cache := make(map[int64]*models.UserGroup)

center/router/router_alert_rule.go

@@ -37,18 +37,6 @@ func (rt *Router) alertRuleGets(c *gin.Context) {
 	ginx.NewRender(c).Data(ars, err)
 }
 
-func getAlertCueEventTimeRange(c *gin.Context) (stime, etime int64) {
-	stime = ginx.QueryInt64(c, "stime", 0)
-	etime = ginx.QueryInt64(c, "etime", 0)
-	if etime == 0 {
-		etime = time.Now().Unix()
-	}
-	if stime == 0 || stime >= etime {
-		stime = etime - 30*24*int64(time.Hour.Seconds())
-	}
-	return
-}
-
 func (rt *Router) alertRuleGetsByGids(c *gin.Context) {
 	gids := str.IdsInt64(ginx.QueryStr(c, "gids", ""), ",")
 	if len(gids) > 0 {
@@ -72,30 +60,9 @@ func (rt *Router) alertRuleGetsByGids(c *gin.Context) {
 	ars, err := models.AlertRuleGetsByBGIds(rt.Ctx, gids)
 	if err == nil {
 		cache := make(map[int64]*models.UserGroup)
-		rids := make([]int64, 0, len(ars))
-		names := make([]string, 0, len(ars))
 		for i := 0; i < len(ars); i++ {
 			ars[i].FillNotifyGroups(rt.Ctx, cache)
 			ars[i].FillSeverities()
-			rids = append(rids, ars[i].Id)
-			names = append(names, ars[i].UpdateBy)
-		}
-
-		stime, etime := getAlertCueEventTimeRange(c)
-		cnt := models.AlertCurEventCountByRuleId(rt.Ctx, rids, stime, etime)
-		if cnt != nil {
-			for i := 0; i < len(ars); i++ {
-				ars[i].CurEventCount = cnt[ars[i].Id]
-			}
-		}
-
-		users := models.UserMapGet(rt.Ctx, "username in (?)", names)
-		if users != nil {
-			for i := 0; i < len(ars); i++ {
-				if user, exist := users[ars[i].UpdateBy]; exist {
-					ars[i].UpdateByNickname = user.Nickname
-				}
-			}
 		}
 	}
 	ginx.NewRender(c).Data(ars, err)
@@ -525,7 +492,7 @@ func (rt *Router) relabelTest(c *gin.Context) {
 
 	labels := make([]prompb.Label, len(f.Tags))
 	for i, tag := range f.Tags {
-		label := strings.SplitN(tag, "=", 2)
+		label := strings.Split(tag, "=")
 		if len(label) != 2 {
 			ginx.Bomb(http.StatusBadRequest, "tag:%s format error", tag)
 		}
@@ -562,15 +529,6 @@ type identListForm struct {
 	IdentList []string `json:"ident_list"`
 }
 
-func containsIdentOperator(s string) bool {
-	pattern := `ident\s*(!=|!~|=~)`
-	matched, err := regexp.MatchString(pattern, s)
-	if err != nil {
-		return false
-	}
-	return matched
-}
-
 func (rt *Router) cloneToMachine(c *gin.Context) {
 	var f identListForm
 	ginx.BindJSON(c, &f)
@@ -592,17 +550,10 @@ func (rt *Router) cloneToMachine(c *gin.Context) {
 	reterr := make(map[string]map[string]string)
 
 	for i := range alertRules {
-		errMsg := make(map[string]string)
+		reterr[alertRules[i].Name] = make(map[string]string)
 
 		if alertRules[i].Cate != "prometheus" {
-			errMsg["all"] = "Only Prometheus rule can be cloned to machines"
-			reterr[alertRules[i].Name] = errMsg
-			continue
-		}
-
-		if containsIdentOperator(alertRules[i].RuleConfig) {
-			errMsg["all"] = "promql is missing ident"
-			reterr[alertRules[i].Name] = errMsg
+			reterr[alertRules[i].Name]["all"] = "Only Prometheus rules can be cloned to machines"
 			continue
 		}
 
@@ -611,7 +562,7 @@ func (rt *Router) cloneToMachine(c *gin.Context) {
 
 			newRule := &models.AlertRule{}
 			if err := copier.Copy(newRule, alertRules[i]); err != nil {
-				errMsg[f.IdentList[j]] = fmt.Sprintf("fail to clone rule, err: %s", err)
+				reterr[alertRules[i].Name][f.IdentList[j]] = fmt.Sprintf("fail to clone rule, err: %s", err)
 				continue
 			}
 
@@ -625,21 +576,17 @@ func (rt *Router) cloneToMachine(c *gin.Context) {
 
 			exist, err := models.AlertRuleExists(rt.Ctx, 0, newRule.GroupId, newRule.DatasourceIdsJson, newRule.Name)
 			if err != nil {
-				errMsg[f.IdentList[j]] = err.Error()
+				reterr[alertRules[i].Name][f.IdentList[j]] = err.Error()
 				continue
 			}
 
 			if exist {
-				errMsg[f.IdentList[j]] = fmt.Sprintf("rule already exists, ruleName: %s", newRule.Name)
+				reterr[alertRules[i].Name][f.IdentList[j]] = fmt.Sprintf("rule already exists, ruleName: %s", newRule.Name)
 				continue
 			}
 
 			newRules = append(newRules, newRule)
 		}
-
-		if len(errMsg) > 0 {
-			reterr[alertRules[i].Name] = errMsg
-		}
 	}
 
 	ginx.NewRender(c).Data(reterr, models.InsertAlertRule(rt.Ctx, newRules))

center/router/router_board.go

@@ -94,14 +94,6 @@ func (rt *Router) boardGet(c *gin.Context) {
 	ginx.NewRender(c).Data(board, nil)
 }
 
-// 根据 bids 参数，获取多个 board
-func (rt *Router) boardGetsByBids(c *gin.Context) {
-	bids := str.IdsInt64(ginx.QueryStr(c, "bids", ""), ",")
-	boards, err := models.BoardGetsByBids(rt.Ctx, bids)
-	ginx.Dangerous(err)
-	ginx.NewRender(c).Data(boards, err)
-}
-
 func (rt *Router) boardPureGet(c *gin.Context) {
 	board, err := models.BoardGetByID(rt.Ctx, ginx.UrlParamInt64(c, "bid"))
 	ginx.Dangerous(err)

center/router/router_builtin_componet.go

@@ -4,15 +4,10 @@ import (
 	"net/http"
 
 	"github.com/ccfos/nightingale/v6/models"
-	"github.com/ccfos/nightingale/v6/pkg/ctx"
-
 	"github.com/gin-gonic/gin"
 	"github.com/toolkits/pkg/ginx"
-	"gorm.io/gorm"
 )
 
-const SYSTEM = "system"
-
 func (rt *Router) builtinComponentsAdd(c *gin.Context) {
 	var lst []models.BuiltinComponent
 	ginx.BindJSON(c, &lst)
@@ -55,31 +50,10 @@ func (rt *Router) builtinComponentsPut(c *gin.Context) {
 		return
 	}
 
-	if bc.CreatedBy == SYSTEM {
-		req.Ident = bc.Ident
-	}
-
 	username := Username(c)
 	req.UpdatedBy = username
 
-	err = models.DB(rt.Ctx).Transaction(func(tx *gorm.DB) error {
-		tCtx := &ctx.Context{
-			DB: tx,
-		}
-
-		txErr := models.BuiltinMetricBatchUpdateColumn(tCtx, "typ", bc.Ident, req.Ident, req.UpdatedBy)
-		if txErr != nil {
-			return txErr
-		}
-
-		txErr = bc.Update(tCtx, req)
-		if txErr != nil {
-			return txErr
-		}
-		return nil
-	})
-
-	ginx.NewRender(c).Message(err)
+	ginx.NewRender(c).Message(bc.Update(rt.Ctx, req))
 }
 
 func (rt *Router) builtinComponentsDel(c *gin.Context) {

center/router/router_builtin_payload.go

@@ -6,7 +6,6 @@ import (
 	"strings"
 	"time"
 
-	"github.com/BurntSushi/toml"
 	"github.com/ccfos/nightingale/v6/models"
 	"github.com/gin-gonic/gin"
 	"github.com/toolkits/pkg/ginx"
@@ -53,15 +52,15 @@ func (rt *Router) builtinPayloadsAdd(c *gin.Context) {
 					}
 
 					bp := models.BuiltinPayload{
-						Type:        lst[i].Type,
-						ComponentID: lst[i].ComponentID,
-						Cate:        lst[i].Cate,
-						Name:        rule.Name,
-						Tags:        rule.AppendTags,
-						UUID:        rule.UUID,
-						Content:     string(contentBytes),
-						CreatedBy:   username,
-						UpdatedBy:   username,
+						Type:      lst[i].Type,
+						Component: lst[i].Component,
+						Cate:      lst[i].Cate,
+						Name:      rule.Name,
+						Tags:      rule.AppendTags,
+						UUID:      rule.UUID,
+						Content:   string(contentBytes),
+						CreatedBy: username,
+						UpdatedBy: username,
 					}
 
 					if err := bp.Add(rt.Ctx, username); err != nil {
@@ -82,15 +81,15 @@ func (rt *Router) builtinPayloadsAdd(c *gin.Context) {
 			}
 
 			bp := models.BuiltinPayload{
-				Type:        lst[i].Type,
-				ComponentID: lst[i].ComponentID,
-				Cate:        lst[i].Cate,
-				Name:        alertRule.Name,
-				Tags:        alertRule.AppendTags,
-				UUID:        alertRule.UUID,
-				Content:     lst[i].Content,
-				CreatedBy:   username,
-				UpdatedBy:   username,
+				Type:      lst[i].Type,
+				Component: lst[i].Component,
+				Cate:      lst[i].Cate,
+				Name:      alertRule.Name,
+				Tags:      alertRule.AppendTags,
+				UUID:      alertRule.UUID,
+				Content:   lst[i].Content,
+				CreatedBy: username,
+				UpdatedBy: username,
 			}
 
 			if err := bp.Add(rt.Ctx, username); err != nil {
@@ -116,15 +115,15 @@ func (rt *Router) builtinPayloadsAdd(c *gin.Context) {
 					}
 
 					bp := models.BuiltinPayload{
-						Type:        lst[i].Type,
-						ComponentID: lst[i].ComponentID,
-						Cate:        lst[i].Cate,
-						Name:        dashboard.Name,
-						Tags:        dashboard.Tags,
-						UUID:        dashboard.UUID,
-						Content:     string(contentBytes),
-						CreatedBy:   username,
-						UpdatedBy:   username,
+						Type:      lst[i].Type,
+						Component: lst[i].Component,
+						Cate:      lst[i].Cate,
+						Name:      dashboard.Name,
+						Tags:      dashboard.Tags,
+						UUID:      dashboard.UUID,
+						Content:   string(contentBytes),
+						CreatedBy: username,
+						UpdatedBy: username,
 					}
 
 					if err := bp.Add(rt.Ctx, username); err != nil {
@@ -145,29 +144,21 @@ func (rt *Router) builtinPayloadsAdd(c *gin.Context) {
 			}
 
 			bp := models.BuiltinPayload{
-				Type:        lst[i].Type,
-				ComponentID: lst[i].ComponentID,
-				Cate:        lst[i].Cate,
-				Name:        dashboard.Name,
-				Tags:        dashboard.Tags,
-				UUID:        dashboard.UUID,
-				Content:     lst[i].Content,
-				CreatedBy:   username,
-				UpdatedBy:   username,
+				Type:      lst[i].Type,
+				Component: lst[i].Component,
+				Cate:      lst[i].Cate,
+				Name:      dashboard.Name,
+				Tags:      dashboard.Tags,
+				UUID:      dashboard.UUID,
+				Content:   lst[i].Content,
+				CreatedBy: username,
+				UpdatedBy: username,
 			}
 
 			if err := bp.Add(rt.Ctx, username); err != nil {
 				reterr[bp.Name] = i18n.Sprintf(c.GetHeader("X-Language"), err.Error())
 			}
 		} else {
-			if lst[i].Type == "collect" {
-				c := make(map[string]interface{})
-				if _, err := toml.Decode(lst[i].Content, &c); err != nil {
-					reterr[lst[i].Name] = err.Error()
-					continue
-				}
-			}
-
 			if err := lst[i].Add(rt.Ctx, username); err != nil {
 				reterr[lst[i].Name] = i18n.Sprintf(c.GetHeader("X-Language"), err.Error())
 			}
@@ -180,20 +171,19 @@ func (rt *Router) builtinPayloadsAdd(c *gin.Context) {
 
 func (rt *Router) builtinPayloadsGets(c *gin.Context) {
 	typ := ginx.QueryStr(c, "type", "")
-	ComponentID := ginx.QueryInt64(c, "component_id", 0)
-
+	component := ginx.QueryStr(c, "component", "")
 	cate := ginx.QueryStr(c, "cate", "")
 	query := ginx.QueryStr(c, "query", "")
 
-	lst, err := models.BuiltinPayloadGets(rt.Ctx, uint64(ComponentID), typ, cate, query)
+	lst, err := models.BuiltinPayloadGets(rt.Ctx, typ, component, cate, query)
 	ginx.NewRender(c).Data(lst, err)
 }
 
 func (rt *Router) builtinPayloadcatesGet(c *gin.Context) {
 	typ := ginx.QueryStr(c, "type", "")
-	ComponentID := ginx.QueryInt64(c, "component_id", 0)
+	component := ginx.QueryStr(c, "component", "")
 
-	cates, err := models.BuiltinPayloadCates(rt.Ctx, typ, uint64(ComponentID))
+	cates, err := models.BuiltinPayloadCates(rt.Ctx, typ, component)
 	ginx.NewRender(c).Data(cates, err)
 }
 
@@ -239,11 +229,6 @@ func (rt *Router) builtinPayloadsPut(c *gin.Context) {
 
 		req.Name = dashboard.Name
 		req.Tags = dashboard.Tags
-	} else if req.Type == "collect" {
-		c := make(map[string]interface{})
-		if _, err := toml.Decode(req.Content, &c); err != nil {
-			ginx.Bomb(http.StatusBadRequest, err.Error())
-		}
 	}
 
 	username := Username(c)
@@ -260,15 +245,3 @@ func (rt *Router) builtinPayloadsDel(c *gin.Context) {
 
 	ginx.NewRender(c).Message(models.BuiltinPayloadDels(rt.Ctx, req.Ids))
 }
-
-func (rt *Router) builtinPayloadsGetByUUIDOrID(c *gin.Context) {
-	uuid := ginx.QueryInt64(c, "uuid", 0)
-	// 优先以 uuid 为准
-	if uuid != 0 {
-		ginx.NewRender(c).Data(models.BuiltinPayloadGet(rt.Ctx, "uuid = ?", uuid))
-		return
-	}
-
-	id := ginx.QueryInt64(c, "id", 0)
-	ginx.NewRender(c).Data(models.BuiltinPayloadGet(rt.Ctx, "id = ?", id))
-}

center/router/router_datasource.go

@@ -92,12 +92,10 @@ func (rt *Router) datasourceUpsert(c *gin.Context) {
 	var err error
 	var count int64
 
-	if !req.ForceSave {
-		err = DatasourceCheck(req)
-		if err != nil {
-			Dangerous(c, err)
-			return
-		}
+	err = DatasourceCheck(req)
+	if err != nil {
+		Dangerous(c, err)
+		return
 	}
 
 	if req.Id == 0 {

center/router/router_funcs.go

@@ -65,23 +65,6 @@ func queryDatasourceIds(c *gin.Context) []int64 {
 	return ids
 }
 
-func queryStrListField(c *gin.Context, fieldName string, sep ...string) []string {
-	str := ginx.QueryStr(c, fieldName, "")
-	if str == "" {
-		return nil
-	}
-
-	lst := []string{str}
-	for _, s := range sep {
-		var newLst []string
-		for _, str := range lst {
-			newLst = append(newLst, strings.Split(str, s)...)
-		}
-		lst = newLst
-	}
-	return lst
-}
-
 type idsForm struct {
 	Ids               []int64 `json:"ids"`
 	IsSyncToFlashDuty bool    `json:"is_sync_to_flashduty"`

center/router/router_heartbeat.go

@@ -6,7 +6,6 @@ import (
 	"errors"
 	"io/ioutil"
 	"sort"
-	"strconv"
 	"strings"
 	"time"
 
@@ -81,48 +80,16 @@ func HandleHeartbeat(c *gin.Context, ctx *ctx.Context, engineName string, metaSe
 	identSet.MSet(items)
 
 	if target, has := targetCache.Get(req.Hostname); has && target != nil {
-		gidsStr := ginx.QueryStr(c, "gid", "")
-		overwriteGids := ginx.QueryBool(c, "overwrite_gids", false)
+		gid := ginx.QueryInt64(c, "gid", 0)
 		hostIp := strings.TrimSpace(req.HostIp)
-		gids := strings.Split(gidsStr, ",")
-
-		if overwriteGids {
-			groupIds := make([]int64, 0)
-			for i := range gids {
-				if gids[i] == "" {
-					continue
-				}
-				groupId, err := strconv.ParseInt(gids[i], 10, 64)
-				if err != nil {
-					logger.Warningf("update target:%s group ids failed, err: %v", req.Hostname, err)
-					continue
-				}
-				groupIds = append(groupIds, groupId)
-			}
-
-			err := models.TargetOverrideBgids(ctx, []string{target.Ident}, groupIds)
-			if err != nil {
-				logger.Warningf("update target:%s group ids failed, err: %v", target.Ident, err)
-			}
-		} else if gidsStr != "" {
-			for i := range gids {
-				groupId, err := strconv.ParseInt(gids[i], 10, 64)
-				if err != nil {
-					logger.Warningf("update target:%s group ids failed, err: %v", req.Hostname, err)
-					continue
-				}
-
-				if !target.MatchGroupId(groupId) {
-					err := models.TargetBindBgids(ctx, []string{target.Ident}, []int64{groupId})
-					if err != nil {
-						logger.Warningf("update target:%s group ids failed, err: %v", target.Ident, err)
-					}
-				}
-			}
-		}
 
 		newTarget := models.Target{}
 		targetNeedUpdate := false
+		if gid != 0 && gid != target.GroupId {
+			newTarget.GroupId = gid
+			targetNeedUpdate = true
+		}
+
 		if hostIp != "" && hostIp != target.HostIp {
 			newTarget.HostIp = hostIp
 			targetNeedUpdate = true

center/router/router_notify_tpl.go

@@ -138,7 +138,7 @@ func (rt *Router) notifyTplPreview(c *gin.Context) {
 			continue
 		}
 
-		arr := strings.SplitN(pair, "=", 2)
+		arr := strings.Split(pair, "=")
 		if len(arr) != 2 {
 			continue
 		}

center/router/router_target.go

@@ -52,8 +52,6 @@ func (rt *Router) targetGets(c *gin.Context) {
 	order := ginx.QueryStr(c, "order", "ident")
 	desc := ginx.QueryBool(c, "desc", false)
 
-	hosts := queryStrListField(c, "hosts", ",", " ", "\n")
-
 	var err error
 	if len(bgids) == 0 {
 		user := c.MustGet("user").(*models.User)
@@ -67,13 +65,11 @@ func (rt *Router) targetGets(c *gin.Context) {
 			bgids = append(bgids, 0)
 		}
 	}
-
 	options := []models.BuildTargetWhereOption{
 		models.BuildTargetWhereWithBgids(bgids),
 		models.BuildTargetWhereWithDsIds(dsIds),
 		models.BuildTargetWhereWithQuery(query),
 		models.BuildTargetWhereWithDowntime(downtime),
-		models.BuildTargetWhereWithHosts(hosts),
 	}
 	total, err := models.TargetTotal(rt.Ctx, options...)
 	ginx.Dangerous(err)
@@ -82,13 +78,6 @@ func (rt *Router) targetGets(c *gin.Context) {
 		ginx.Offset(c, limit), order, desc, options...)
 	ginx.Dangerous(err)
 
-	tgs, err := models.TargetBusiGroupsGetAll(rt.Ctx)
-	ginx.Dangerous(err)
-
-	for _, t := range list {
-		t.GroupIds = tgs[t.Ident]
-	}
-
 	if err == nil {
 		now := time.Now()
 		cache := make(map[int64]*models.BusiGroup)
@@ -393,15 +382,8 @@ type targetBgidForm struct {
 	Bgid    int64    `json:"bgid"`
 }
 
-type targetBgidsForm struct {
-	Idents  []string `json:"idents" binding:"required_without=HostIps"`
-	HostIps []string `json:"host_ips" binding:"required_without=Idents"`
-	Bgids   []int64  `json:"bgids"`
-	Action  string   `json:"action"` // add del reset
-}
-
-func (rt *Router) targetBindBgids(c *gin.Context) {
-	var f targetBgidsForm
+func (rt *Router) targetUpdateBgid(c *gin.Context) {
+	var f targetBgidForm
 	var err error
 	var failedResults = make(map[string]string)
 	ginx.BindJSON(c, &f)
@@ -417,24 +399,35 @@ func (rt *Router) targetBindBgids(c *gin.Context) {
 	}
 
 	user := c.MustGet("user").(*models.User)
-	if !user.IsAdmin() {
-		// 普通用户，检查用户是否有权限操作所有请求的业务组
-		existing, _, err := models.SeparateTargetIdents(rt.Ctx, f.Idents)
+	if user.IsAdmin() {
+		ginx.NewRender(c).Data(failedResults, models.TargetUpdateBgid(rt.Ctx, f.Idents, f.Bgid, false))
+		return
+	}
+
+	if f.Bgid > 0 {
+		// 把要操作的机器分成两部分，一部分是bgid为0，需要管理员分配，另一部分bgid>0，说明是业务组内部想调整
+		// 比如原来分配给didiyun的机器，didiyun的管理员想把部分机器调整到didiyun-ceph下
+		// 对于调整的这种情况，当前登录用户要对这批机器有操作权限，同时还要对目标BG有操作权限
+		orphans, err := models.IdentsFilter(rt.Ctx, f.Idents, "group_id = ?", 0)
 		ginx.Dangerous(err)
-		rt.checkTargetPerm(c, existing)
 
-		var groupIds []int64
-		if f.Action == "reset" {
-			// 如果是复写，则需要检查用户是否有权限操作机器之前的业务组
-			bgids, err := models.TargetGroupIdsGetByIdents(rt.Ctx, f.Idents)
+		// 机器里边存在未归组的，登录用户就需要是admin
+		if len(orphans) > 0 && !user.IsAdmin() {
+			can, err := user.CheckPerm(rt.Ctx, "/targets/bind")
 			ginx.Dangerous(err)
-
-			groupIds = append(groupIds, bgids...)
+			if !can {
+				ginx.Bomb(http.StatusForbidden, "No permission. Only admin can assign BG")
+			}
 		}
-		groupIds = append(groupIds, f.Bgids...)
 
-		for _, bgid := range groupIds {
-			bg := BusiGroup(rt.Ctx, bgid)
+		reBelongs, err := models.IdentsFilter(rt.Ctx, f.Idents, "group_id > ?", 0)
+		ginx.Dangerous(err)
+
+		if len(reBelongs) > 0 {
+			// 对于这些要重新分配的机器，操作者要对这些机器本身有权限，同时要对目标bgid有权限
+			rt.checkTargetPerm(c, f.Idents)
+
+			bg := BusiGroup(rt.Ctx, f.Bgid)
 			can, err := user.CanDoBusiGroup(rt.Ctx, bg, "rw")
 			ginx.Dangerous(err)
 
@@ -442,24 +435,14 @@ func (rt *Router) targetBindBgids(c *gin.Context) {
 				ginx.Bomb(http.StatusForbidden, "No permission. You are not admin of BG(%s)", bg.Name)
 			}
 		}
-
-		can, err := user.CheckPerm(rt.Ctx, "/targets/bind")
-		ginx.Dangerous(err)
-		if !can {
-			ginx.Bomb(http.StatusForbidden, "No permission. Only admin can assign BG")
-		}
+	} else if f.Bgid == 0 {
+		// 退还机器
+		rt.checkTargetPerm(c, f.Idents)
+	} else {
+		ginx.Bomb(http.StatusBadRequest, "invalid bgid")
 	}
 
-	switch f.Action {
-	case "add":
-		ginx.NewRender(c).Data(failedResults, models.TargetBindBgids(rt.Ctx, f.Idents, f.Bgids))
-	case "del":
-		ginx.NewRender(c).Data(failedResults, models.TargetUnbindBgids(rt.Ctx, f.Idents, f.Bgids))
-	case "reset":
-		ginx.NewRender(c).Data(failedResults, models.TargetOverrideBgids(rt.Ctx, f.Idents, f.Bgids))
-	default:
-		ginx.Bomb(http.StatusBadRequest, "invalid action")
-	}
+	ginx.NewRender(c).Data(failedResults, models.TargetUpdateBgid(rt.Ctx, f.Idents, f.Bgid, false))
 }
 
 func (rt *Router) targetUpdateBgidByService(c *gin.Context) {
@@ -478,7 +461,7 @@ func (rt *Router) targetUpdateBgidByService(c *gin.Context) {
 		ginx.Bomb(http.StatusBadRequest, err.Error())
 	}
 
-	ginx.NewRender(c).Data(failedResults, models.TargetOverrideBgids(rt.Ctx, f.Idents, []int64{f.Bgid}))
+	ginx.NewRender(c).Data(failedResults, models.TargetUpdateBgid(rt.Ctx, f.Idents, f.Bgid, false))
 }
 
 type identsForm struct {
@@ -502,7 +485,7 @@ func (rt *Router) targetDel(c *gin.Context) {
 		ginx.Bomb(http.StatusBadRequest, err.Error())
 	}
 
-	ginx.NewRender(c).Data(failedResults, models.TargetDel(rt.Ctx, f.Idents, rt.TargetDeleteHook))
+	ginx.NewRender(c).Data(failedResults, models.TargetDel(rt.Ctx, f.Idents))
 }
 
 func (rt *Router) targetDelByService(c *gin.Context) {
@@ -521,7 +504,7 @@ func (rt *Router) targetDelByService(c *gin.Context) {
 		ginx.Bomb(http.StatusBadRequest, err.Error())
 	}
 
-	ginx.NewRender(c).Data(failedResults, models.TargetDel(rt.Ctx, f.Idents, rt.TargetDeleteHook))
+	ginx.NewRender(c).Data(failedResults, models.TargetDel(rt.Ctx, f.Idents))
 }
 
 func (rt *Router) checkTargetPerm(c *gin.Context, idents []string) {

center/router/router_task.go

@@ -126,7 +126,7 @@ func (rt *Router) taskAdd(c *gin.Context) {
 	rt.checkTargetPerm(c, f.Hosts)
 
 	// call ibex
-	taskId, err := sender.TaskAdd(f, user.Username, rt.Ctx.IsCenter)
+	taskId, err := sender.TaskAdd(rt.Ctx, f, user.Username, rt.Ctx.IsCenter)
 	ginx.Dangerous(err)
 
 	if taskId <= 0 {

cli/upgrade/upgrade.go

@@ -18,7 +18,7 @@ func Upgrade(configFile string) error {
 		return err
 	}
 
-	ctx := ctx.NewContext(context.Background(), db, true)
+	ctx := ctx.NewContext(context.Background(), db, nil, true)
 	for _, cluster := range config.Clusters {
 		count, err := models.GetDatasourcesCountByName(ctx, cluster.Name)
 		if err != nil {

cmd/edge/edge.go

@@ -12,6 +12,7 @@ import (
 	"github.com/ccfos/nightingale/v6/center/metas"
 	"github.com/ccfos/nightingale/v6/conf"
 	"github.com/ccfos/nightingale/v6/dumper"
+	"github.com/ccfos/nightingale/v6/ibex"
 	"github.com/ccfos/nightingale/v6/memsto"
 	"github.com/ccfos/nightingale/v6/pkg/ctx"
 	"github.com/ccfos/nightingale/v6/pkg/httpx"
@@ -22,8 +23,6 @@ import (
 	"github.com/ccfos/nightingale/v6/pushgw/writer"
 	"github.com/ccfos/nightingale/v6/storage"
 	"github.com/ccfos/nightingale/v6/tdengine"
-
-	"github.com/flashcatcloud/ibex/src/cmd/ibex"
 )
 
 func Initialize(configDir string, cryptoKey string) (func(), error) {
@@ -40,7 +39,6 @@ func Initialize(configDir string, cryptoKey string) (func(), error) {
 	if len(config.CenterApi.Addrs) < 1 {
 		return nil, errors.New("failed to init config: the CenterApi configuration is missing")
 	}
-	ctx := ctx.NewContext(context.Background(), nil, false, config.CenterApi)
 
 	var redis storage.Redis
 	redis, err = storage.NewRedis(config.Redis)
@@ -48,6 +46,8 @@ func Initialize(configDir string, cryptoKey string) (func(), error) {
 		return nil, err
 	}
 
+	ctx := ctx.NewContext(context.Background(), nil, redis, false, config.CenterApi)
+
 	syncStats := memsto.NewSyncStats()
 
 	targetCache := memsto.NewTargetCache(ctx, syncStats, redis)
@@ -82,12 +82,12 @@ func Initialize(configDir string, cryptoKey string) (func(), error) {
 		alertrtRouter.Config(r)
 
 		if config.Ibex.Enable {
-			ibex.ServerStart(false, nil, redis, config.HTTP.APIForService.BasicAuth, config.Alert.Heartbeat, &config.CenterApi, r, nil, config.Ibex, config.HTTP.Port)
+			ibex.ServerStart(ctx, false, nil, redis, config.HTTP.APIForService.BasicAuth, config.Alert.Heartbeat, &config.CenterApi, r, nil, config.Ibex, config.HTTP.Port)
 		}
 	}
 
 	dumper.ConfigRouter(r)
-	httpClean := httpx.Init(config.HTTP, r)
+	httpClean := httpx.Init(config.HTTP, context.Background(), r)
 
 	return func() {
 		logxClean()

cmd/ibex/main.go

@@ -0,0 +1,119 @@
+package main
+
+import (
+	"fmt"
+	"os"
+
+	"github.com/ccfos/nightingale/v6/ibex/agentd"
+	"github.com/ccfos/nightingale/v6/ibex/server"
+
+	"github.com/toolkits/pkg/net/tcpx"
+	"github.com/toolkits/pkg/runner"
+	"github.com/urfave/cli/v2"
+)
+
+// VERSION go build -ldflags "-X main.VERSION=x.x.x"
+var VERSION = "not specified"
+
+func main() {
+	app := cli.NewApp()
+	app.Name = "ibex"
+	app.Version = VERSION
+	app.Usage = "Ibex, running scripts on large scale machines"
+	app.Commands = []*cli.Command{
+		newCenterServerCmd(),
+		newEdgeServerCmd(),
+		newAgentdCmd(),
+	}
+	app.Run(os.Args)
+}
+
+func newCenterServerCmd() *cli.Command {
+	return &cli.Command{
+		Name:  "server",
+		Usage: "Run server",
+		Flags: []cli.Flag{
+			&cli.StringFlag{
+				Name:    "conf",
+				Aliases: []string{"c"},
+				Usage:   "specify configuration file(.json,.yaml,.toml)",
+			},
+		},
+		Action: func(c *cli.Context) error {
+			printEnv()
+
+			tcpx.WaitHosts()
+
+			var opts []server.ServerOption
+			if c.String("conf") != "" {
+				opts = append(opts, server.SetConfigFile(c.String("conf")))
+			}
+			opts = append(opts, server.SetVersion(VERSION))
+
+			server.Run(true, opts...)
+			return nil
+		},
+	}
+}
+
+func newEdgeServerCmd() *cli.Command {
+	return &cli.Command{
+		Name:  "edge server",
+		Usage: "Run edge server",
+		Flags: []cli.Flag{
+			&cli.StringFlag{
+				Name:    "conf",
+				Aliases: []string{"c"},
+				Usage:   "specify configuration file(.json,.yaml,.toml)",
+			},
+		},
+		Action: func(c *cli.Context) error {
+			printEnv()
+
+			tcpx.WaitHosts()
+
+			var opts []server.ServerOption
+			if c.String("conf") != "" {
+				opts = append(opts, server.SetConfigFile(c.String("conf")))
+			}
+			opts = append(opts, server.SetVersion(VERSION))
+
+			server.Run(false, opts...)
+			return nil
+		},
+	}
+}
+
+func newAgentdCmd() *cli.Command {
+	return &cli.Command{
+		Name:  "agentd",
+		Usage: "Run agentd",
+		Flags: []cli.Flag{
+			&cli.StringFlag{
+				Name:    "conf",
+				Aliases: []string{"c"},
+				Usage:   "specify configuration file(.json,.yaml,.toml)",
+			},
+		},
+		Action: func(c *cli.Context) error {
+			printEnv()
+
+			var opts []agentd.AgentdOption
+			if c.String("conf") != "" {
+				opts = append(opts, agentd.SetConfigFile(c.String("conf")))
+			}
+			opts = append(opts, agentd.SetVersion(VERSION))
+
+			agentd.Run(opts...)
+			return nil
+		},
+	}
+}
+
+func printEnv() {
+	runner.Init()
+	fmt.Println("runner.cwd:", runner.Cwd)
+	fmt.Println("runner.hostname:", runner.Hostname)
+	fmt.Println("runner.fd_limits:", runner.FdLimits())
+	fmt.Println("runner.vm_limits:", runner.VMLimits())
+}

docker/compose-postgres/initsql_for_postgres/a-n9e-for-Postgres.sql

@@ -561,7 +561,7 @@ CREATE TABLE alert_cur_event (
     target_note varchar(191) not null default '' ,
     first_trigger_time bigint,
     trigger_time bigint not null,
-    trigger_value varchar(2048) not null,
+    trigger_value varchar(255) not null,
     annotations text not null ,
     rule_config text not null ,
     tags varchar(1024) not null default '' ,
@@ -621,7 +621,7 @@ CREATE TABLE alert_his_event (
     target_note varchar(191) not null default '' ,
     first_trigger_time bigint,
     trigger_time bigint not null,
-    trigger_value varchar(2048) not null,
+    trigger_value varchar(255) not null,
     recover_time bigint not null default 0,
     last_eval_time bigint not null default 0 ,
     tags varchar(1024) not null default '' ,

docker/initsql/a-n9e.sql

@@ -453,7 +453,7 @@ CREATE TABLE `alert_cur_event` (
     `target_note` varchar(191) not null default '' comment 'target note',
     `first_trigger_time` bigint,
     `trigger_time` bigint not null,
-    `trigger_value` text not null,
+    `trigger_value` varchar(255) not null,
     `annotations` text not null comment 'annotations',
     `rule_config` text not null comment 'annotations',
     `tags` varchar(1024) not null default '' comment 'merge data_tags rule_tags, split by ,,',
@@ -493,7 +493,7 @@ CREATE TABLE `alert_his_event` (
     `target_note` varchar(191) not null default '' comment 'target note',
     `first_trigger_time` bigint,
     `trigger_time` bigint not null,
-    `trigger_value` text not null,
+    `trigger_value` varchar(255) not null,
     `recover_time` bigint not null default 0,
     `last_eval_time` bigint not null default 0 comment 'for time filter',
     `tags` varchar(1024) not null default '' comment 'merge data_tags rule_tags, split by ,,',
@@ -528,7 +528,6 @@ CREATE TABLE `builtin_components` (
 
 CREATE TABLE `builtin_payloads` (
   `id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '''unique identifier''',
-  `component_id` bigint(20) NOT NULL DEFAULT 0 COMMENT 'component_id',
   `uuid` bigint(20) NOT NULL COMMENT '''uuid of payload''',
   `type` varchar(191) NOT NULL COMMENT '''type of payload''',
   `component` varchar(191) NOT NULL COMMENT '''component of payload''',

docker/initsql/sqlite.sql

@@ -389,7 +389,7 @@ CREATE TABLE `alert_cur_event` (
     `target_note` varchar(191) not null default '',
     `first_trigger_time` bigint,
     `trigger_time` bigint not null,
-    `trigger_value` varchar(2048) not null,
+    `trigger_value` varchar(255) not null,
     `annotations` text not null,
     `rule_config` text not null,
     `tags` varchar(1024) not null default ''
@@ -427,7 +427,7 @@ CREATE TABLE `alert_his_event` (
     `target_note` varchar(191) not null default '',
     `first_trigger_time` bigint,
     `trigger_time` bigint not null,
-    `trigger_value` varchar(2048) not null,
+    `trigger_value` varchar(255) not null,
     `recover_time` bigint not null default 0,
     `last_eval_time` bigint not null default 0,
     `tags` varchar(1024) not null default '',

docker/migratesql/migrate.sql

@@ -103,17 +103,4 @@ CREATE TABLE notification_record (
 
 
 /* v7.3.0 2024-08-26 */
-ALTER TABLE `target` ADD COLUMN `host_tags` TEXT COMMENT 'global labels set in conf file';
-
-/* v7.3.4 2024-08-28 */
-ALTER TABLE `builtin_payloads` ADD COLUMN `component_id` bigint(20) NOT NULL DEFAULT 0 COMMENT 'component_id';
-
-/* v7.4.0 2024-09-20 */
-CREATE TABLE `target_busi_group` (
-  `id` bigint NOT NULL AUTO_INCREMENT,
-  `target_ident` varchar(191) NOT NULL,
-  `group_id` bigint NOT NULL,
-  `update_at` bigint NOT NULL,
-  PRIMARY KEY (`id`),
-  UNIQUE KEY `idx_target_group` (`target_ident`,`group_id`)
-) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4;
+ALTER TABLE `target` ADD COLUMN `host_tags` TEXT COMMENT 'global labels set in conf file';

etc/ibex/agentd.toml

@@ -0,0 +1,38 @@
+# debug, release
+RunMode = "debug"
+
+# task meta storage dir
+MetaDir = "./meta"
+
+[HTTP]
+Enable = true
+# http listening address
+Host = "0.0.0.0"
+# http listening port
+Port = 2090
+# https cert file path
+CertFile = ""
+# https key file path
+KeyFile = ""
+# whether print access log
+PrintAccessLog = true
+# whether enable pprof
+PProf = false
+# http graceful shutdown timeout, unit: s
+ShutdownTimeout = 30
+# max content length: 64M
+MaxContentLength = 67108864
+# http server read timeout, unit: s
+ReadTimeout = 20
+# http server write timeout, unit: s
+WriteTimeout = 40
+# http server idle timeout, unit: s
+IdleTimeout = 120
+
+[Heartbeat]
+# unit: ms
+Interval = 1000
+# rpc servers
+Servers = ["127.0.0.1:20090"]
+# $ip or $hostname or specified string
+Host = "$hostname"

etc/ibex/ibex-agentd.service

@@ -0,0 +1,20 @@
+[Unit]
+Description="ibex-agentd"
+After=network.target
+
+[Service]
+Type=simple
+
+ExecStart=/root/gopath/ibex/ibex agentd
+WorkingDirectory=/root/gopath/ibex
+
+Restart=on-failure
+SuccessExitStatus=0
+LimitNOFILE=65536
+StandardOutput=syslog
+StandardError=syslog
+SyslogIdentifier=ibex-agentd
+
+
+[Install]
+WantedBy=multi-user.target

etc/ibex/ibex-server.service

@@ -0,0 +1,20 @@
+[Unit]
+Description="ibex-server"
+After=network.target
+
+[Service]
+Type=simple
+
+ExecStart=/root/gopath/ibex/ibex server
+WorkingDirectory=/root/gopath/ibex
+
+Restart=on-failure
+SuccessExitStatus=0
+LimitNOFILE=65536
+StandardOutput=syslog
+StandardError=syslog
+SyslogIdentifier=ibex-server
+
+
+[Install]
+WantedBy=multi-user.target

etc/ibex/server.toml

@@ -0,0 +1,86 @@
+# debug, release
+RunMode = "debug"
+
+[Log]
+# log write dir
+Dir = "logs-server"
+# log level: DEBUG INFO WARNING ERROR
+Level = "DEBUG"
+# stdout, stderr, file
+Output = "stdout"
+# # rotate by time
+# KeepHours: 4
+# # rotate by size
+# RotateNum = 3
+# # unit: MB
+# RotateSize = 256
+
+[HTTP]
+Enable = true
+# http listening address
+Host = "0.0.0.0"
+# http listening port
+Port = 10090
+# https cert file path
+CertFile = ""
+# https key file path
+KeyFile = ""
+# whether print access log
+PrintAccessLog = true
+# whether enable pprof
+PProf = false
+# http graceful shutdown timeout, unit: s
+ShutdownTimeout = 30
+# max content length: 64M
+MaxContentLength = 67108864
+# http server read timeout, unit: s
+ReadTimeout = 20
+# http server write timeout, unit: s
+WriteTimeout = 40
+# http server idle timeout, unit: s
+IdleTimeout = 120
+
+[BasicAuth]
+# using when call apis
+ibex = "ibex"
+
+[RPC]
+Listen = "0.0.0.0:20090"
+
+[Heartbeat]
+# auto detect if blank
+IP = ""
+# unit: ms
+Interval = 1000
+
+[Output]
+# database | remote
+ComeFrom = "database"
+AgtdPort = 2090
+
+[DB]
+# postgres: host=%s port=%s user=%s dbname=%s password=%s sslmode=%s
+# postgres: DSN="host=127.0.0.1 port=5432 user=root dbname=n9e_v6 password=1234 sslmode=disable"
+DSN="root:1234@tcp(127.0.0.1:3306)/ibex?charset=utf8mb4&parseTime=True&loc=Local&allowNativePasswords=true"
+# enable debug mode or not
+Debug = false
+# mysql postgres
+DBType = "mysql"
+# unit: s
+MaxLifetime = 7200
+# max open connections
+MaxOpenConns = 150
+# max idle connections
+MaxIdleConns = 50
+# table prefix
+TablePrefix = ""
+
+[Redis]
+# address, ip:port or ip1:port,ip2:port for cluster and sentinel(SentinelAddrs)
+Address = "127.0.0.1:6379"
+# Username = ""
+# Password = ""
+# DB = 0
+# UseTLS = false
+# TLSMinVersion = "1.2"
+# standalone cluster sentinel

go.mod

@@ -3,12 +3,11 @@ module github.com/ccfos/nightingale/v6
 go 1.18
 
 require (
-	github.com/BurntSushi/toml v0.3.1
+	github.com/BurntSushi/toml v1.3.2
 	github.com/coreos/go-oidc v2.2.1+incompatible
 	github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc
 	github.com/dgrijalva/jwt-go v3.2.0+incompatible
 	github.com/expr-lang/expr v1.16.1
-	github.com/flashcatcloud/ibex v1.3.5
 	github.com/gin-contrib/pprof v1.4.0
 	github.com/gin-gonic/gin v1.9.1
 	github.com/go-ldap/ldap/v3 v3.4.4
@@ -34,6 +33,7 @@ require (
 	github.com/spaolacci/murmur3 v1.1.0
 	github.com/tidwall/gjson v1.14.0
 	github.com/toolkits/pkg v1.3.6
+	github.com/urfave/cli/v2 v2.27.4
 	golang.org/x/exp v0.0.0-20230713183714-613f0c0eb8a1
 	golang.org/x/oauth2 v0.10.0
 	gopkg.in/gomail.v2 v2.0.0-20160411212932-81ebce5c23df
@@ -44,6 +44,12 @@ require (
 	gorm.io/gorm v1.25.7-0.20240204074919-46816ad31dde
 )
 
+require (
+	github.com/cpuguy83/go-md2man/v2 v2.0.4 // indirect
+	github.com/russross/blackfriday/v2 v2.1.0 // indirect
+	github.com/xrash/smetrics v0.0.0-20240521201337-686a1a2994c1 // indirect
+)
+
 require (
 	github.com/Azure/go-ntlmssp v0.0.0-20220621081337-cb9428e4ac1e // indirect
 	github.com/beorn7/perks v1.0.1 // indirect
@@ -90,7 +96,7 @@ require (
 	github.com/tidwall/match v1.1.1 // indirect
 	github.com/tidwall/pretty v1.2.0 // indirect
 	github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
-	github.com/ugorji/go/codec v1.2.11 // indirect
+	github.com/ugorji/go/codec v1.2.11
 	go.uber.org/atomic v1.11.0 // indirect
 	go.uber.org/automaxprocs v1.5.2 // indirect
 	golang.org/x/arch v0.3.0 // indirect

go.sum

@@ -5,8 +5,9 @@ github.com/Azure/azure-sdk-for-go/sdk/internal v1.3.0 h1:sXr+ck84g/ZlZUOZiNELInm
 github.com/Azure/go-ntlmssp v0.0.0-20220621081337-cb9428e4ac1e h1:NeAW1fUYUEWhft7pkxDf6WoUvEZJ/uOKsvtpjLnn8MU=
 github.com/Azure/go-ntlmssp v0.0.0-20220621081337-cb9428e4ac1e/go.mod h1:chxPXzSsl7ZWRAuOIE23GDNzjWuZquvFlgA8xmpunjU=
 github.com/AzureAD/microsoft-authentication-library-for-go v1.0.0 h1:OBhqkivkhkMqLPymWEppkm7vgPQY2XsHoEkaMQ0AdZY=
-github.com/BurntSushi/toml v0.3.1 h1:WXkYYl6Yr3qBf1K79EBnL4mak0OimBfB0XUf9Vl28OQ=
 github.com/BurntSushi/toml v0.3.1/go.mod h1:xHWCNGjB5oqiDr8zfno3MHue2Ht5sIBksp03qcyfWMU=
+github.com/BurntSushi/toml v1.3.2 h1:o7IhLm0Msx3BaB+n3Ag7L8EVlByGnpq14C4YWiu/gL8=
+github.com/BurntSushi/toml v1.3.2/go.mod h1:CxXYINrC8qIiEnFrOxCa7Jy5BFHlXnUU2pbicEuybxQ=
 github.com/Masterminds/semver/v3 v3.1.1 h1:hLg3sBzpNErnxhQtUy/mmLR2I9foDujNK030IGemrRc=
 github.com/Masterminds/semver/v3 v3.1.1/go.mod h1:VPu/7SZ7ePZ3QOrcuXROw5FAcLl4a0cBrbBpGY/8hQs=
 github.com/alecthomas/units v0.0.0-20211218093645-b94a6e3cc137 h1:s6gZFSlWYmbqAuRjVTiNNhvNRfY2Wxp9nhfyel4rklc=
@@ -29,6 +30,8 @@ github.com/coreos/go-oidc v2.2.1+incompatible h1:mh48q/BqXqgjVHpy2ZY7WnWAbenxRjs
 github.com/coreos/go-oidc v2.2.1+incompatible/go.mod h1:CgnwVTmzoESiwO9qyAFEMiHoZ1nMCKZlZ9V6mm3/LKc=
 github.com/coreos/go-systemd v0.0.0-20190321100706-95778dfbb74e/go.mod h1:F5haX7vjVVG0kc13fIWeqUViNPyEJxv/OmvnBo0Yme4=
 github.com/coreos/go-systemd v0.0.0-20190719114852-fd7a80b32e1f/go.mod h1:F5haX7vjVVG0kc13fIWeqUViNPyEJxv/OmvnBo0Yme4=
+github.com/cpuguy83/go-md2man/v2 v2.0.4 h1:wfIWP927BUkWJb2NmU/kNDYIBTh/ziUX91+lVfRxZq4=
+github.com/cpuguy83/go-md2man/v2 v2.0.4/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
 github.com/creack/pty v1.1.7/go.mod h1:lj5s0c3V2DBrqTV7llrYr5NG6My20zk30Fl46Y7DoTY=
 github.com/creack/pty v1.1.9/go.mod h1:oKZEueFk5CKHvIhNR5MUki03XCEU+Q6VDXinZuGJ33E=
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
@@ -47,8 +50,6 @@ github.com/fatih/camelcase v1.0.0 h1:hxNvNX/xYBp0ovncs8WyWZrOrpBNub/JfaMvbURyft8
 github.com/fatih/camelcase v1.0.0/go.mod h1:yN2Sb0lFhZJUdVvtELVWefmrXpuZESvPmqwoZc+/fpc=
 github.com/fatih/structs v1.1.0 h1:Q7juDM0QtcnhCpeyLGQKyg4TOIghuNXrkL32pHAUMxo=
 github.com/fatih/structs v1.1.0/go.mod h1:9NiDSp5zOcgEDl+j00MP/WkGVPOlPRLejGD8Ga6PJ7M=
-github.com/flashcatcloud/ibex v1.3.5 h1:8GOOf5+aJT0TP/MC6izz7CO5JKJSdKVFBwL0vQp93Nc=
-github.com/flashcatcloud/ibex v1.3.5/go.mod h1:T8hbMUySK2q6cXUaYp0AUVeKkU9Od2LjzwmB5lmTRBM=
 github.com/gabriel-vasile/mimetype v1.4.2 h1:w5qFW6JKBz9Y393Y4q372O9A7cUSequkh1Q7OhCmWKU=
 github.com/gabriel-vasile/mimetype v1.4.2/go.mod h1:zApsH/mKG4w07erKIaJPFiX0Tsq9BFQgN3qGY5GnNgA=
 github.com/garyburd/redigo v1.6.2/go.mod h1:NR3MbYisc3/PwhQ00EMzDiPmrwpPxAn5GI05/YaO1SY=
@@ -262,6 +263,8 @@ github.com/rogpeppe/go-internal v1.10.0 h1:TMyTOH3F/DB16zRVcYyreMH6GnZZrwQVAoYjR
 github.com/rs/xid v1.2.1/go.mod h1:+uKXf+4Djp6Md1KODXJxgGQPKngRmWyn10oCKFzNHOQ=
 github.com/rs/zerolog v1.13.0/go.mod h1:YbFCdg8HfsridGWAh22vktObvhZbQsZXe4/zB0OKkWU=
 github.com/rs/zerolog v1.15.0/go.mod h1:xYTKnLHcpfU2225ny5qZjxnj9NvkumZYjJHlAThCjNc=
+github.com/russross/blackfriday/v2 v2.1.0 h1:JIOH55/0cWyOuilr9/qlrm0BSXldqnqwMsf35Ld67mk=
+github.com/russross/blackfriday/v2 v2.1.0/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
 github.com/satori/go.uuid v1.2.0/go.mod h1:dA0hQrYB0VpLJoorglMZABFdXlWrHn1NEOzdhQKdks0=
 github.com/shopspring/decimal v0.0.0-20180709203117-cd690d0c9e24/go.mod h1:M+9NzErvs504Cn4c5DxATwIqPbtswREoFCre64PpcG4=
 github.com/shopspring/decimal v1.2.0 h1:abSATXmQEYyShuxI4/vyW3tV1MrKAJzCZ/0zLUXYbsQ=
@@ -302,6 +305,10 @@ github.com/ugorji/go v1.2.7/go.mod h1:nF9osbDWLy6bDVv/Rtoh6QgnvNDpmCalQV5urGCCS6
 github.com/ugorji/go/codec v1.2.7/go.mod h1:WGN1fab3R1fzQlVQTkfxVtIBhWDRqOviHU95kRgeqEY=
 github.com/ugorji/go/codec v1.2.11 h1:BMaWp1Bb6fHwEtbplGBGJ498wD+LKlNSl25MjdZY4dU=
 github.com/ugorji/go/codec v1.2.11/go.mod h1:UNopzCgEMSXjBc6AOMqYvWC1ktqTAfzJZUZgYf6w6lg=
+github.com/urfave/cli/v2 v2.27.4 h1:o1owoI+02Eb+K107p27wEX9Bb8eqIoZCfLXloLUSWJ8=
+github.com/urfave/cli/v2 v2.27.4/go.mod h1:m4QzxcD2qpra4z7WhzEGn74WZLViBnMpb1ToCAKdGRQ=
+github.com/xrash/smetrics v0.0.0-20240521201337-686a1a2994c1 h1:gEOO8jv9F4OT7lGCjxCBTO/36wtF6j2nSip77qHd4x4=
+github.com/xrash/smetrics v0.0.0-20240521201337-686a1a2994c1/go.mod h1:Ohn+xnUBiLI6FVj/9LpzZWtj1/D6lUovWYBkxHVV3aM=
 github.com/yuin/goldmark v1.1.27/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 github.com/yuin/goldmark v1.2.1/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 github.com/yuin/goldmark v1.4.13/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY=

ibex/agentd/agentd.go

@@ -0,0 +1,117 @@
+package agentd
+
+import (
+	"context"
+	"fmt"
+	"log"
+	"os"
+	"os/signal"
+	"path/filepath"
+	"syscall"
+
+	"github.com/toolkits/pkg/i18n"
+
+	"github.com/ccfos/nightingale/v6/ibex/agentd/config"
+	"github.com/ccfos/nightingale/v6/ibex/agentd/router"
+	"github.com/ccfos/nightingale/v6/ibex/agentd/timer"
+	"github.com/ccfos/nightingale/v6/pkg/httpx"
+)
+
+type Agentd struct {
+	ConfigFile string
+	Version    string
+}
+
+type AgentdOption func(*Agentd)
+
+func SetConfigFile(f string) AgentdOption {
+	return func(s *Agentd) {
+		s.ConfigFile = f
+	}
+}
+
+func SetVersion(v string) AgentdOption {
+	return func(s *Agentd) {
+		s.Version = v
+	}
+}
+
+// Run run agentd
+func Run(opts ...AgentdOption) {
+	code := 1
+	sc := make(chan os.Signal, 1)
+	signal.Notify(sc, syscall.SIGHUP, syscall.SIGINT, syscall.SIGTERM, syscall.SIGQUIT)
+
+	agentd := Agentd{
+		ConfigFile: filepath.Join("etc", "ibex", "agentd.toml"),
+		Version:    "not specified",
+	}
+
+	for _, opt := range opts {
+		opt(&agentd)
+	}
+
+	cleanFunc, err := agentd.initialize()
+	if err != nil {
+		fmt.Println("agentd init fail:", err)
+		os.Exit(code)
+	}
+
+EXIT:
+	for {
+		sig := <-sc
+		fmt.Println("received signal:", sig.String())
+		switch sig {
+		case syscall.SIGQUIT, syscall.SIGTERM, syscall.SIGINT:
+			code = 0
+			break EXIT
+		case syscall.SIGHUP:
+			// reload configuration?
+		default:
+			break EXIT
+		}
+	}
+
+	cleanFunc()
+	fmt.Println("agentd exited")
+	os.Exit(code)
+}
+
+func (s Agentd) initialize() (func(), error) {
+	fns := Functions{}
+	ctx, cancel := context.WithCancel(context.Background())
+	fns.Add(cancel)
+
+	log.SetFlags(log.Ldate | log.Ltime | log.Lshortfile)
+
+	// parse config file
+	config.MustLoad(s.ConfigFile)
+
+	// init i18n
+	i18n.Init()
+
+	// init http server
+	r := router.New(s.Version)
+	httpClean := httpx.Init(config.C.HTTP, ctx, r)
+	fns.Add(httpClean)
+
+	go timer.Heartbeat(ctx)
+
+	return fns.Ret(), nil
+}
+
+type Functions struct {
+	List []func()
+}
+
+func (fs *Functions) Add(f func()) {
+	fs.List = append(fs.List, f)
+}
+
+func (fs *Functions) Ret() func() {
+	return func() {
+		for i := 0; i < len(fs.List); i++ {
+			fs.List[i]()
+		}
+	}
+}

ibex/agentd/client/cli.go

@@ -0,0 +1,110 @@
+package client
+
+import (
+	"bufio"
+	"io"
+	"log"
+	"net"
+	"net/rpc"
+	"reflect"
+	"time"
+
+	"github.com/toolkits/pkg/net/gobrpc"
+	"github.com/ugorji/go/codec"
+
+	"github.com/ccfos/nightingale/v6/ibex/agentd/config"
+)
+
+var cli *gobrpc.RPCClient
+
+func getCli() *gobrpc.RPCClient {
+	if cli != nil {
+		return cli
+	}
+
+	// detect the fastest server
+	var (
+		address  string
+		client   *rpc.Client
+		duration int64 = 999999999999
+	)
+
+	// auto close other slow server
+	acm := make(map[string]*rpc.Client)
+
+	l := len(config.C.Heartbeat.Servers)
+	for i := 0; i < l; i++ {
+		addr := config.C.Heartbeat.Servers[i]
+		begin := time.Now()
+		conn, err := net.DialTimeout("tcp", addr, time.Second*5)
+		if err != nil {
+			log.Printf("W: dial %s fail: %s", addr, err)
+			continue
+		}
+
+		var bufConn = struct {
+			io.Closer
+			*bufio.Reader
+			*bufio.Writer
+		}{conn, bufio.NewReader(conn), bufio.NewWriter(conn)}
+
+		var mh codec.MsgpackHandle
+		mh.MapType = reflect.TypeOf(map[string]interface{}(nil))
+
+		rpcCodec := codec.MsgpackSpecRpc.ClientCodec(bufConn, &mh)
+		c := rpc.NewClientWithCodec(rpcCodec)
+
+		acm[addr] = c
+
+		var out string
+		err = c.Call("Server.Ping", "", &out)
+		if err != nil {
+			log.Printf("W: ping %s fail: %s", addr, err)
+			continue
+		}
+		use := time.Since(begin).Nanoseconds()
+
+		if use < duration {
+			address = addr
+			client = c
+			duration = use
+		}
+	}
+
+	if address == "" {
+		log.Println("E: no job server found")
+		return nil
+	}
+
+	log.Printf("I: choose server: %s, duration: %dms", address, duration/1000000)
+
+	for addr, c := range acm {
+		if addr == address {
+			continue
+		}
+		c.Close()
+	}
+
+	cli = gobrpc.NewRPCClient(address, client, 5*time.Second)
+	return cli
+}
+
+// GetCli 探测所有server端的延迟，自动选择最快的
+func GetCli() *gobrpc.RPCClient {
+	for {
+		c := getCli()
+		if c != nil {
+			return c
+		}
+
+		time.Sleep(time.Second * 10)
+	}
+}
+
+// CloseCli 关闭客户端连接
+func CloseCli() {
+	if cli != nil {
+		cli.Close()
+		cli = nil
+	}
+}

ibex/agentd/client/meta.go

@@ -0,0 +1,31 @@
+package client
+
+import (
+	"fmt"
+	"log"
+
+	"github.com/ccfos/nightingale/v6/ibex/types"
+)
+
+// Meta 从Server端获取任务元信息
+func Meta(id int64) (script string, args string, account string, stdin string, err error) {
+	var resp types.TaskMetaResponse
+	err = GetCli().Call("Server.GetTaskMeta", id, &resp)
+	if err != nil {
+		log.Println("E: rpc call Server.GetTaskMeta:", err)
+		CloseCli()
+		return
+	}
+
+	if resp.Message != "" {
+		log.Println("E: rpc call Server.GetTaskMeta:", resp.Message)
+		err = fmt.Errorf(resp.Message)
+		return
+	}
+
+	script = resp.Script
+	args = resp.Args
+	account = resp.Account
+	stdin = resp.Stdin
+	return
+}

ibex/agentd/config/config.go

@@ -0,0 +1,140 @@
+package config
+
+import (
+	"fmt"
+	"log"
+	"net"
+	"os"
+	"strings"
+	"sync"
+
+	"github.com/koding/multiconfig"
+	"github.com/toolkits/pkg/file"
+
+	"github.com/ccfos/nightingale/v6/pkg/httpx"
+)
+
+var (
+	C    = new(Config)
+	once sync.Once
+)
+
+func MustLoad(fpaths ...string) {
+	once.Do(func() {
+		loaders := []multiconfig.Loader{
+			&multiconfig.TagLoader{},
+			&multiconfig.EnvironmentLoader{},
+		}
+
+		for _, fpath := range fpaths {
+			handled := false
+
+			if strings.HasSuffix(fpath, "toml") {
+				loaders = append(loaders, &multiconfig.TOMLLoader{Path: fpath})
+				handled = true
+			}
+			if strings.HasSuffix(fpath, "conf") {
+				loaders = append(loaders, &multiconfig.TOMLLoader{Path: fpath})
+				handled = true
+			}
+			if strings.HasSuffix(fpath, "json") {
+				loaders = append(loaders, &multiconfig.JSONLoader{Path: fpath})
+				handled = true
+			}
+			if strings.HasSuffix(fpath, "yaml") {
+				loaders = append(loaders, &multiconfig.YAMLLoader{Path: fpath})
+				handled = true
+			}
+
+			if !handled {
+				fmt.Println("config file invalid, valid file exts: .conf,.yaml,.toml,.json")
+				os.Exit(1)
+			}
+		}
+
+		m := multiconfig.DefaultLoader{
+			Loader:    multiconfig.MultiLoader(loaders...),
+			Validator: multiconfig.MultiValidator(&multiconfig.RequiredValidator{}),
+		}
+
+		m.MustLoad(C)
+
+		if C.Heartbeat.Host == "" {
+			fmt.Println("heartbeat.host is blank")
+			os.Exit(1)
+		}
+
+		if C.Heartbeat.Host == "$ip" {
+			C.Heartbeat.Endpoint = fmt.Sprint(GetOutboundIP())
+			if C.Heartbeat.Endpoint == "" {
+				fmt.Println("ip auto got is blank")
+				os.Exit(1)
+			}
+			fmt.Println("host.ip:", C.Heartbeat.Endpoint)
+		}
+
+		host, err := C.GetHost()
+		if err != nil {
+			log.Println("E: failed to GetHost:", err)
+			os.Exit(1)
+		}
+
+		fmt.Println("host:", host)
+
+		if C.MetaDir == "" {
+			C.MetaDir = "./meta"
+		}
+
+		C.MetaDir, err = file.RealPath(C.MetaDir)
+		if err != nil {
+			log.Println("E: failed to get real path of MetaDir:", err)
+			os.Exit(1)
+		}
+		file.EnsureDir(C.MetaDir)
+		file.EnsureDirRW(C.MetaDir)
+	})
+}
+
+type Config struct {
+	RunMode   string
+	MetaDir   string
+	Heartbeat Heartbeat
+	HTTP      httpx.Config
+}
+
+type Heartbeat struct {
+	Interval int64
+	Servers  []string
+	Host     string
+	Endpoint string
+}
+
+func (c *Config) IsDebugMode() bool {
+	return c.RunMode == "debug"
+}
+
+func (c *Config) GetHost() (string, error) {
+	if c.Heartbeat.Host == "$ip" {
+		return c.Heartbeat.Endpoint, nil
+	}
+
+	if c.Heartbeat.Host == "$hostname" {
+		return os.Hostname()
+	}
+
+	return c.Heartbeat.Host, nil
+}
+
+// Get preferred outbound ip of this machine
+func GetOutboundIP() net.IP {
+	conn, err := net.Dial("udp", "8.8.8.8:80")
+	if err != nil {
+		fmt.Println("auto get outbound ip fail:", err)
+		os.Exit(1)
+	}
+	defer conn.Close()
+
+	localAddr := conn.LocalAddr().(*net.UDPAddr)
+
+	return localAddr.IP
+}

ibex/agentd/router/router.go

@@ -0,0 +1,60 @@
+package router
+
+import (
+	"fmt"
+	"os"
+	"strings"
+
+	"github.com/gin-contrib/pprof"
+	"github.com/gin-gonic/gin"
+
+	"github.com/ccfos/nightingale/v6/ibex/agentd/config"
+	"github.com/ccfos/nightingale/v6/pkg/aop"
+)
+
+func New(version string) *gin.Engine {
+	gin.SetMode(config.C.RunMode)
+
+	loggerMid := aop.Logger()
+	recoveryMid := aop.Recovery()
+
+	if strings.ToLower(config.C.RunMode) == "release" {
+		aop.DisableConsoleColor()
+	}
+
+	r := gin.New()
+
+	r.Use(recoveryMid)
+
+	// whether print access log
+	if config.C.HTTP.PrintAccessLog {
+		r.Use(loggerMid)
+	}
+
+	configRoute(r, version)
+
+	return r
+}
+
+func configRoute(r *gin.Engine, version string) {
+	if config.C.HTTP.PProf {
+		pprof.Register(r, "/debug/pprof")
+	}
+
+	r.GET("/ping", func(c *gin.Context) {
+		c.String(200, "pong")
+	})
+
+	r.GET("/pid", func(c *gin.Context) {
+		c.String(200, fmt.Sprintf("%d", os.Getpid()))
+	})
+
+	r.GET("/addr", func(c *gin.Context) {
+		c.String(200, c.Request.RemoteAddr)
+	})
+
+	r.GET("/version", func(c *gin.Context) {
+		c.String(200, version)
+	})
+
+}

ibex/agentd/timer/cmd_nix.go

@@ -0,0 +1,18 @@
+//go:build !windows
+// +build !windows
+
+package timer
+
+import (
+	"os/exec"
+	"syscall"
+)
+
+func CmdStart(cmd *exec.Cmd) error {
+	cmd.SysProcAttr = &syscall.SysProcAttr{Setpgid: true}
+	return cmd.Start()
+}
+
+func CmdKill(cmd *exec.Cmd) error {
+	return syscall.Kill(-cmd.Process.Pid, syscall.SIGKILL)
+}

ibex/agentd/timer/cmd_windows.go

@@ -0,0 +1,16 @@
+//go:build windows
+// +build windows
+
+package timer
+
+import (
+	"os/exec"
+)
+
+func CmdStart(cmd *exec.Cmd) error {
+	return cmd.Start()
+}
+
+func CmdKill(cmd *exec.Cmd) error {
+	return cmd.Process.Kill()
+}

ibex/agentd/timer/heartbeat.go

@@ -0,0 +1,74 @@
+package timer
+
+import (
+	"context"
+	"log"
+	"time"
+
+	"github.com/ccfos/nightingale/v6/ibex/agentd/client"
+	"github.com/ccfos/nightingale/v6/ibex/agentd/config"
+	"github.com/ccfos/nightingale/v6/ibex/types"
+)
+
+func Heartbeat(ctx context.Context) {
+	interval := time.Duration(config.C.Heartbeat.Interval) * time.Millisecond
+	for {
+		select {
+		case <-ctx.Done():
+			return
+		case <-time.After(interval):
+			heartbeat()
+		}
+	}
+}
+
+func heartbeat() {
+	ident, err := config.C.GetHost()
+	if err != nil {
+		log.Println("E: GetHost fail:", err)
+		return
+	}
+
+	req := types.ReportRequest{
+		Ident:       ident,
+		ReportTasks: Locals.ReportTasks(),
+	}
+
+	var resp types.ReportResponse
+	err = client.GetCli().Call("Server.Report", req, &resp)
+	if err != nil {
+		log.Println("E: rpc call Server.Report fail:", err)
+		client.CloseCli()
+		return
+	}
+
+	if resp.Message != "" {
+		log.Println("E: error from server:", resp.Message)
+		return
+	}
+
+	assigned := make(map[int64]struct{})
+
+	if resp.AssignTasks != nil {
+		count := len(resp.AssignTasks)
+		for i := 0; i < count; i++ {
+			at := resp.AssignTasks[i]
+			assigned[at.Id] = struct{}{}
+			Locals.AssignTask(at)
+		}
+	}
+
+	if len(assigned) > 0 {
+		log.Println("D: assigned tasks:", mapKeys(assigned))
+	}
+
+	Locals.Clean(assigned)
+}
+
+func mapKeys(m map[int64]struct{}) []int64 {
+	lst := make([]int64, 0, len(m))
+	for k := range m {
+		lst = append(lst, k)
+	}
+	return lst
+}

ibex/agentd/timer/task.go

@@ -0,0 +1,333 @@
+package timer
+
+import (
+	"bytes"
+	"fmt"
+	"log"
+	"os/exec"
+	"os/user"
+	"path"
+	"strings"
+	"sync"
+
+	"github.com/toolkits/pkg/file"
+	"github.com/toolkits/pkg/runner"
+	"github.com/toolkits/pkg/sys"
+
+	"github.com/ccfos/nightingale/v6/ibex/agentd/client"
+	"github.com/ccfos/nightingale/v6/ibex/agentd/config"
+)
+
+type Task struct {
+	sync.Mutex
+
+	Id     int64
+	Clock  int64
+	Action string
+	Status string
+
+	alive  bool
+	Cmd    *exec.Cmd
+	Stdout bytes.Buffer
+	Stderr bytes.Buffer
+	Stdin  *bytes.Reader
+
+	Args     string
+	Account  string
+	StdinStr string
+}
+
+func (t *Task) SetStatus(status string) {
+	t.Lock()
+	t.Status = status
+	t.Unlock()
+}
+
+func (t *Task) GetStatus() string {
+	t.Lock()
+	s := t.Status
+	t.Unlock()
+	return s
+}
+
+func (t *Task) GetAlive() bool {
+	t.Lock()
+	pa := t.alive
+	t.Unlock()
+	return pa
+}
+
+func (t *Task) SetAlive(pa bool) {
+	t.Lock()
+	t.alive = pa
+	t.Unlock()
+}
+
+func (t *Task) GetStdout() string {
+	t.Lock()
+	out := t.Stdout.String()
+	t.Unlock()
+	return out
+}
+
+func (t *Task) GetStderr() string {
+	t.Lock()
+	out := t.Stderr.String()
+	t.Unlock()
+	return out
+}
+
+func (t *Task) ResetBuff() {
+	t.Lock()
+	t.Stdout.Reset()
+	t.Stderr.Reset()
+	t.Unlock()
+}
+
+func (t *Task) doneBefore() bool {
+	doneFlag := path.Join(config.C.MetaDir, fmt.Sprint(t.Id), fmt.Sprintf("%d.done", t.Clock))
+	return file.IsExist(doneFlag)
+}
+
+func (t *Task) loadResult() {
+	metadir := config.C.MetaDir
+
+	doneFlag := path.Join(metadir, fmt.Sprint(t.Id), fmt.Sprintf("%d.done", t.Clock))
+	stdoutFile := path.Join(metadir, fmt.Sprint(t.Id), "stdout")
+	stderrFile := path.Join(metadir, fmt.Sprint(t.Id), "stderr")
+
+	var err error
+
+	t.Status, err = file.ReadStringTrim(doneFlag)
+	if err != nil {
+		log.Printf("E: read file %s fail %v", doneFlag, err)
+	}
+	stdout, err := file.ReadString(stdoutFile)
+	if err != nil {
+		log.Printf("E: read file %s fail %v", stdoutFile, err)
+	}
+	stderr, err := file.ReadString(stderrFile)
+	if err != nil {
+		log.Printf("E: read file %s fail %v", stderrFile, err)
+	}
+
+	t.Stdout = *bytes.NewBufferString(stdout)
+	t.Stderr = *bytes.NewBufferString(stderr)
+}
+
+func (t *Task) prepare() error {
+	if t.Account != "" {
+		// already prepared
+		return nil
+	}
+
+	IdDir := path.Join(config.C.MetaDir, fmt.Sprint(t.Id))
+	err := file.EnsureDir(IdDir)
+	if err != nil {
+		log.Printf("E: mkdir -p %s fail: %v", IdDir, err)
+		return err
+	}
+
+	writeFlag := path.Join(IdDir, ".write")
+	if file.IsExist(writeFlag) {
+		// 从磁盘读取
+		argsFile := path.Join(IdDir, "args")
+		args, err := file.ReadStringTrim(argsFile)
+		if err != nil {
+			log.Printf("E: read %s fail %v", argsFile, err)
+			return err
+		}
+
+		accountFile := path.Join(IdDir, "account")
+		account, err := file.ReadStringTrim(accountFile)
+		if err != nil {
+			log.Printf("E: read %s fail %v", accountFile, err)
+			return err
+		}
+
+		stdinFile := path.Join(IdDir, "stdin")
+		stdin, err := file.ReadStringTrim(stdinFile)
+		if err != nil {
+			log.Printf("E: read %s fail %v", stdinFile, err)
+			return err
+		}
+
+		t.Args = args
+		t.Account = account
+		t.StdinStr = stdin
+
+	} else {
+		// 从远端读取，再写入磁盘
+		script, args, account, stdin, err := client.Meta(t.Id)
+		if err != nil {
+			log.Println("E: query task meta fail:", err)
+			return err
+		}
+
+		scriptFile := path.Join(IdDir, "script")
+		_, err = file.WriteString(scriptFile, script)
+		if err != nil {
+			log.Printf("E: write script to %s fail: %v", scriptFile, err)
+			return err
+		}
+
+		out, err := sys.CmdOutTrim("chmod", "+x", scriptFile)
+		if err != nil {
+			log.Printf("E: chmod +x %s fail %v. output: %s", scriptFile, err, out)
+			return err
+		}
+
+		argsFile := path.Join(IdDir, "args")
+		_, err = file.WriteString(argsFile, args)
+		if err != nil {
+			log.Printf("E: write args to %s fail: %v", argsFile, err)
+			return err
+		}
+
+		accountFile := path.Join(IdDir, "account")
+		_, err = file.WriteString(accountFile, account)
+		if err != nil {
+			log.Printf("E: write account to %s fail: %v", accountFile, err)
+			return err
+		}
+
+		stdinFile := path.Join(IdDir, "stdin")
+		_, err = file.WriteString(stdinFile, stdin)
+		if err != nil {
+			log.Printf("E: write tags to %s fail: %v", stdinFile, err)
+			return err
+		}
+
+		_, err = file.WriteString(writeFlag, "")
+		if err != nil {
+			log.Printf("E: create %s flag file fail: %v", writeFlag, err)
+			return err
+		}
+
+		t.Args = args
+		t.Account = account
+		t.StdinStr = stdin
+	}
+
+	t.Stdin = bytes.NewReader([]byte(t.StdinStr))
+
+	return nil
+}
+
+func (t *Task) start() {
+	if t.GetAlive() {
+		return
+	}
+
+	err := t.prepare()
+	if err != nil {
+		return
+	}
+
+	args := t.Args
+	if args != "" {
+		args = strings.Replace(args, ",,", "' '", -1)
+		args = "'" + args + "'"
+	}
+
+	scriptFile := path.Join(config.C.MetaDir, fmt.Sprint(t.Id), "script")
+	if !path.IsAbs(scriptFile) {
+		scriptFile = path.Join(runner.Cwd, scriptFile)
+	}
+
+	sh := fmt.Sprintf("%s %s", scriptFile, args)
+	var cmd *exec.Cmd
+
+	loginUser, err := user.Current()
+	if err != nil {
+		log.Println("E: cannot get current login user:", err)
+		return
+	}
+
+	if loginUser.Username == "root" {
+		// current login user is root
+		if t.Account == "root" {
+			cmd = exec.Command("sh", "-c", sh)
+			cmd.Dir = loginUser.HomeDir
+		} else {
+			cmd = exec.Command("su", "-c", sh, "-", t.Account)
+		}
+	} else {
+		// current login user not root
+		cmd = exec.Command("sh", "-c", sh)
+		cmd.Dir = loginUser.HomeDir
+	}
+
+	cmd.Stdout = &t.Stdout
+	cmd.Stderr = &t.Stderr
+	cmd.Stdin = t.Stdin
+
+	t.Cmd = cmd
+
+	err = CmdStart(cmd)
+	if err != nil {
+		log.Printf("E: cannot start cmd of task[%d]: %v", t.Id, err)
+		return
+	}
+
+	go runProcess(t)
+}
+
+func (t *Task) kill() {
+	go killProcess(t)
+}
+
+func runProcess(t *Task) {
+	t.SetAlive(true)
+	defer t.SetAlive(false)
+
+	err := t.Cmd.Wait()
+	if err != nil {
+		if strings.Contains(err.Error(), "signal: killed") {
+			t.SetStatus("killed")
+			log.Printf("D: process of task[%d] killed", t.Id)
+		} else if strings.Contains(err.Error(), "signal: terminated") {
+			// kill children process manually
+			t.SetStatus("killed")
+			log.Printf("D: process of task[%d] terminated", t.Id)
+		} else {
+			t.SetStatus("failed")
+			log.Printf("D: process of task[%d] return error: %v", t.Id, err)
+		}
+	} else {
+		t.SetStatus("success")
+		log.Printf("D: process of task[%d] done", t.Id)
+	}
+
+	persistResult(t)
+}
+
+func persistResult(t *Task) {
+	metadir := config.C.MetaDir
+
+	stdout := path.Join(metadir, fmt.Sprint(t.Id), "stdout")
+	stderr := path.Join(metadir, fmt.Sprint(t.Id), "stderr")
+	doneFlag := path.Join(metadir, fmt.Sprint(t.Id), fmt.Sprintf("%d.done", t.Clock))
+
+	file.WriteString(stdout, t.GetStdout())
+	file.WriteString(stderr, t.GetStderr())
+	file.WriteString(doneFlag, t.GetStatus())
+}
+
+func killProcess(t *Task) {
+	t.SetAlive(true)
+	defer t.SetAlive(false)
+
+	log.Printf("D: begin kill process of task[%d]", t.Id)
+
+	err := CmdKill(t.Cmd)
+	if err != nil {
+		t.SetStatus("killfailed")
+		log.Printf("D: kill process of task[%d] fail: %v", t.Id, err)
+	} else {
+		t.SetStatus("killed")
+		log.Printf("D: process of task[%d] killed", t.Id)
+	}
+
+	persistResult(t)
+}

ibex/agentd/timer/tasks.go

@@ -0,0 +1,120 @@
+package timer
+
+import (
+	"log"
+
+	"github.com/ccfos/nightingale/v6/ibex/types"
+)
+
+type LocalTasksT struct {
+	M map[int64]*Task
+}
+
+var Locals = &LocalTasksT{M: make(map[int64]*Task)}
+
+func (lt *LocalTasksT) ReportTasks() []types.ReportTask {
+	ret := make([]types.ReportTask, 0, len(lt.M))
+	for id, t := range lt.M {
+		rt := types.ReportTask{Id: id, Clock: t.Clock}
+
+		rt.Status = t.GetStatus()
+		if rt.Status == "running" || rt.Status == "killing" {
+			// intermediate state
+			continue
+		}
+
+		rt.Stdout = t.GetStdout()
+		rt.Stderr = t.GetStderr()
+
+		stdoutLen := len(rt.Stdout)
+		stderrLen := len(rt.Stderr)
+
+		// 输出太长的话，截断，要不然把数据库撑爆了
+		if stdoutLen > 65535 {
+			start := stdoutLen - 65535
+			rt.Stdout = rt.Stdout[start:]
+		}
+
+		if stderrLen > 65535 {
+			start := stderrLen - 65535
+			rt.Stderr = rt.Stderr[start:]
+		}
+
+		ret = append(ret, rt)
+	}
+
+	return ret
+}
+
+func (lt *LocalTasksT) GetTask(id int64) (*Task, bool) {
+	t, found := lt.M[id]
+	return t, found
+}
+
+func (lt *LocalTasksT) SetTask(t *Task) {
+	lt.M[t.Id] = t
+}
+
+func (lt *LocalTasksT) AssignTask(at types.AssignTask) {
+	local, found := lt.GetTask(at.Id)
+	if found {
+		if local.Clock == at.Clock && local.Action == at.Action {
+			// ignore repeat task
+			return
+		}
+
+		local.Clock = at.Clock
+		local.Action = at.Action
+	} else {
+		if at.Action == "kill" {
+			// no process in local, no need kill
+			return
+		}
+		local = &Task{
+			Id:     at.Id,
+			Clock:  at.Clock,
+			Action: at.Action,
+		}
+		lt.SetTask(local)
+
+		if local.doneBefore() {
+			local.loadResult()
+			return
+		}
+	}
+
+	if local.Action == "kill" {
+		local.SetStatus("killing")
+		local.kill()
+	} else if local.Action == "start" {
+		local.SetStatus("running")
+		local.start()
+	} else {
+		log.Printf("W: unknown action: %s of task %d", at.Action, at.Id)
+	}
+}
+
+func (lt *LocalTasksT) Clean(assigned map[int64]struct{}) {
+	del := make(map[int64]struct{})
+
+	for id := range lt.M {
+		if _, found := assigned[id]; !found {
+			del[id] = struct{}{}
+		}
+	}
+
+	for id := range del {
+		// 远端已经不关注这个任务了，但是本地来看，任务还是running的
+		// 可能是远端认为超时了，此时本地不能删除，仍然要继续上报
+		if lt.M[id].GetStatus() == "running" {
+			continue
+		}
+
+		lt.M[id].ResetBuff()
+		cmd := lt.M[id].Cmd
+		delete(lt.M, id)
+		if cmd != nil && cmd.Process != nil {
+			cmd.Process.Release()
+		}
+	}
+}

ibex/ibex.go

@@ -0,0 +1,82 @@
+package ibex
+
+import (
+	"fmt"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"os"
+	"strings"
+
+	"github.com/ccfos/nightingale/v6/ibex/server/config"
+	"github.com/ccfos/nightingale/v6/ibex/server/router"
+	"github.com/ccfos/nightingale/v6/ibex/server/rpc"
+	"github.com/ccfos/nightingale/v6/ibex/server/timer"
+	"github.com/ccfos/nightingale/v6/storage"
+
+	"github.com/ccfos/nightingale/v6/alert/aconf"
+	n9eRouter "github.com/ccfos/nightingale/v6/center/router"
+	"github.com/ccfos/nightingale/v6/conf"
+	n9eConf "github.com/ccfos/nightingale/v6/conf"
+	"github.com/gin-gonic/gin"
+	"github.com/redis/go-redis/v9"
+	"gorm.io/gorm"
+)
+
+var (
+	HttpPort int
+)
+
+func ServerStart(ctx *ctx.Context, isCenter bool, db *gorm.DB, rc redis.Cmdable, basicAuth gin.Accounts, heartbeat aconf.HeartbeatConfig,
+	api *n9eConf.CenterApi, r *gin.Engine, centerRouter *n9eRouter.Router, ibex conf.Ibex, httpPort int) {
+	config.C.IsCenter = isCenter
+	config.C.BasicAuth = make(gin.Accounts)
+	if len(basicAuth) > 0 {
+		config.C.BasicAuth = basicAuth
+	}
+
+	config.C.Heartbeat.IP = heartbeat.IP
+	config.C.Heartbeat.Interval = heartbeat.Interval
+	config.C.Heartbeat.LocalAddr = schedulerAddrGet(ibex.RPCListen)
+	HttpPort = httpPort
+
+	config.C.Output.ComeFrom = ibex.Output.ComeFrom
+	config.C.Output.AgtdPort = ibex.Output.AgtdPort
+
+	rou := router.NewRouter(ctx)
+
+	if centerRouter != nil {
+		rou.ConfigRouter(r, centerRouter)
+	} else {
+		rou.ConfigRouter(r)
+	}
+
+	ctx.Redis = rc
+	if err := storage.IdInit(ctx.Redis); err != nil {
+		fmt.Println("cannot init id generator: ", err)
+		os.Exit(1)
+	}
+
+	rpc.Start(ibex.RPCListen, ctx)
+
+	if isCenter {
+		go timer.Heartbeat(ctx)
+		go timer.Schedule(ctx)
+		go timer.CleanLong(ctx)
+	} else {
+		config.C.CenterApi = *api
+	}
+
+	timer.CacheHostDoing(ctx)
+	timer.ReportResult(ctx)
+}
+
+func schedulerAddrGet(rpcListen string) string {
+	ip := fmt.Sprint(config.GetOutboundIP())
+	if ip == "" {
+		fmt.Println("heartbeat ip auto got is blank")
+		os.Exit(1)
+	}
+
+	port := strings.Split(rpcListen, ":")[1]
+	localAddr := ip + ":" + port
+	return localAddr
+}

ibex/server/config/config.go

@@ -0,0 +1,135 @@
+package config
+
+import (
+	"fmt"
+	"net"
+	"os"
+	"strings"
+	"sync"
+
+	"github.com/ccfos/nightingale/v6/pkg/httpx"
+	"github.com/ccfos/nightingale/v6/pkg/logx"
+
+	"github.com/ccfos/nightingale/v6/conf"
+	"github.com/ccfos/nightingale/v6/pkg/ormx"
+	"github.com/ccfos/nightingale/v6/storage"
+	"github.com/gin-gonic/gin"
+	"github.com/koding/multiconfig"
+)
+
+var (
+	C    = new(Config)
+	once sync.Once
+)
+
+func MustLoad(fpaths ...string) {
+	once.Do(func() {
+		loaders := []multiconfig.Loader{
+			&multiconfig.TagLoader{},
+			&multiconfig.EnvironmentLoader{},
+		}
+
+		for _, fpath := range fpaths {
+			handled := false
+
+			if strings.HasSuffix(fpath, "toml") {
+				loaders = append(loaders, &multiconfig.TOMLLoader{Path: fpath})
+				handled = true
+			}
+			if strings.HasSuffix(fpath, "conf") {
+				loaders = append(loaders, &multiconfig.TOMLLoader{Path: fpath})
+				handled = true
+			}
+			if strings.HasSuffix(fpath, "json") {
+				loaders = append(loaders, &multiconfig.JSONLoader{Path: fpath})
+				handled = true
+			}
+			if strings.HasSuffix(fpath, "yaml") {
+				loaders = append(loaders, &multiconfig.YAMLLoader{Path: fpath})
+				handled = true
+			}
+
+			if !handled {
+				fmt.Println("config file invalid, valid file exts: .conf,.yaml,.toml,.json")
+				os.Exit(1)
+			}
+		}
+
+		m := multiconfig.DefaultLoader{
+			Loader:    multiconfig.MultiLoader(loaders...),
+			Validator: multiconfig.MultiValidator(&multiconfig.RequiredValidator{}),
+		}
+
+		m.MustLoad(C)
+
+		if C.Heartbeat.IP == "" {
+			// auto detect
+			C.Heartbeat.IP = fmt.Sprint(GetOutboundIP())
+
+			if C.Heartbeat.IP == "" {
+				fmt.Println("heartbeat ip auto got is blank")
+				os.Exit(1)
+			}
+		}
+
+		port := strings.Split(C.RPC.Listen, ":")[1]
+		endpoint := C.Heartbeat.IP + ":" + port
+		C.Heartbeat.LocalAddr = endpoint
+
+		// 正常情况肯定不是127.0.0.1，但是，如果就是单机部署，并且这个机器没有网络，比如本地调试并且本机没网的时候
+		// if C.Heartbeat.IP == "127.0.0.1" {
+		// 	fmt.Println("heartbeat ip is 127.0.0.1 and it is useless, so, exit")
+		// 	os.Exit(1)
+		// }
+
+		fmt.Println("heartbeat.ip:", C.Heartbeat.IP)
+		fmt.Printf("heartbeat.interval: %dms\n", C.Heartbeat.Interval)
+	})
+}
+
+type Config struct {
+	RunMode   string
+	RPC       RPC
+	Heartbeat Heartbeat
+	Output    Output
+	IsCenter  bool
+	CenterApi conf.CenterApi
+	Log       logx.Config
+	HTTP      httpx.Config
+	BasicAuth gin.Accounts
+	DB        ormx.DBConfig
+	Redis     storage.RedisConfig
+}
+
+type RPC struct {
+	Listen string
+}
+
+type Heartbeat struct {
+	IP        string
+	Interval  int64
+	LocalAddr string
+}
+
+type Output struct {
+	ComeFrom string
+	AgtdPort int
+}
+
+func (c *Config) IsDebugMode() bool {
+	return c.RunMode == "debug"
+}
+
+// Get preferred outbound ip of this machine
+func GetOutboundIP() net.IP {
+	conn, err := net.Dial("udp", "8.8.8.8:80")
+	if err != nil {
+		fmt.Println("auto get outbound ip fail:", err)
+		os.Exit(1)
+	}
+	defer conn.Close()
+
+	localAddr := conn.LocalAddr().(*net.UDPAddr)
+
+	return localAddr.IP
+}

ibex/server/logic/logic_scheduler.go

@@ -0,0 +1,144 @@
+package logic
+
+import (
+	"github.com/ccfos/nightingale/v6/models"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"github.com/toolkits/pkg/logger"
+	"github.com/toolkits/pkg/slice"
+	"github.com/toolkits/pkg/str"
+)
+
+func ScheduleTask(ctx *ctx.Context, id int64) {
+	logger.Debugf("task[%d] scheduling...", id)
+
+	count, err := models.WaitingHostCount(ctx, id)
+	if err != nil {
+		logger.Errorf("cannot get task[%d] waiting host count: %v", id, err)
+		return
+	}
+
+	if count == 0 {
+		cleanDoneTask(ctx, id)
+		return
+	}
+
+	action, err := models.TaskActionGet(ctx, "id=?", id)
+	if err != nil {
+		logger.Errorf("cannot get task[%d] action: %v", id, err)
+		return
+	}
+
+	if action == nil {
+		logger.Errorf("[W] no action found of task[%d]", id)
+		return
+	}
+
+	switch action.Action {
+	case "start":
+		startTask(ctx, id, action)
+	case "pause":
+		return
+	case "cancel":
+		return
+	case "kill":
+		return
+	default:
+		logger.Errorf("unknown action: %s of task[%d]", action.Action, id)
+	}
+}
+
+func cleanDoneTask(ctx *ctx.Context, id int64) {
+	ingCount, err := models.IngStatusHostCount(ctx, id)
+	if err != nil {
+		logger.Errorf("cannot get task[%d] ing status host count: %v", id, err)
+		return
+	}
+
+	if ingCount > 0 {
+		return
+	}
+
+	err = models.CleanDoneTask(ctx, id)
+	if err != nil {
+		logger.Errorf("cannot clean done task[%d]: %v", id, err)
+	}
+
+	logger.Debugf("task[%d] done", id)
+}
+
+func startTask(ctx *ctx.Context, id int64, action *models.TaskAction) {
+	meta, err := models.TaskMetaGetByID(ctx, id)
+	if err != nil {
+		logger.Errorf("cannot get task[%d] meta: %v", id, err)
+		return
+	}
+
+	if meta == nil {
+		logger.Errorf("task[%d] meta lost", id)
+		return
+	}
+
+	count, err := models.UnexpectedHostCount(ctx, id)
+	if err != nil {
+		logger.Errorf("cannot get task[%d] unexpected host count: %v", id, err)
+		return
+	}
+
+	if count > int64(meta.Tolerance) {
+		err = action.Update(ctx, "pause")
+		if err != nil {
+			logger.Errorf("cannot update task[%d] action to 'pause': %v", id, err)
+		}
+
+		return
+	}
+
+	waitings, err := models.WaitingHostList(ctx, id)
+	if err != nil {
+		logger.Errorf("cannot get task[%d] waiting host: %v", id, err)
+		return
+	}
+
+	waitingsCount := len(waitings)
+	if waitingsCount == 0 {
+		return
+	}
+
+	doingsCount, err := models.TableRecordCount(ctx, models.TaskHostDoing{}.TableName(), "id=?", id)
+	if err != nil {
+		logger.Errorf("cannot get task[%d] doing host count: %v", id, err)
+		return
+	}
+
+	need := meta.Batch - int(doingsCount)
+	if meta.Batch == 0 {
+		need = waitingsCount
+	}
+
+	if need <= 0 {
+		return
+	}
+
+	if need > waitingsCount {
+		need = waitingsCount
+	}
+
+	arr := str.ParseCommaTrim(meta.Pause)
+	end := need
+	for i := 0; i < need; i++ {
+		if slice.ContainsString(arr, waitings[i].Host) {
+			end = i + 1
+			err = action.Update(ctx, "pause")
+			if err != nil {
+				logger.Errorf("cannot update task[%d] action to 'pause': %v", id, err)
+				return
+			}
+			break
+		}
+	}
+
+	err = models.RunWaitingHosts(ctx, waitings[:end])
+	if err != nil {
+		logger.Errorf("cannot run waiting hosts: %v", err)
+	}
+}

ibex/server/logic/logic_timeout.go

@@ -0,0 +1,45 @@
+package logic
+
+import (
+	"github.com/ccfos/nightingale/v6/models"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"time"
+
+	"github.com/toolkits/pkg/logger"
+)
+
+func CheckTimeout(ctx *ctx.Context, id int64) {
+	meta, err := models.TaskMetaGetByID(ctx, id)
+	if err != nil {
+		logger.Errorf("cannot get task[%d] meta: %v", id, err)
+		return
+	}
+
+	if meta == nil {
+		logger.Errorf("task[%d] meta lost", id)
+		return
+	}
+
+	hosts, err := models.TableRecordGets[[]models.TaskHostDoing](ctx, models.TaskHostDoing{}.TableName(), "id=?", id)
+	if err != nil {
+		logger.Errorf("cannot get task[%d] doing host list: %v", id, err)
+		return
+	}
+
+	count := len(hosts)
+	if count == 0 {
+		return
+	}
+
+	// 3s: task dispatch duration: web -> db -> scheduler -> executor
+	timeout := int64(meta.Timeout + 3)
+	now := time.Now().Unix()
+	for i := 0; i < count; i++ {
+		if now-hosts[i].Clock > timeout {
+			err = models.MarkDoneStatus(ctx, hosts[i].Id, hosts[i].Clock, hosts[i].Host, "timeout", "", "")
+			if err != nil {
+				logger.Errorf("cannot mark task[%d] done status: %v", id, err)
+			}
+		}
+	}
+}

ibex/server/router/funcs.go

@@ -0,0 +1,40 @@
+package router
+
+import (
+	"github.com/ccfos/nightingale/v6/models"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"net/http"
+	"strings"
+
+	"github.com/toolkits/pkg/errorx"
+)
+
+func TaskMeta(ctx *ctx.Context, id int64) *models.TaskMeta {
+	obj, err := models.TaskMetaGet(ctx, "id = ?", id)
+	errorx.Dangerous(err)
+
+	if obj == nil {
+		errorx.Bomb(http.StatusNotFound, "no such task meta")
+	}
+
+	return obj
+}
+
+func cleanHosts(formHosts []string) []string {
+	cnt := len(formHosts)
+	arr := make([]string, 0, cnt)
+	for i := 0; i < cnt; i++ {
+		item := strings.TrimSpace(formHosts[i])
+		if item == "" {
+			continue
+		}
+
+		if strings.HasPrefix(item, "#") {
+			continue
+		}
+
+		arr = append(arr, item)
+	}
+
+	return arr
+}

ibex/server/router/handler.go

@@ -0,0 +1,612 @@
+package router
+
+import (
+	"fmt"
+	"github.com/ccfos/nightingale/v6/models"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"strconv"
+
+	"io/ioutil"
+	"net/http"
+	"time"
+
+	"github.com/ccfos/nightingale/v6/ibex/server/config"
+	"github.com/ccfos/nightingale/v6/storage"
+
+	"github.com/gin-gonic/gin"
+	"github.com/toolkits/pkg/errorx"
+	"github.com/toolkits/pkg/ginx"
+	"github.com/toolkits/pkg/logger"
+	"github.com/toolkits/pkg/slice"
+	"github.com/toolkits/pkg/str"
+)
+
+func (rou *Router) taskStdout(c *gin.Context) {
+	meta := TaskMeta(rou.ctx, UrlParamsInt64(c, "id"))
+	stdouts, err := meta.Stdouts(rou.ctx)
+	ginx.NewRender(c).Data(stdouts, err)
+}
+
+func (rou *Router) taskStderr(c *gin.Context) {
+	meta := TaskMeta(rou.ctx, UrlParamsInt64(c, "id"))
+	stderrs, err := meta.Stderrs(rou.ctx)
+	ginx.NewRender(c).Data(stderrs, err)
+}
+
+// TODO: 不能只判断task_action，还应该看所有的host执行情况
+func (rou *Router) taskState(c *gin.Context) {
+	action, err := models.TaskActionGet(rou.ctx, "id=?", UrlParamsInt64(c, "id"))
+	if err != nil {
+		ginx.NewRender(c).Data("", err)
+		return
+	}
+
+	state := "done"
+	if action != nil {
+		state = action.Action
+	}
+
+	ginx.NewRender(c).Data(state, err)
+}
+
+func (rou *Router) taskResult(c *gin.Context) {
+	id := UrlParamsInt64(c, "id")
+
+	hosts, err := models.TaskHostStatus(rou.ctx, id)
+	if err != nil {
+		errorx.Bomb(500, "load task hosts of %d occur error %v", id, err)
+	}
+
+	ss := make(map[string][]string)
+	total := len(hosts)
+	for i := 0; i < total; i++ {
+		s := hosts[i].Status
+		ss[s] = append(ss[s], hosts[i].Host)
+	}
+
+	ginx.NewRender(c).Data(ss, nil)
+}
+
+func (rou *Router) taskHostOutput(c *gin.Context) {
+	obj, err := models.TaskHostGet(rou.ctx, UrlParamsInt64(c, "id"), ginx.UrlParamStr(c, "host"))
+	ginx.NewRender(c).Data(obj, err)
+}
+
+func (rou *Router) taskHostStdout(c *gin.Context) {
+	id := UrlParamsInt64(c, "id")
+	host := ginx.UrlParamStr(c, "host")
+
+	if config.C.Output.ComeFrom == "database" || config.C.Output.ComeFrom == "" {
+		obj, err := models.TaskHostGet(rou.ctx, id, host)
+		ginx.NewRender(c).Data(obj.Stdout, err)
+		return
+	}
+
+	if config.C.Output.AgtdPort <= 0 || config.C.Output.AgtdPort > 65535 {
+		ginx.NewRender(c).Message(fmt.Errorf("remotePort(%d) invalid", config.C.Output.AgtdPort))
+		return
+	}
+
+	url := fmt.Sprintf("http://%s:%d/output/%d/stdout.json", host, config.C.Output.AgtdPort, id)
+	client := &http.Client{
+		CheckRedirect: func(req *http.Request, via []*http.Request) error {
+			return http.ErrUseLastResponse
+		},
+	}
+	resp, err := client.Get(url)
+	errorx.Dangerous(err)
+
+	defer resp.Body.Close()
+
+	bs, err := ioutil.ReadAll(resp.Body)
+	errorx.Dangerous(err)
+
+	c.Writer.Header().Set("Content-Type", "application/json; charset=UTF-8")
+	c.Writer.Write(bs)
+}
+
+func (rou *Router) taskHostStderr(c *gin.Context) {
+	id := UrlParamsInt64(c, "id")
+	host := ginx.UrlParamStr(c, "host")
+
+	if config.C.Output.ComeFrom == "database" || config.C.Output.ComeFrom == "" {
+		obj, err := models.TaskHostGet(rou.ctx, id, host)
+		ginx.NewRender(c).Data(obj.Stderr, err)
+		return
+	}
+
+	if config.C.Output.AgtdPort <= 0 || config.C.Output.AgtdPort > 65535 {
+		ginx.NewRender(c).Message(fmt.Errorf("remotePort(%d) invalid", config.C.Output.AgtdPort))
+		return
+	}
+
+	url := fmt.Sprintf("http://%s:%d/output/%d/stderr.json", host, config.C.Output.AgtdPort, id)
+	client := &http.Client{
+		CheckRedirect: func(req *http.Request, via []*http.Request) error {
+			return http.ErrUseLastResponse
+		},
+	}
+	resp, err := client.Get(url)
+	errorx.Dangerous(err)
+
+	defer resp.Body.Close()
+
+	bs, err := ioutil.ReadAll(resp.Body)
+	errorx.Dangerous(err)
+
+	c.Writer.Header().Set("Content-Type", "application/json; charset=UTF-8")
+	c.Writer.Write(bs)
+}
+
+func (rou *Router) taskStdoutTxt(c *gin.Context) {
+	id := UrlParamsInt64(c, "id")
+
+	meta, err := models.TaskMetaGet(rou.ctx, "id = ?", id)
+	if err != nil {
+		c.String(500, err.Error())
+		return
+	}
+
+	if meta == nil {
+		c.String(404, "no such task")
+		return
+	}
+
+	stdouts, err := meta.Stdouts(rou.ctx)
+	if err != nil {
+		c.String(500, err.Error())
+		return
+	}
+
+	w := c.Writer
+
+	w.Header().Set("Content-Type", "text/plain; charset=utf-8")
+	count := len(stdouts)
+	for i := 0; i < count; i++ {
+		if i != 0 {
+			w.Write([]byte("\n\n"))
+		}
+
+		w.Write([]byte(stdouts[i].Host + ":\n"))
+		w.Write([]byte(stdouts[i].Stdout))
+	}
+}
+
+func (rou *Router) taskStderrTxt(c *gin.Context) {
+	id := UrlParamsInt64(c, "id")
+
+	meta, err := models.TaskMetaGet(rou.ctx, "id = ?", id)
+	if err != nil {
+		c.String(500, err.Error())
+		return
+	}
+
+	if meta == nil {
+		c.String(404, "no such task")
+		return
+	}
+
+	stderrs, err := meta.Stderrs(rou.ctx)
+	if err != nil {
+		c.String(500, err.Error())
+		return
+	}
+
+	w := c.Writer
+
+	w.Header().Set("Content-Type", "text/plain; charset=utf-8")
+	count := len(stderrs)
+	for i := 0; i < count; i++ {
+		if i != 0 {
+			w.Write([]byte("\n\n"))
+		}
+
+		w.Write([]byte(stderrs[i].Host + ":\n"))
+		w.Write([]byte(stderrs[i].Stderr))
+	}
+}
+
+type TaskStdoutData struct {
+	Host   string `json:"host"`
+	Stdout string `json:"stdout"`
+}
+
+type TaskStderrData struct {
+	Host   string `json:"host"`
+	Stderr string `json:"stderr"`
+}
+
+func (rou *Router) taskStdoutJSON(c *gin.Context) {
+	task := TaskMeta(rou.ctx, UrlParamsInt64(c, "id"))
+
+	host := ginx.QueryStr(c, "host", "")
+
+	var hostsLen int
+	var ret []TaskStdoutData
+
+	if host != "" {
+		obj, err := models.TaskHostGet(rou.ctx, task.Id, host)
+		if err != nil {
+			ginx.NewRender(c).Data("", err)
+			return
+		} else if obj == nil {
+			ginx.NewRender(c).Data("", fmt.Errorf("task: %d, host(%s) not eixsts", task.Id, host))
+			return
+		} else {
+			ret = append(ret, TaskStdoutData{
+				Host:   host,
+				Stdout: obj.Stdout,
+			})
+		}
+	} else {
+		hosts, err := models.TaskHostGets(rou.ctx, task.Id)
+		if err != nil {
+			ginx.NewRender(c).Data("", err)
+			return
+		}
+
+		hostsLen = len(hosts)
+
+		ret = make([]TaskStdoutData, 0, hostsLen)
+		for i := 0; i < hostsLen; i++ {
+			ret = append(ret, TaskStdoutData{
+				Host:   hosts[i].Host,
+				Stdout: hosts[i].Stdout,
+			})
+		}
+	}
+
+	ginx.NewRender(c).Data(ret, nil)
+}
+
+func (rou *Router) taskStderrJSON(c *gin.Context) {
+	task := TaskMeta(rou.ctx, UrlParamsInt64(c, "id"))
+
+	host := ginx.QueryStr(c, "host", "")
+
+	var hostsLen int
+	var ret []TaskStderrData
+
+	if host != "" {
+		obj, err := models.TaskHostGet(rou.ctx, task.Id, host)
+		if err != nil {
+			ginx.NewRender(c).Data("", err)
+			return
+		} else if obj == nil {
+			ginx.NewRender(c).Data("", fmt.Errorf("task: %d, host(%s) not eixsts", task.Id, host))
+			return
+		} else {
+			ret = append(ret, TaskStderrData{
+				Host:   host,
+				Stderr: obj.Stderr,
+			})
+		}
+	} else {
+		hosts, err := models.TaskHostGets(rou.ctx, task.Id)
+		if err != nil {
+			ginx.NewRender(c).Data("", err)
+			return
+		}
+
+		hostsLen = len(hosts)
+
+		ret = make([]TaskStderrData, 0, hostsLen)
+		for i := 0; i < hostsLen; i++ {
+			ret = append(ret, TaskStderrData{
+				Host:   hosts[i].Host,
+				Stderr: hosts[i].Stderr,
+			})
+		}
+	}
+
+	ginx.NewRender(c).Data(ret, nil)
+}
+
+type taskForm struct {
+	Title          string   `json:"title" binding:"required"`
+	Account        string   `json:"account" binding:"required"`
+	Batch          int      `json:"batch"`
+	Tolerance      int      `json:"tolerance"`
+	Timeout        int      `json:"timeout"`
+	Pause          string   `json:"pause"`
+	Script         string   `json:"script" binding:"required"`
+	Args           string   `json:"args"`
+	Stdin          string   `json:"stdin"`
+	Action         string   `json:"action" binding:"required"`
+	Creator        string   `json:"creator" binding:"required"`
+	Hosts          []string `json:"hosts" binding:"required"`
+	AlertTriggered bool     `json:"alert_triggered"`
+}
+
+func (rou *Router) taskAdd(c *gin.Context) {
+	var f taskForm
+	ginx.BindJSON(c, &f)
+
+	hosts := cleanHosts(f.Hosts)
+	if len(hosts) == 0 {
+		errorx.Bomb(http.StatusBadRequest, "arg(hosts) empty")
+	}
+
+	taskMeta := &models.TaskMeta{
+		Title:     f.Title,
+		Account:   f.Account,
+		Batch:     f.Batch,
+		Tolerance: f.Tolerance,
+		Timeout:   f.Timeout,
+		Pause:     f.Pause,
+		Script:    f.Script,
+		Args:      f.Args,
+		Stdin:     f.Stdin,
+		Creator:   f.Creator,
+	}
+
+	err := taskMeta.CleanFields()
+	ginx.Dangerous(err)
+	taskMeta.HandleFH(hosts[0])
+
+	authUser := c.MustGet(gin.AuthUserKey).(string)
+	// 任务类型分为"告警规则触发"和"n9e center用户下发"两种；
+	// 边缘机房"告警规则触发"的任务不需要规划，并且它可能是失联的，无法使用db资源，所以放入redis缓存中，直接下发给agentd执行
+	if !config.C.IsCenter && f.AlertTriggered {
+		if err := taskMeta.Create(rou.ctx); err != nil {
+			// 当网络不连通时，生成唯一的id，防止边缘机房中不同任务的id相同；
+			// 方法是，redis自增id去防止同一个机房的不同n9e edge生成的id相同；
+			// 但没法防止不同边缘机房生成同样的id，所以，生成id的数据不会上报存入数据库，只用于闭环执行。
+			taskMeta.Id, err = storage.IdGet(rou.ctx.Redis)
+			ginx.Dangerous(err)
+		}
+		if err == nil {
+			taskHost := models.TaskHost{
+				Id:     taskMeta.Id,
+				Host:   hosts[0],
+				Status: "running",
+			}
+			if err = taskHost.Create(rou.ctx); err != nil {
+				logger.Warningf("task_add_fail: authUser=%s title=%s err=%s", authUser, taskMeta.Title, err.Error())
+			}
+		}
+
+		// 缓存任务元信息和待下发的任务
+		err = taskMeta.Cache(rou.ctx, hosts[0])
+		ginx.Dangerous(err)
+
+	} else {
+		// 如果是中心机房，还是保持之前的逻辑
+		err = taskMeta.Save(rou.ctx, hosts, f.Action)
+		ginx.Dangerous(err)
+	}
+
+	logger.Infof("task_add_succ: authUser=%s title=%s", authUser, taskMeta.Title)
+
+	ginx.NewRender(c).Data(taskMeta.Id, err)
+}
+
+func (rou *Router) taskGet(c *gin.Context) {
+	meta := TaskMeta(rou.ctx, UrlParamsInt64(c, "id"))
+
+	hosts, err := meta.Hosts(rou.ctx)
+	errorx.Dangerous(err)
+
+	action, err := meta.Action(rou.ctx)
+	errorx.Dangerous(err)
+
+	actionStr := ""
+	if action != nil {
+		actionStr = action.Action
+	} else {
+		meta.Done = true
+	}
+
+	ginx.NewRender(c).Data(gin.H{
+		"meta":   meta,
+		"hosts":  hosts,
+		"action": actionStr,
+	}, nil)
+}
+
+// 传进来一堆ids，返回已经done的任务的ids
+func (rou *Router) doneIds(c *gin.Context) {
+	ids := ginx.QueryStr(c, "ids", "")
+	if ids == "" {
+		errorx.Dangerous("arg(ids) empty")
+	}
+
+	idsint64 := str.IdsInt64(ids, ",")
+	if len(idsint64) == 0 {
+		errorx.Dangerous("arg(ids) empty")
+	}
+
+	exists, err := models.TaskActionExistsIds(rou.ctx, idsint64)
+	errorx.Dangerous(err)
+
+	dones := slice.SubInt64(idsint64, exists)
+	ginx.NewRender(c).Data(gin.H{
+		"list": dones,
+	}, nil)
+}
+
+func (rou *Router) taskGets(c *gin.Context) {
+	query := ginx.QueryStr(c, "query", "")
+	limit := ginx.QueryInt(c, "limit", 20)
+	creator := ginx.QueryStr(c, "creator", "")
+	days := ginx.QueryInt64(c, "days", 7)
+
+	before := time.Unix(time.Now().Unix()-days*24*3600, 0)
+
+	total, err := models.TaskMetaTotal(rou.ctx, creator, query, before)
+	errorx.Dangerous(err)
+
+	list, err := models.TaskMetaGets(rou.ctx, creator, query, before, limit, ginx.Offset(c, limit))
+	errorx.Dangerous(err)
+
+	cnt := len(list)
+	ids := make([]int64, cnt)
+	for i := 0; i < cnt; i++ {
+		ids[i] = list[i].Id
+	}
+
+	exists, err := models.TaskActionExistsIds(rou.ctx, ids)
+	errorx.Dangerous(err)
+
+	for i := 0; i < cnt; i++ {
+		if slice.ContainsInt64(exists, list[i].Id) {
+			list[i].Done = false
+		} else {
+			list[i].Done = true
+		}
+	}
+
+	ginx.NewRender(c).Data(gin.H{
+		"total": total,
+		"list":  list,
+	}, nil)
+}
+
+type actionForm struct {
+	Action string `json:"action"`
+}
+
+func (rou *Router) taskAction(c *gin.Context) {
+	meta := TaskMeta(rou.ctx, UrlParamsInt64(c, "id"))
+
+	var f actionForm
+	ginx.BindJSON(c, &f)
+
+	action, err := models.TaskActionGet(rou.ctx, "id=?", meta.Id)
+	errorx.Dangerous(err)
+
+	if action == nil {
+		errorx.Bomb(200, "task already finished, no more action can do")
+	}
+
+	ginx.NewRender(c).Message(action.Update(rou.ctx, f.Action))
+}
+
+func (rou *Router) taskHostAction(c *gin.Context) {
+	host := ginx.UrlParamStr(c, "host")
+	meta := TaskMeta(rou.ctx, UrlParamsInt64(c, "id"))
+
+	noopWhenDone(rou.ctx, meta.Id)
+
+	var f actionForm
+	ginx.BindJSON(c, &f)
+
+	if f.Action == "ignore" {
+		errorx.Dangerous(meta.IgnoreHost(rou.ctx, host))
+
+		action, err := models.TaskActionGet(rou.ctx, "id=?", meta.Id)
+		errorx.Dangerous(err)
+
+		if action != nil && action.Action == "pause" {
+			ginx.NewRender(c).Data("you can click start to run the task", nil)
+			return
+		}
+	}
+
+	if f.Action == "kill" {
+		errorx.Dangerous(meta.KillHost(rou.ctx, host))
+	}
+
+	if f.Action == "redo" {
+		errorx.Dangerous(meta.RedoHost(rou.ctx, host))
+	}
+
+	ginx.NewRender(c).Message(nil)
+}
+
+func noopWhenDone(ctx *ctx.Context, id int64) {
+	action, err := models.TaskActionGet(ctx, "id=?", id)
+	errorx.Dangerous(err)
+
+	if action == nil {
+		errorx.Bomb(200, "task already finished, no more taskAction can do")
+	}
+}
+
+type sqlCondForm struct {
+	Table string
+	Where string
+	Args  []interface{}
+}
+
+func (rou *Router) tableRecordListGet(c *gin.Context) {
+	var f sqlCondForm
+	ginx.BindJSON(c, &f)
+	switch f.Table {
+	case models.TaskHostDoing{}.TableName():
+		lst, err := models.TableRecordGets[[]models.TaskHostDoing](rou.ctx, f.Table, f.Where, f.Args)
+		ginx.NewRender(c).Data(lst, err)
+	case models.TaskMeta{}.TableName():
+		lst, err := models.TableRecordGets[[]models.TaskMeta](rou.ctx, f.Table, f.Where, f.Args)
+		ginx.NewRender(c).Data(lst, err)
+	default:
+		ginx.Bomb(http.StatusBadRequest, "table[%v] not support", f.Table)
+	}
+}
+
+func (rou *Router) tableRecordCount(c *gin.Context) {
+	var f sqlCondForm
+	ginx.BindJSON(c, &f)
+	ginx.NewRender(c).Data(models.TableRecordCount(rou.ctx, f.Table, f.Where, f.Args))
+}
+
+type markDoneForm struct {
+	Id     int64
+	Clock  int64
+	Host   string
+	Status string
+	Stdout string
+	Stderr string
+}
+
+func (rou *Router) markDone(c *gin.Context) {
+	var f markDoneForm
+	ginx.BindJSON(c, &f)
+	ginx.NewRender(c).Message(models.MarkDoneStatus(rou.ctx, f.Id, f.Clock, f.Host, f.Status, f.Stdout, f.Stderr))
+}
+
+func (rou *Router) taskMetaAdd(c *gin.Context) {
+	var f models.TaskMeta
+	ginx.BindJSON(c, &f)
+	err := f.Create(rou.ctx)
+	ginx.NewRender(c).Data(f.Id, err)
+}
+
+func (rou *Router) taskHostAdd(c *gin.Context) {
+	var f models.TaskHost
+	ginx.BindJSON(c, &f)
+	ginx.NewRender(c).Message(f.Upsert(rou.ctx))
+}
+
+func (rou *Router) taskHostUpsert(c *gin.Context) {
+	var f []models.TaskHost
+	ginx.BindJSON(c, &f)
+	ginx.NewRender(c).Data(models.TaskHostUpserts(rou.ctx, f))
+}
+
+func UrlParamsInt64(c *gin.Context, field string) int64 {
+
+	var params []gin.Param
+	for _, p := range c.Params {
+		if p.Key == "id" {
+			params = append(params, p)
+		}
+	}
+
+	var strval string
+	if len(params) == 1 {
+		strval = ginx.UrlParamStr(c, field)
+	} else if len(params) == 2 {
+		strval = params[1].Value
+	} else {
+		logger.Warningf("url param[%+v] not ok", params)
+		errorx.Bomb(http.StatusBadRequest, "url param[%s] is blank", field)
+	}
+
+	intval, err := strconv.ParseInt(strval, 10, 64)
+	if err != nil {
+		errorx.Bomb(http.StatusBadRequest, "cannot convert %s to int64", strval)
+	}
+
+	return intval
+}

ibex/server/router/router.go

@@ -0,0 +1,132 @@
+package router
+
+import (
+	"fmt"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+
+	"os"
+	"strings"
+
+	"github.com/ccfos/nightingale/v6/ibex/server/config"
+	"github.com/ccfos/nightingale/v6/pkg/aop"
+
+	"github.com/ccfos/nightingale/v6/center/router"
+	"github.com/gin-contrib/pprof"
+	"github.com/gin-gonic/gin"
+)
+
+func New(ctx *ctx.Context, version string) *gin.Engine {
+	gin.SetMode(config.C.RunMode)
+
+	loggerMid := aop.Logger()
+	recoveryMid := aop.Recovery()
+
+	if strings.ToLower(config.C.RunMode) == "release" {
+		aop.DisableConsoleColor()
+	}
+
+	r := gin.New()
+
+	r.Use(recoveryMid)
+
+	// whether print access log
+	if config.C.HTTP.PrintAccessLog {
+		r.Use(loggerMid)
+	}
+
+	rou := NewRouter(ctx)
+
+	rou.configBaseRouter(r, version)
+	rou.ConfigRouter(r)
+
+	return r
+}
+
+type Router struct {
+	ctx *ctx.Context
+}
+
+func NewRouter(ctx *ctx.Context) *Router {
+	return &Router{
+		ctx: ctx,
+	}
+}
+
+func (rou *Router) configBaseRouter(r *gin.Engine, version string) {
+	if config.C.HTTP.PProf {
+		pprof.Register(r, "/debug/pprof")
+	}
+
+	r.GET("/ping", func(c *gin.Context) {
+		c.String(200, "pong")
+	})
+
+	r.GET("/pid", func(c *gin.Context) {
+		c.String(200, fmt.Sprintf("%d", os.Getpid()))
+	})
+
+	r.GET("/addr", func(c *gin.Context) {
+		c.String(200, c.Request.RemoteAddr)
+	})
+
+	r.GET("/version", func(c *gin.Context) {
+		c.String(200, version)
+	})
+}
+
+func (rou *Router) ConfigRouter(r *gin.Engine, rts ...*router.Router) {
+
+	if len(rts) > 0 {
+		rt := rts[0]
+		pagesPrefix := "/api/n9e/busi-group/:id"
+		pages := r.Group(pagesPrefix)
+		{
+			pages.GET("/task/:id", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskGet)
+			pages.PUT("/task/:id/action", rt.Auth(), rt.User(), rt.Perm("/job-tasks/put"), rt.Bgrw(), rou.taskAction)
+			pages.GET("/task/:id/stdout", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskStdout)
+			pages.GET("/task/:id/stderr", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskStderr)
+			pages.GET("/task/:id/state", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskState)
+			pages.GET("/task/:id/result", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskResult)
+			pages.PUT("/task/:id/host/:host/action", rt.Auth(), rt.User(), rt.Perm("/job-tasks/put"), rt.Bgrw(), rou.taskHostAction)
+			pages.GET("/task/:id/host/:host/output", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskHostOutput)
+			pages.GET("/task/:id/host/:host/stdout", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskHostStdout)
+			pages.GET("/task/:id/host/:host/stderr", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskHostStderr)
+			pages.GET("/task/:id/stdout.txt", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskStdoutTxt)
+			pages.GET("/task/:id/stderr.txt", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskStderrTxt)
+			pages.GET("/task/:id/stdout.json", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskStdoutJSON)
+			pages.GET("/task/:id/stderr.json", rt.Auth(), rt.User(), rt.Perm("/job-tasks"), rou.taskStderrJSON)
+		}
+	}
+
+	api := r.Group("/ibex/v1")
+	if len(config.C.BasicAuth) > 0 {
+		api = r.Group("/ibex/v1", gin.BasicAuth(config.C.BasicAuth))
+	}
+	{
+		api.POST("/tasks", rou.taskAdd)
+		api.GET("/tasks", rou.taskGets)
+		api.GET("/tasks/done-ids", rou.doneIds)
+		api.GET("/task/:id", rou.taskGet)
+		api.PUT("/task/:id/action", rou.taskAction)
+		api.GET("/task/:id/stdout", rou.taskStdout)
+		api.GET("/task/:id/stderr", rou.taskStderr)
+		api.GET("/task/:id/state", rou.taskState)
+		api.GET("/task/:id/result", rou.taskResult)
+		api.PUT("/task/:id/host/:host/action", rou.taskHostAction)
+		api.GET("/task/:id/host/:host/output", rou.taskHostOutput)
+		api.GET("/task/:id/host/:host/stdout", rou.taskHostStdout)
+		api.GET("/task/:id/host/:host/stderr", rou.taskHostStderr)
+		api.GET("/task/:id/stdout.txt", rou.taskStdoutTxt)
+		api.GET("/task/:id/stderr.txt", rou.taskStderrTxt)
+		api.GET("/task/:id/stdout.json", rou.taskStdoutJSON)
+		api.GET("/task/:id/stderr.json", rou.taskStderrJSON)
+
+		// api for edge server
+		api.POST("/table/record/list", rou.tableRecordListGet)
+		api.POST("/table/record/count", rou.tableRecordCount)
+		api.POST("/mark/done", rou.markDone)
+		api.POST("/task/meta", rou.taskMetaAdd)
+		api.POST("/task/host/", rou.taskHostAdd)
+		api.POST("/task/hosts/upsert", rou.taskHostUpsert)
+	}
+}

ibex/server/rpc/method.go

@@ -0,0 +1,93 @@
+package rpc
+
+import (
+	"fmt"
+	"github.com/ccfos/nightingale/v6/models"
+	"os"
+
+	"github.com/toolkits/pkg/logger"
+
+	"github.com/ccfos/nightingale/v6/ibex/types"
+)
+
+// Ping return string 'pong', just for test
+func (*Server) Ping(input string, output *string) error {
+	*output = "pong"
+	return nil
+}
+
+func (*Server) GetTaskMeta(id int64, resp *types.TaskMetaResponse) error {
+	meta, err := models.TaskMetaGetByID(ctxC, id)
+	if err != nil {
+		resp.Message = err.Error()
+		return nil
+	}
+
+	if meta == nil {
+		resp.Message = fmt.Sprintf("task %d not found", id)
+		return nil
+	}
+
+	resp.Script = meta.Script
+	resp.Args = meta.Args
+	resp.Account = meta.Account
+	resp.Stdin = meta.Stdin
+
+	return nil
+}
+
+func (*Server) Report(req types.ReportRequest, resp *types.ReportResponse) error {
+	if req.ReportTasks != nil && len(req.ReportTasks) > 0 {
+		err := handleDoneTask(req)
+		if err != nil {
+			resp.Message = err.Error()
+			return nil
+		}
+	}
+
+	doings := models.GetDoingCache(req.Ident)
+
+	tasks := make([]types.AssignTask, 0, len(doings))
+	for _, doing := range doings {
+		tasks = append(tasks, types.AssignTask{
+			Id:     doing.Id,
+			Clock:  doing.Clock,
+			Action: doing.Action,
+		})
+	}
+	resp.AssignTasks = tasks
+
+	return nil
+}
+
+func handleDoneTask(req types.ReportRequest) error {
+	count := len(req.ReportTasks)
+	val, ok := os.LookupEnv("CONTINUOUS_OUTPUT")
+	for i := 0; i < count; i++ {
+		t := req.ReportTasks[i]
+		if ok && val == "1" && t.Status == "running" {
+			err := models.RealTimeUpdateOutput(ctxC, t.Id, req.Ident, t.Stdout, t.Stderr)
+			if err != nil {
+				logger.Errorf("cannot update output, id:%d, hostname:%s, clock:%d, status:%s, err: %v", t.Id, req.Ident, t.Clock, t.Status, err)
+				return err
+			}
+		} else {
+			if t.Status == "success" || t.Status == "failed" {
+				exist, isEdgeAlertTriggered := models.CheckExistAndEdgeAlertTriggered(req.Ident, t.Id)
+				// ibex agent可能会重复上报结果，如果任务已经不在task_host_doing缓存中了，说明该任务已经MarkDone了，不需要再处理
+				if !exist {
+					continue
+				}
+
+				err := models.MarkDoneStatus(ctxC, t.Id, t.Clock, req.Ident, t.Status, t.Stdout, t.Stderr, isEdgeAlertTriggered)
+				if err != nil {
+					logger.Errorf("cannot mark task done, id:%d, hostname:%s, clock:%d, status:%s, err: %v", t.Id, req.Ident, t.Clock, t.Status, err)
+					return err
+				}
+			}
+		}
+
+	}
+
+	return nil
+}

ibex/server/rpc/rpc.go

@@ -0,0 +1,61 @@
+package rpc
+
+import (
+	"bufio"
+	"fmt"
+	"io"
+	"net"
+	"net/rpc"
+	"os"
+	"reflect"
+	"time"
+
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+
+	"github.com/toolkits/pkg/logger"
+	"github.com/ugorji/go/codec"
+)
+
+type Server int
+
+var ctxC *ctx.Context
+
+func Start(listen string, ctx *ctx.Context) {
+	ctxC = ctx
+	go serve(listen)
+}
+
+func serve(listen string) {
+	server := rpc.NewServer()
+	server.Register(new(Server))
+
+	l, err := net.Listen("tcp", listen)
+	if err != nil {
+		fmt.Printf("fail to listen on: %s, error: %v\n", listen, err)
+		os.Exit(1)
+	}
+
+	fmt.Println("rpc.listening:", listen)
+
+	var mh codec.MsgpackHandle
+	mh.MapType = reflect.TypeOf(map[string]interface{}(nil))
+
+	duration := time.Duration(100) * time.Millisecond
+
+	for {
+		conn, err := l.Accept()
+		if err != nil {
+			logger.Warningf("listener accept error: %v", err)
+			time.Sleep(duration)
+			continue
+		}
+
+		var bufconn = struct {
+			io.Closer
+			*bufio.Reader
+			*bufio.Writer
+		}{conn, bufio.NewReader(conn), bufio.NewWriter(conn)}
+
+		go server.ServeCodec(codec.MsgpackSpecRpc.ServerCodec(bufconn, &mh))
+	}
+}

ibex/server/server.go

@@ -0,0 +1,159 @@
+package server
+
+import (
+	"context"
+	"fmt"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"os"
+	"os/signal"
+	"path/filepath"
+	"syscall"
+
+	"github.com/ccfos/nightingale/v6/ibex/server/config"
+	"github.com/ccfos/nightingale/v6/ibex/server/router"
+	"github.com/ccfos/nightingale/v6/ibex/server/rpc"
+	"github.com/ccfos/nightingale/v6/ibex/server/timer"
+	"github.com/ccfos/nightingale/v6/pkg/httpx"
+	"github.com/ccfos/nightingale/v6/pkg/logx"
+	"github.com/ccfos/nightingale/v6/storage"
+
+	"github.com/toolkits/pkg/i18n"
+)
+
+type Server struct {
+	ConfigFile string
+	Version    string
+}
+
+type ServerOption func(*Server)
+
+func SetConfigFile(f string) ServerOption {
+	return func(s *Server) {
+		s.ConfigFile = f
+	}
+}
+
+func SetVersion(v string) ServerOption {
+	return func(s *Server) {
+		s.Version = v
+	}
+}
+
+// Run run server
+func Run(isCenter bool, opts ...ServerOption) {
+	code := 1
+	sc := make(chan os.Signal, 1)
+	signal.Notify(sc, syscall.SIGHUP, syscall.SIGINT, syscall.SIGTERM, syscall.SIGQUIT)
+
+	server := Server{
+		ConfigFile: filepath.Join("etc", "ibex", "server.toml"),
+		Version:    "not specified",
+	}
+
+	for _, opt := range opts {
+		opt(&server)
+	}
+
+	// parse config file
+	config.MustLoad(server.ConfigFile)
+	config.C.IsCenter = isCenter
+
+	cleanFunc, err := server.initialize()
+	if err != nil {
+		fmt.Println("server init fail:", err)
+		os.Exit(code)
+	}
+
+EXIT:
+	for {
+		sig := <-sc
+		fmt.Println("received signal:", sig.String())
+		switch sig {
+		case syscall.SIGQUIT, syscall.SIGTERM, syscall.SIGINT:
+			code = 0
+			break EXIT
+		case syscall.SIGHUP:
+			// reload configuration?
+		default:
+			break EXIT
+		}
+	}
+
+	cleanFunc()
+	fmt.Println("server exited")
+	os.Exit(code)
+}
+
+func (s Server) initialize() (func(), error) {
+	fns := Functions{}
+	bgCtx, cancel := context.WithCancel(context.Background())
+	fns.Add(cancel)
+
+	// init i18n
+	i18n.Init()
+
+	// init logger
+	loggerClean, err := logx.Init(config.C.Log)
+	if err != nil {
+		return fns.Ret(), err
+	} else {
+		fns.Add(loggerClean)
+	}
+
+	var ctxC *ctx.Context
+
+	var redis storage.Redis
+	if redis, err = storage.NewRedis(config.C.Redis); err != nil {
+		return fns.Ret(), err
+	}
+
+	// init database
+	if config.C.IsCenter {
+		db, err := storage.New(config.C.DB)
+		if err != nil {
+			return fns.Ret(), err
+		}
+		ctxC = ctx.NewContext(context.Background(), db, redis, true, config.C.CenterApi)
+	} else {
+		ctxC = ctx.NewContext(context.Background(), nil, redis, false, config.C.CenterApi)
+	}
+
+	if err := storage.IdInit(ctxC.Redis); err != nil {
+		fmt.Println("cannot init id generator: ", err)
+		os.Exit(1)
+	}
+
+	timer.CacheHostDoing(ctxC)
+	timer.ReportResult(ctxC)
+	if config.C.IsCenter {
+		go timer.Heartbeat(ctxC)
+		go timer.Schedule(ctxC)
+		go timer.CleanLong(ctxC)
+	}
+	// init http server
+	r := router.New(ctxC, s.Version)
+	httpClean := httpx.Init(config.C.HTTP, bgCtx, r)
+	fns.Add(httpClean)
+
+	// start rpc server
+	rpc.Start(config.C.RPC.Listen, ctxC)
+
+	// release all the resources
+	return fns.Ret(), nil
+}
+
+type Functions struct {
+	List []func()
+}
+
+func (fs *Functions) Add(f func()) {
+	fs.List = append(fs.List, f)
+}
+
+func (fs *Functions) Ret() func() {
+	return func() {
+		for i := 0; i < len(fs.List); i++ {
+			fs.List[i]()
+		}
+	}
+}

ibex/server/timer/timer_heartbeat.go

@@ -0,0 +1,76 @@
+package timer
+
+import (
+	"github.com/ccfos/nightingale/v6/models"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"time"
+
+	"github.com/ccfos/nightingale/v6/ibex/server/config"
+
+	"github.com/toolkits/pkg/logger"
+)
+
+func Heartbeat(ctx *ctx.Context) {
+	if config.C.Heartbeat.Interval == 0 {
+		config.C.Heartbeat.Interval = 1000
+	}
+
+	for {
+		heartbeat(ctx)
+		time.Sleep(time.Duration(config.C.Heartbeat.Interval) * time.Millisecond)
+	}
+}
+
+func heartbeat(ctx *ctx.Context) {
+	ident := config.C.Heartbeat.LocalAddr
+
+	err := models.TaskSchedulerHeartbeat(ctx, ident)
+	if err != nil {
+		logger.Errorf("task scheduler(%s) cannot heartbeat: %v", ident, err)
+		return
+	}
+
+	dss, err := models.DeadTaskSchedulers(ctx)
+	if err != nil {
+		logger.Errorf("cannot get dead task schedulers: %v", err)
+		return
+	}
+
+	cnt := len(dss)
+	if cnt == 0 {
+		return
+	}
+
+	for i := 0; i < cnt; i++ {
+		ids, err := models.TasksOfScheduler(ctx, dss[i])
+		if err != nil {
+			logger.Errorf("cannot get tasks of scheduler(%s): %v", dss[i], err)
+			return
+		}
+
+		if len(ids) == 0 {
+			err = models.DelDeadTaskScheduler(ctx, dss[i])
+			if err != nil {
+				logger.Errorf("cannot del dead task scheduler(%s): %v", dss[i], err)
+				return
+			}
+		}
+
+		takeOverTasks(ctx, ident, dss[i], ids)
+	}
+}
+
+func takeOverTasks(ctx *ctx.Context, alive, dead string, ids []int64) {
+	count := len(ids)
+	for i := 0; i < count; i++ {
+		success, err := models.TakeOverTask(ctx, ids[i], dead, alive)
+		if err != nil {
+			logger.Errorf("cannot take over task: %v", err)
+			return
+		}
+
+		if success {
+			logger.Infof("%s take over task[%d] of %s", alive, ids[i], dead)
+		}
+	}
+}

ibex/server/timer/timer_host_doing.go

@@ -0,0 +1,53 @@
+package timer
+
+import (
+	"fmt"
+	"github.com/ccfos/nightingale/v6/models"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"time"
+
+	"github.com/toolkits/pkg/logger"
+)
+
+// CacheHostDoing 缓存task_host_doing表全部内容，减轻DB压力
+func CacheHostDoing(ctx *ctx.Context) {
+	if err := cacheHostDoing(ctx); err != nil {
+		fmt.Println("cannot cache task_host_doing data: ", err)
+	}
+	go loopCacheHostDoing(ctx)
+}
+
+func loopCacheHostDoing(ctx *ctx.Context) {
+	for {
+		time.Sleep(time.Millisecond * 400)
+		if err := cacheHostDoing(ctx); err != nil {
+			logger.Warning("cannot cache task_host_doing data: ", err)
+		}
+	}
+}
+
+func cacheHostDoing(ctx *ctx.Context) error {
+	doingsFromDb, err := models.TableRecordGets[[]models.TaskHostDoing](ctx, models.TaskHostDoing{}.TableName(), "")
+	if err != nil {
+		logger.Errorf("models.TableRecordGets fail: %v", err)
+	}
+
+	doingsFromRedis, err := models.CacheRecordGets[models.TaskHostDoing](ctx)
+	if err != nil {
+		logger.Errorf("models.CacheRecordGets fail: %v", err)
+	}
+
+	set := make(map[string][]models.TaskHostDoing)
+	for _, doing := range doingsFromDb {
+		doing.AlertTriggered = false
+		set[doing.Host] = append(set[doing.Host], doing)
+	}
+	for _, doing := range doingsFromRedis {
+		doing.AlertTriggered = true
+		set[doing.Host] = append(set[doing.Host], doing)
+	}
+
+	models.SetDoingCache(set)
+
+	return err
+}

ibex/server/timer/timer_report_result.go

@@ -0,0 +1,27 @@
+package timer
+
+import (
+	"fmt"
+	"github.com/ccfos/nightingale/v6/models"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"time"
+
+	"github.com/toolkits/pkg/logger"
+)
+
+func ReportResult(ctx *ctx.Context) {
+	if err := models.ReportCacheResult(ctx); err != nil {
+		fmt.Println("cannot report task_host result from alter trigger: ", err)
+	}
+	go loopReport(ctx)
+}
+
+func loopReport(ctx *ctx.Context) {
+	d := time.Duration(2) * time.Second
+	for {
+		time.Sleep(d)
+		if err := models.ReportCacheResult(ctx); err != nil {
+			logger.Warning("cannot report task_host result from alter trigger: ", err)
+		}
+	}
+}

ibex/server/timer/timer_scheduler.go

@@ -0,0 +1,79 @@
+package timer
+
+import (
+	"github.com/ccfos/nightingale/v6/models"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"time"
+
+	"github.com/ccfos/nightingale/v6/ibex/server/config"
+	"github.com/ccfos/nightingale/v6/ibex/server/logic"
+
+	"github.com/toolkits/pkg/logger"
+)
+
+func Schedule(ctx *ctx.Context) {
+	for {
+		scheduleOrphan(ctx)
+		scheduleMine(ctx)
+		time.Sleep(time.Second)
+	}
+}
+
+func scheduleMine(ctx *ctx.Context) {
+	ids, err := models.TasksOfScheduler(ctx, config.C.Heartbeat.LocalAddr)
+	if err != nil {
+		logger.Errorf("cannot get tasks of scheduler(%s): %v", config.C.Heartbeat.LocalAddr, err)
+		return
+	}
+
+	count := len(ids)
+	for i := 0; i < count; i++ {
+		logic.CheckTimeout(ctx, ids[i])
+		logic.ScheduleTask(ctx, ids[i])
+	}
+}
+
+func scheduleOrphan(ctx *ctx.Context) {
+	ids, err := models.OrphanTaskIds(ctx)
+	if err != nil {
+		logger.Errorf("cannot get orphan task ids: %v", err)
+		return
+	}
+
+	count := len(ids)
+	if count == 0 {
+		return
+	}
+
+	logger.Debug("orphan task ids:", ids)
+
+	for i := 0; i < count; i++ {
+		action, err := models.TaskActionGet(ctx, "id=?", ids[i])
+		if err != nil {
+			logger.Errorf("cannot get task[%d] action: %v", ids[i], err)
+			continue
+		}
+
+		if action == nil {
+			continue
+		}
+
+		if action.Action == "pause" {
+			continue
+		}
+
+		mine, err := models.TakeOverTask(ctx, ids[i], "", config.C.Heartbeat.LocalAddr)
+		if err != nil {
+			logger.Errorf("cannot take over task[%d]: %v", ids[i], err)
+			continue
+		}
+
+		if !mine {
+			continue
+		}
+
+		logger.Debugf("task[%d] is mine", ids[i])
+
+		logic.ScheduleTask(ctx, ids[i])
+	}
+}

ibex/server/timer/timer_task_cleaner.go

@@ -0,0 +1,38 @@
+package timer
+
+import (
+	"github.com/ccfos/nightingale/v6/models"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"time"
+
+	"github.com/toolkits/pkg/logger"
+)
+
+func CleanLong(ctx *ctx.Context) {
+	d := time.Duration(24) * time.Hour
+	for {
+		cleanLongTask(ctx)
+		time.Sleep(d)
+	}
+}
+
+func cleanLongTask(ctx *ctx.Context) {
+	ids, err := models.LongTaskIds(ctx)
+	if err != nil {
+		logger.Error("LongTaskIds:", err)
+		return
+	}
+
+	if ids == nil {
+		return
+	}
+
+	count := len(ids)
+	for i := 0; i < count; i++ {
+		action := models.TaskAction{Id: ids[i]}
+		err = action.Update(ctx, "cancel")
+		if err != nil {
+			logger.Errorf("cannot cancel long task[%d]: %v", ids[i], err)
+		}
+	}
+}

ibex/types/types.go

@@ -0,0 +1,33 @@
+package types
+
+type TaskMetaResponse struct {
+	Message string
+	Script  string
+	Args    string
+	Account string
+	Stdin   string
+}
+
+type ReportTask struct {
+	Id     int64
+	Clock  int64
+	Status string
+	Stdout string
+	Stderr string
+}
+
+type ReportRequest struct {
+	Ident       string
+	ReportTasks []ReportTask
+}
+
+type AssignTask struct {
+	Id     int64
+	Clock  int64
+	Action string
+}
+
+type ReportResponse struct {
+	Message     string
+	AssignTasks []AssignTask
+}

integrations/IPMI/dashboards/IPMI_by_prometheus.json

@@ -1,463 +0,0 @@
-{
-    "name": "IPMI for Prometheus",
-    "ident": "",
-    "configs": {
-        "version": "2.0.0",
-        "links": [],
-        "var": [
-            {
-                "name": "node",
-                "type": "query",
-                "datasource": {
-                    "cate": "prometheus"
-                },
-                "definition": "label_values(ipmi_bmc_info, ident)",
-                "reg": "",
-                "multi": false
-            }
-        ],
-        "panels": [
-            {
-                "type": "gauge",
-                "id": "f975fded-f57e-4a6e-80b4-50d5be6dd84c",
-                "layout": {
-                    "h": 7,
-                    "w": 24,
-                    "x": 0,
-                    "y": 0,
-                    "i": "f975fded-f57e-4a6e-80b4-50d5be6dd84c",
-                    "isResizable": true
-                },
-                "version": "2.0.0",
-                "datasourceCate": "prometheus",
-                "targets": [
-                    {
-                        "refId": "A",
-                        "expr": "ipmi_temperature_celsius{ident='$node'}",
-                        "legend": "{{name}}"
-                    }
-                ],
-                "transformations": [
-                    {
-                        "id": "organize",
-                        "options": {}
-                    }
-                ],
-                "name": "Temperatures",
-                "links": [],
-                "custom": {
-                    "textMode": "valueAndName",
-                    "calc": "avg"
-                },
-                "options": {
-                    "valueMappings": [],
-                    "standardOptions": {
-                        "util": "none"
-                    },
-                    "thresholds": {
-                        "steps": [
-                            {
-                                "color": "green",
-                                "value": null,
-                                "type": "base"
-                            },
-                            {
-                                "color": "red",
-                                "value": 80
-                            }
-                        ]
-                    }
-                }
-            },
-            {
-                "type": "timeseries",
-                "id": "681f1191-4777-4377-8b77-404d9f036406",
-                "layout": {
-                    "h": 5,
-                    "w": 12,
-                    "x": 0,
-                    "y": 7,
-                    "i": "681f1191-4777-4377-8b77-404d9f036406",
-                    "isResizable": true
-                },
-                "version": "2.0.0",
-                "datasourceCate": "prometheus",
-                "targets": [
-                    {
-                        "refId": "A",
-                        "expr": "ipmi_power_watts{ident='$node'}",
-                        "legend": "{{name}}"
-                    }
-                ],
-                "transformations": [
-                    {
-                        "id": "organize",
-                        "options": {}
-                    }
-                ],
-                "name": "Power",
-                "links": [],
-                "options": {
-                    "tooltip": {
-                        "mode": "all",
-                        "sort": "none"
-                    },
-                    "legend": {
-                        "displayMode": "hidden"
-                    },
-                    "standardOptions": {},
-                    "thresholds": {
-                        "steps": [
-                            {
-                                "color": "#634CD9",
-                                "value": null,
-                                "type": "base"
-                            }
-                        ]
-                    }
-                },
-                "custom": {
-                    "drawStyle": "lines",
-                    "lineInterpolation": "smooth",
-                    "spanNulls": false,
-                    "lineWidth": 1,
-                    "fillOpacity": 0.5,
-                    "gradientMode": "none",
-                    "stack": "off",
-                    "scaleDistribution": {
-                        "type": "linear"
-                    }
-                }
-            },
-            {
-                "type": "timeseries",
-                "id": "feede24c-8296-4127-982e-08cfc4151933",
-                "layout": {
-                    "h": 5,
-                    "w": 12,
-                    "x": 12,
-                    "y": 7,
-                    "i": "feede24c-8296-4127-982e-08cfc4151933",
-                    "isResizable": true
-                },
-                "version": "2.0.0",
-                "datasourceCate": "prometheus",
-                "targets": [
-                    {
-                        "refId": "A",
-                        "expr": "ipmi_power_watts{ident='$node'} * 30 * 24 ",
-                        "legend": "{{name}}"
-                    }
-                ],
-                "transformations": [
-                    {
-                        "id": "organize",
-                        "options": {}
-                    }
-                ],
-                "name": "Power usage 30d",
-                "links": [],
-                "options": {
-                    "tooltip": {
-                        "mode": "all",
-                        "sort": "none"
-                    },
-                    "legend": {
-                        "displayMode": "hidden"
-                    },
-                    "standardOptions": {},
-                    "thresholds": {
-                        "steps": [
-                            {
-                                "color": "#634CD9",
-                                "value": null,
-                                "type": "base"
-                            }
-                        ]
-                    }
-                },
-                "custom": {
-                    "drawStyle": "lines",
-                    "lineInterpolation": "smooth",
-                    "spanNulls": false,
-                    "lineWidth": 1,
-                    "fillOpacity": 0.5,
-                    "gradientMode": "none",
-                    "stack": "off",
-                    "scaleDistribution": {
-                        "type": "linear"
-                    }
-                }
-            },
-            {
-                "type": "timeseries",
-                "id": "9e11e7f5-ed3c-49eb-8a72-ee76c8700c24",
-                "layout": {
-                    "h": 7,
-                    "w": 12,
-                    "x": 0,
-                    "y": 12,
-                    "i": "9e11e7f5-ed3c-49eb-8a72-ee76c8700c24",
-                    "isResizable": true
-                },
-                "version": "2.0.0",
-                "datasourceCate": "prometheus",
-                "targets": [
-                    {
-                        "refId": "A",
-                        "expr": "ipmi_temperature_celsius{ident='$node'}",
-                        "legend": "{{name}}"
-                    }
-                ],
-                "transformations": [
-                    {
-                        "id": "organize",
-                        "options": {}
-                    }
-                ],
-                "name": "Temperatures",
-                "links": [],
-                "description": "",
-                "options": {
-                    "tooltip": {
-                        "mode": "multi"
-                    },
-                    "legend": {
-                        "displayMode": "list",
-                        "placement": "bottom"
-                    },
-                    "standardOptions": {
-                        "util": "none"
-                    },
-                    "thresholds": {
-                        "steps": [
-                            {
-                                "color": "green",
-                                "value": null,
-                                "type": "base"
-                            },
-                            {
-                                "color": "red",
-                                "value": 80
-                            }
-                        ]
-                    }
-                },
-                "custom": {
-                    "drawStyle": "lines",
-                    "lineInterpolation": "linear",
-                    "spanNulls": false,
-                    "lineWidth": 1,
-                    "fillOpacity": 0.5,
-                    "gradientMode": "none",
-                    "stack": "off",
-                    "scaleDistribution": {
-                        "type": "linear"
-                    }
-                }
-            },
-            {
-                "type": "timeseries",
-                "id": "95c734f7-26cb-41a7-8376-49332cc220c2",
-                "layout": {
-                    "h": 7,
-                    "w": 12,
-                    "x": 12,
-                    "y": 12,
-                    "i": "95c734f7-26cb-41a7-8376-49332cc220c2",
-                    "isResizable": true
-                },
-                "version": "2.0.0",
-                "datasourceCate": "prometheus",
-                "targets": [
-                    {
-                        "refId": "A",
-                        "expr": "ipmi_power_watts{ident='$node'}",
-                        "legend": "{{name}}"
-                    }
-                ],
-                "transformations": [
-                    {
-                        "id": "organize",
-                        "options": {}
-                    }
-                ],
-                "name": "Power",
-                "links": [],
-                "description": "",
-                "options": {
-                    "tooltip": {
-                        "mode": "multi"
-                    },
-                    "legend": {
-                        "displayMode": "list",
-                        "placement": "bottom"
-                    },
-                    "standardOptions": {
-                        "util": "none"
-                    },
-                    "thresholds": {
-                        "steps": [
-                            {
-                                "color": "green",
-                                "value": null,
-                                "type": "base"
-                            },
-                            {
-                                "color": "red",
-                                "value": 80
-                            }
-                        ]
-                    }
-                },
-                "custom": {
-                    "drawStyle": "lines",
-                    "lineInterpolation": "linear",
-                    "spanNulls": false,
-                    "lineWidth": 1,
-                    "fillOpacity": 0.01,
-                    "gradientMode": "none",
-                    "stack": "off",
-                    "scaleDistribution": {
-                        "type": "linear"
-                    }
-                }
-            },
-            {
-                "type": "timeseries",
-                "id": "0313f34f-afcf-41e9-8f69-9a3dbd4b2e56",
-                "layout": {
-                    "h": 7,
-                    "w": 12,
-                    "x": 0,
-                    "y": 19,
-                    "i": "0313f34f-afcf-41e9-8f69-9a3dbd4b2e56",
-                    "isResizable": true
-                },
-                "version": "2.0.0",
-                "datasourceCate": "prometheus",
-                "targets": [
-                    {
-                        "refId": "A",
-                        "expr": "ipmi_fan_speed_rpm{ident='$node'}",
-                        "legend": "{{name}}"
-                    }
-                ],
-                "transformations": [
-                    {
-                        "id": "organize",
-                        "options": {}
-                    }
-                ],
-                "name": "Fans",
-                "links": [],
-                "description": "",
-                "options": {
-                    "tooltip": {
-                        "mode": "multi"
-                    },
-                    "legend": {
-                        "displayMode": "list",
-                        "placement": "bottom"
-                    },
-                    "standardOptions": {
-                        "util": "none"
-                    },
-                    "thresholds": {
-                        "steps": [
-                            {
-                                "color": "green",
-                                "value": null,
-                                "type": "base"
-                            },
-                            {
-                                "color": "red",
-                                "value": 80
-                            }
-                        ]
-                    }
-                },
-                "custom": {
-                    "drawStyle": "lines",
-                    "lineInterpolation": "linear",
-                    "spanNulls": false,
-                    "lineWidth": 1,
-                    "fillOpacity": 0.5,
-                    "gradientMode": "none",
-                    "stack": "off",
-                    "scaleDistribution": {
-                        "type": "linear"
-                    }
-                }
-            },
-            {
-                "type": "timeseries",
-                "id": "29ee004d-a95c-405d-97d1-d715fab4e1de",
-                "layout": {
-                    "h": 7,
-                    "w": 12,
-                    "x": 12,
-                    "y": 19,
-                    "i": "29ee004d-a95c-405d-97d1-d715fab4e1de",
-                    "isResizable": true
-                },
-                "version": "2.0.0",
-                "datasourceCate": "prometheus",
-                "targets": [
-                    {
-                        "refId": "A",
-                        "expr": "ipmi_voltage_volts{ident='$node',name!~\"Voltage 1|Voltage 2\"}",
-                        "legend": "{{name}}"
-                    }
-                ],
-                "transformations": [
-                    {
-                        "id": "organize",
-                        "options": {}
-                    }
-                ],
-                "name": "Voltages",
-                "links": [],
-                "description": "",
-                "options": {
-                    "tooltip": {
-                        "mode": "multi"
-                    },
-                    "legend": {
-                        "displayMode": "list",
-                        "placement": "bottom"
-                    },
-                    "standardOptions": {
-                        "util": "none"
-                    },
-                    "thresholds": {
-                        "steps": [
-                            {
-                                "color": "green",
-                                "value": null,
-                                "type": "base"
-                            },
-                            {
-                                "color": "red",
-                                "value": 80
-                            }
-                        ]
-                    }
-                },
-                "custom": {
-                    "drawStyle": "lines",
-                    "lineInterpolation": "linear",
-                    "spanNulls": false,
-                    "lineWidth": 1,
-                    "fillOpacity": 0.5,
-                    "gradientMode": "none",
-                    "stack": "off",
-                    "scaleDistribution": {
-                        "type": "linear"
-                    }
-                }
-            }
-        ]
-    },
-    "uuid": 1727587308068775200
-}

integrations/Linux/alerts/常用中文告警规则-采集器Categraf.json

@@ -1425,7 +1425,7 @@
         "rule_config": {
             "queries": [
                 {
-                    "prom_ql": "increase(kernel_vmstat_oom_kill[2m]) > 0",
+                    "prom_ql": "kernel_vmstat_oom_kill != 0",
                     "severity": 2
                 }
             ]
@@ -2139,4 +2139,4 @@
         "update_by": "",
         "uuid": 1717556327737117000
     }
-]
+]

integrations/Linux/metrics/categraf-base.json

@@ -259,11 +259,11 @@
         "uuid": 1717556327796195000,
         "collector": "Categraf",
         "typ": "Linux",
-        "name": "1分钟内 OOM 次数统计",
+        "name": "OOM 次数统计",
         "unit": "none",
         "note": "取自 `/proc/vmstat`，需要较高版本的内核，没记错的话应该是 4.13 以上版本",
         "lang": "zh_CN",
-        "expression": "increase(kernel_vmstat_oom_kill[1m])",
+        "expression": "kernel_vmstat_oom_kill",
         "created_at": 0,
         "created_by": "",
         "updated_at": 0,
@@ -1334,4 +1334,4 @@
         "updated_at": 0,
         "updated_by": ""
     }
-]
+]

memsto/target_cache.go

@@ -160,9 +160,8 @@ func (tc *TargetCacheType) syncTargets() error {
 	}
 
 	m := make(map[string]*models.Target)
-
-	metaMap := tc.GetHostMetas(lst)
-	if len(metaMap) > 0 {
+	if tc.ctx.IsCenter {
+		metaMap := tc.GetHostMetas(lst)
 		for i := 0; i < len(lst); i++ {
 			if meta, ok := metaMap[lst[i].Ident]; ok {
 				lst[i].FillMeta(meta)

models/alert_cur_event.go

@@ -67,7 +67,6 @@ type AlertCurEvent struct {
 	Claimant           string            `json:"claimant" gorm:"-"`
 	SubRuleId          int64             `json:"sub_rule_id" gorm:"-"`
 	ExtraInfo          []string          `json:"extra_info" gorm:"-"`
-	Target             *Target           `json:"target" gorm:"-"`
 }
 
 func (e *AlertCurEvent) TableName() string {
@@ -342,7 +341,7 @@ func (e *AlertCurEvent) DB2Mem() {
 			continue
 		}
 
-		arr := strings.SplitN(pair, "=", 2)
+		arr := strings.Split(pair, "=")
 		if len(arr) != 2 {
 			continue
 		}
@@ -419,8 +418,7 @@ func (e *AlertCurEvent) FillNotifyGroups(ctx *ctx.Context, cache map[int64]*User
 	return nil
 }
 
-func AlertCurEventTotal(ctx *ctx.Context, prods []string, bgids []int64, stime, etime int64,
-	severity int, dsIds []int64, cates []string, ruleId int64, query string) (int64, error) {
+func AlertCurEventTotal(ctx *ctx.Context, prods []string, bgids []int64, stime, etime int64, severity int, dsIds []int64, cates []string, query string) (int64, error) {
 	session := DB(ctx).Model(&AlertCurEvent{})
 	if stime != 0 && etime != 0 {
 		session = session.Where("trigger_time between ? and ?", stime, etime)
@@ -445,10 +443,6 @@ func AlertCurEventTotal(ctx *ctx.Context, prods []string, bgids []int64, stime,
 		session = session.Where("cate in ?", cates)
 	}
 
-	if ruleId > 0 {
-		session = session.Where("rule_id = ?", ruleId)
-	}
-
 	if query != "" {
 		arr := strings.Fields(query)
 		for i := 0; i < len(arr); i++ {
@@ -460,9 +454,7 @@ func AlertCurEventTotal(ctx *ctx.Context, prods []string, bgids []int64, stime,
 	return Count(session)
 }
 
-func AlertCurEventsGet(ctx *ctx.Context, prods []string, bgids []int64, stime, etime int64,
-	severity int, dsIds []int64, cates []string, ruleId int64, query string, limit, offset int) (
-	[]AlertCurEvent, error) {
+func AlertCurEventGets(ctx *ctx.Context, prods []string, bgids []int64, stime, etime int64, severity int, dsIds []int64, cates []string, query string, limit, offset int) ([]AlertCurEvent, error) {
 	session := DB(ctx).Model(&AlertCurEvent{})
 	if stime != 0 && etime != 0 {
 		session = session.Where("trigger_time between ? and ?", stime, etime)
@@ -487,10 +479,6 @@ func AlertCurEventsGet(ctx *ctx.Context, prods []string, bgids []int64, stime, e
 		session = session.Where("cate in ?", cates)
 	}
 
-	if ruleId > 0 {
-		session = session.Where("rule_id = ?", ruleId)
-	}
-
 	if query != "" {
 		arr := strings.Fields(query)
 		for i := 0; i < len(arr); i++ {
@@ -511,26 +499,6 @@ func AlertCurEventsGet(ctx *ctx.Context, prods []string, bgids []int64, stime, e
 	return lst, err
 }
 
-func AlertCurEventCountByRuleId(ctx *ctx.Context, rids []int64, stime, etime int64) map[int64]int64 {
-	type Row struct {
-		RuleId int64
-		Cnt    int64
-	}
-	var rows []Row
-	err := DB(ctx).Model(&AlertCurEvent{}).Select("rule_id, count(*) as cnt").
-		Where("trigger_time between ? and ?", stime, etime).Group("rule_id").Find(&rows).Error
-	if err != nil {
-		logger.Errorf("Failed to count group by rule_id: %v", err)
-		return nil
-	}
-
-	curEventTotalByRid := make(map[int64]int64, len(rids))
-	for _, r := range rows {
-		curEventTotalByRid[r.RuleId] = r.Cnt
-	}
-	return curEventTotalByRid
-}
-
 func AlertCurEventDel(ctx *ctx.Context, ids []int64) error {
 	if len(ids) == 0 {
 		return nil

models/alert_his_event.go

@@ -121,9 +121,7 @@ func (e *AlertHisEvent) FillNotifyGroups(ctx *ctx.Context, cache map[int64]*User
 
 // }
 
-func AlertHisEventTotal(
-	ctx *ctx.Context, prods []string, bgids []int64, stime, etime int64, severity int,
-	recovered int, dsIds []int64, cates []string, ruleId int64, query string) (int64, error) {
+func AlertHisEventTotal(ctx *ctx.Context, prods []string, bgids []int64, stime, etime int64, severity int, recovered int, dsIds []int64, cates []string, query string) (int64, error) {
 	session := DB(ctx).Model(&AlertHisEvent{}).Where("last_eval_time between ? and ?", stime, etime)
 
 	if len(prods) > 0 {
@@ -150,10 +148,6 @@ func AlertHisEventTotal(
 		session = session.Where("cate in ?", cates)
 	}
 
-	if ruleId > 0 {
-		session = session.Where("rule_id = ?", ruleId)
-	}
-
 	if query != "" {
 		arr := strings.Fields(query)
 		for i := 0; i < len(arr); i++ {
@@ -165,9 +159,7 @@ func AlertHisEventTotal(
 	return Count(session)
 }
 
-func AlertHisEventGets(ctx *ctx.Context, prods []string, bgids []int64, stime, etime int64,
-	severity int, recovered int, dsIds []int64, cates []string, ruleId int64, query string,
-	limit, offset int) ([]AlertHisEvent, error) {
+func AlertHisEventGets(ctx *ctx.Context, prods []string, bgids []int64, stime, etime int64, severity int, recovered int, dsIds []int64, cates []string, query string, limit, offset int) ([]AlertHisEvent, error) {
 	session := DB(ctx).Where("last_eval_time between ? and ?", stime, etime)
 
 	if len(prods) != 0 {
@@ -194,10 +186,6 @@ func AlertHisEventGets(ctx *ctx.Context, prods []string, bgids []int64, stime, e
 		session = session.Where("cate in ?", cates)
 	}
 
-	if ruleId > 0 {
-		session = session.Where("rule_id = ?", ruleId)
-	}
-
 	if query != "" {
 		arr := strings.Fields(query)
 		for i := 0; i < len(arr); i++ {

models/alert_rule.go

@@ -98,8 +98,6 @@ type AlertRule struct {
 	UpdateAt              int64                  `json:"update_at"`
 	UpdateBy              string                 `json:"update_by"`
 	UUID                  int64                  `json:"uuid" gorm:"-"` // tpl identifier
-	CurEventCount         int64                  `json:"cur_event_count" gorm:"-"`
-	UpdateByNickname      string                 `json:"update_by_nickname" gorm:"-"` // for fe
 }
 
 type Tpl struct {
@@ -163,12 +161,10 @@ type Trigger struct {
 	Duration int    `json:"duration,omitempty"`
 	Percent  int    `json:"percent,omitempty"`
 	Joins    []Join `json:"joins"`
-	JoinRef  string `json:"join_ref"`
 }
 
 type Join struct {
 	JoinType string   `json:"join_type"`
-	Ref      string   `json:"ref"`
 	On       []string `json:"on"`
 }
 
@@ -180,10 +176,9 @@ func GetHostsQuery(queries []HostQuery) []map[string]interface{} {
 		case "group_ids":
 			ids := ParseInt64(q.Values)
 			if q.Op == "==" {
-				m["target_busi_group.group_id in (?)"] = ids
+				m["group_id in (?)"] = ids
 			} else {
-				m["target.ident not in (select target_ident "+
-					"from target_busi_group where group_id in (?))"] = ids
+				m["group_id not in (?)"] = ids
 			}
 		case "tags":
 			lst := []string{}
@@ -823,8 +818,7 @@ func AlertRuleGetsAll(ctx *ctx.Context) ([]*AlertRule, error) {
 	return lst, nil
 }
 
-func AlertRulesGetsBy(ctx *ctx.Context, prods []string, query, algorithm, cluster string,
-	cates []string, disabled int) ([]*AlertRule, error) {
+func AlertRulesGetsBy(ctx *ctx.Context, prods []string, query, algorithm, cluster string, cates []string, disabled int) ([]*AlertRule, error) {
 	session := DB(ctx)
 
 	if len(prods) > 0 {

models/board.go

@@ -286,53 +286,3 @@ func BoardSetHide(ctx *ctx.Context, ids []int64) error {
 		return nil
 	})
 }
-
-func BoardGetsByBids(ctx *ctx.Context, bids []int64) ([]map[string]interface{}, error) {
-	var boards []Board
-	err := DB(ctx).Where("id IN ?", bids).Find(&boards).Error
-	if err != nil {
-		return nil, err
-	}
-
-	// 收集所有唯一的 group_id
-	groupIDs := make([]int64, 0)
-	groupIDSet := make(map[int64]struct{})
-	for _, board := range boards {
-		if _, exists := groupIDSet[board.GroupId]; !exists {
-			groupIDs = append(groupIDs, board.GroupId)
-			groupIDSet[board.GroupId] = struct{}{}
-		}
-	}
-
-	// 一次性查询所有需要的 BusiGroup
-	var busiGroups []BusiGroup
-	err = DB(ctx).Where("id IN ?", groupIDs).Find(&busiGroups).Error
-	if err != nil {
-		return nil, err
-	}
-
-	// 创建 group_id 到 BusiGroup 的映射
-	groupMap := make(map[int64]BusiGroup)
-	for _, bg := range busiGroups {
-		groupMap[bg.Id] = bg
-	}
-
-	result := make([]map[string]interface{}, 0, len(boards))
-	for _, board := range boards {
-		busiGroup, exists := groupMap[board.GroupId]
-		if !exists {
-			// 处理找不到对应 BusiGroup 的情况
-			continue
-		}
-
-		item := map[string]interface{}{
-			"busi_group_name": busiGroup.Name,
-			"busi_group_id":   busiGroup.Id,
-			"board_id":        board.Id,
-			"board_name":      board.Name,
-		}
-		result = append(result, item)
-	}
-
-	return result, nil
-}

models/builtin_component.go

@@ -11,7 +11,7 @@ import (
 // BuiltinComponent represents a builtin component along with its metadata.
 type BuiltinComponent struct {
 	ID        uint64 `json:"id" gorm:"primaryKey;type:bigint;autoIncrement;comment:'unique identifier'"`
-	Ident     string `json:"ident" gorm:"type:varchar(191);not null;uniqueIndex:idx_ident,sort:asc;comment:'identifier of component'"`
+	Ident     string `json:"ident" gorm:"type:varchar(191);not null;index:idx_ident,sort:asc;comment:'identifier of component'"`
 	Logo      string `json:"logo" gorm:"type:varchar(191);not null;comment:'logo of component'"`
 	Readme    string `json:"readme" gorm:"type:text;not null;comment:'readme of component'"`
 	CreatedAt int64  `json:"created_at" gorm:"type:bigint;not null;default:0;comment:'create time'"`

models/builtin_metrics.go

@@ -2,7 +2,6 @@ package models
 
 import (
 	"errors"
-	"fmt"
 	"strings"
 	"time"
 
@@ -218,10 +217,3 @@ func BuiltinMetricCollectors(ctx *ctx.Context, lang, typ, query string) ([]strin
 	err := session.Select("distinct(collector)").Pluck("collector", &collectors).Error
 	return collectors, err
 }
-
-func BuiltinMetricBatchUpdateColumn(ctx *ctx.Context, col, old, new, updatedBy string) error {
-	if old == new {
-		return nil
-	}
-	return DB(ctx).Model(&BuiltinMetric{}).Where(fmt.Sprintf("%s = ?", col), old).Updates(map[string]interface{}{col: new, "updated_by": updatedBy}).Error
-}

models/builtin_payload.go

@@ -9,19 +9,18 @@ import (
 )
 
 type BuiltinPayload struct {
-	ID          int64  `json:"id" gorm:"primaryKey;type:bigint;autoIncrement;comment:'unique identifier'"`
-	Type        string `json:"type" gorm:"type:varchar(191);not null;index:idx_type,sort:asc;comment:'type of payload'"`                // Alert Dashboard Collet
-	Component   string `json:"component" gorm:"type:varchar(191);not null;index:idx_component,sort:asc;comment:'component of payload'"` //
-	ComponentID uint64 `json:"component_id" gorm:"type:bigint;index:idx_component,sort:asc;comment:'component_id of payload'"`          // ComponentID which the payload belongs to
-	Cate        string `json:"cate" gorm:"type:varchar(191);not null;comment:'category of payload'"`                                    // categraf_v1 telegraf_v1
-	Name        string `json:"name" gorm:"type:varchar(191);not null;index:idx_buildinpayload_name,sort:asc;comment:'name of payload'"` //
-	Tags        string `json:"tags" gorm:"type:varchar(191);not null;default:'';comment:'tags of payload'"`                             // {"host":"
-	Content     string `json:"content" gorm:"type:longtext;not null;comment:'content of payload'"`
-	UUID        int64  `json:"uuid" gorm:"type:bigint;not null;index:idx_uuid;comment:'uuid of payload'"`
-	CreatedAt   int64  `json:"created_at" gorm:"type:bigint;not null;default:0;comment:'create time'"`
-	CreatedBy   string `json:"created_by" gorm:"type:varchar(191);not null;default:'';comment:'creator'"`
-	UpdatedAt   int64  `json:"updated_at" gorm:"type:bigint;not null;default:0;comment:'update time'"`
-	UpdatedBy   string `json:"updated_by" gorm:"type:varchar(191);not null;default:'';comment:'updater'"`
+	ID        int64  `json:"id" gorm:"primaryKey;type:bigint;autoIncrement;comment:'unique identifier'"`
+	Type      string `json:"type" gorm:"type:varchar(191);not null;index:idx_type,sort:asc;comment:'type of payload'"`                // Alert Dashboard Collet
+	Component string `json:"component" gorm:"type:varchar(191);not null;index:idx_component,sort:asc;comment:'component of payload'"` // Host MySQL Redis
+	Cate      string `json:"cate" gorm:"type:varchar(191);not null;comment:'category of payload'"`                                    // categraf_v1 telegraf_v1
+	Name      string `json:"name" gorm:"type:varchar(191);not null;index:idx_buildinpayload_name,sort:asc;comment:'name of payload'"`                //
+	Tags      string `json:"tags" gorm:"type:varchar(191);not null;default:'';comment:'tags of payload'"`                             // {"host":"
+	Content   string `json:"content" gorm:"type:longtext;not null;comment:'content of payload'"`
+	UUID      int64  `json:"uuid" gorm:"type:bigint;not null;index:idx_uuid;comment:'uuid of payload'"`
+	CreatedAt int64  `json:"created_at" gorm:"type:bigint;not null;default:0;comment:'create time'"`
+	CreatedBy string `json:"created_by" gorm:"type:varchar(191);not null;default:'';comment:'creator'"`
+	UpdatedAt int64  `json:"updated_at" gorm:"type:bigint;not null;default:0;comment:'update time'"`
+	UpdatedBy string `json:"updated_by" gorm:"type:varchar(191);not null;default:'';comment:'updater'"`
 }
 
 func (bp *BuiltinPayload) TableName() string {
@@ -34,8 +33,9 @@ func (bp *BuiltinPayload) Verify() error {
 		return errors.New("type is blank")
 	}
 
-	if bp.ComponentID == 0 {
-		return errors.New("component_id is blank")
+	bp.Component = strings.TrimSpace(bp.Component)
+	if bp.Component == "" {
+		return errors.New("component is blank")
 	}
 
 	if bp.Name == "" {
@@ -47,7 +47,7 @@ func (bp *BuiltinPayload) Verify() error {
 
 func BuiltinPayloadExists(ctx *ctx.Context, bp *BuiltinPayload) (bool, error) {
 	var count int64
-	err := DB(ctx).Model(bp).Where("type = ? AND component_id = ? AND name = ? AND cate = ?", bp.Type, bp.ComponentID, bp.Name, bp.Cate).Count(&count).Error
+	err := DB(ctx).Model(bp).Where("type = ? AND component = ? AND name = ? AND cate = ?", bp.Type, bp.Component, bp.Name, bp.Cate).Count(&count).Error
 	if err != nil {
 		return false, err
 	}
@@ -78,7 +78,7 @@ func (bp *BuiltinPayload) Update(ctx *ctx.Context, req BuiltinPayload) error {
 		return err
 	}
 
-	if bp.Type != req.Type || bp.ComponentID != req.ComponentID || bp.Name != req.Name {
+	if bp.Type != req.Type || bp.Component != req.Component || bp.Name != req.Name {
 		exists, err := BuiltinPayloadExists(ctx, &req)
 		if err != nil {
 			return err
@@ -117,13 +117,13 @@ func BuiltinPayloadGet(ctx *ctx.Context, where string, args ...interface{}) (*Bu
 	return &bp, nil
 }
 
-func BuiltinPayloadGets(ctx *ctx.Context, componentId uint64, typ, cate, query string) ([]*BuiltinPayload, error) {
+func BuiltinPayloadGets(ctx *ctx.Context, typ, component, cate, query string) ([]*BuiltinPayload, error) {
 	session := DB(ctx)
 	if typ != "" {
 		session = session.Where("type = ?", typ)
 	}
-	if componentId != 0 {
-		session = session.Where("component_id = ?", componentId)
+	if component != "" {
+		session = session.Where("component = ?", component)
 	}
 
 	if cate != "" {
@@ -144,9 +144,9 @@ func BuiltinPayloadGets(ctx *ctx.Context, componentId uint64, typ, cate, query s
 }
 
 // get cates of BuiltinPayload by type and component, return []string
-func BuiltinPayloadCates(ctx *ctx.Context, typ string, componentID uint64) ([]string, error) {
+func BuiltinPayloadCates(ctx *ctx.Context, typ, component string) ([]string, error) {
 	var cates []string
-	err := DB(ctx).Model(new(BuiltinPayload)).Where("type = ? and component_id = ?", typ, componentID).Distinct("cate").Pluck("cate", &cates).Error
+	err := DB(ctx).Model(new(BuiltinPayload)).Where("type = ? and component = ?", typ, component).Distinct("cate").Pluck("cate", &cates).Error
 	return cates, err
 }
 
@@ -163,37 +163,3 @@ func BuiltinPayloadComponents(ctx *ctx.Context, typ, cate string) (string, error
 	}
 	return components[0], nil
 }
-
-// InitBuiltinPayloads 兼容新旧 BuiltinPayload 格式
-func InitBuiltinPayloads(ctx *ctx.Context) error {
-	var lst []*BuiltinPayload
-
-	components, err := BuiltinComponentGets(ctx, "")
-	if err != nil {
-		return err
-	}
-
-	identToId := make(map[string]uint64)
-	for _, component := range components {
-		identToId[component.Ident] = component.ID
-	}
-
-	err = DB(ctx).Where("component_id = 0 or component_id is NULL").Find(&lst).Error
-	if err != nil {
-		return err
-	}
-
-	for _, bp := range lst {
-		componentId, ok := identToId[bp.Component]
-		if !ok {
-			continue
-		}
-		bp.ComponentID = componentId
-	}
-
-	if len(lst) == 0 {
-		return nil
-	}
-
-	return DB(ctx).Save(&lst).Error
-}

models/busi_group.go

@@ -134,7 +134,7 @@ func (bg *BusiGroup) Del(ctx *ctx.Context) error {
 		return errors.New("Some alert subscribes still in the BusiGroup")
 	}
 
-	has, err = Exists(DB(ctx).Model(&TargetBusiGroup{}).Where("group_id=?", bg.Id))
+	has, err = Exists(DB(ctx).Model(&Target{}).Where("group_id=?", bg.Id))
 	if err != nil {
 		return err
 	}

models/datasource.go

@@ -35,7 +35,6 @@ type Datasource struct {
 	UpdatedBy      string                 `json:"updated_by"`
 	IsDefault      bool                   `json:"is_default"`
 	Transport      *http.Transport        `json:"-" gorm:"-"`
-	ForceSave      bool                   `json:"force_save" gorm:"-"`
 }
 
 type Auth struct {

models/ibex_models.go

@@ -0,0 +1,69 @@
+package models
+
+import (
+	"encoding/json"
+	"fmt"
+
+	"github.com/ccfos/nightingale/v6/ibex/server/config"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"github.com/ccfos/nightingale/v6/pkg/poster"
+
+	"gorm.io/gorm"
+)
+
+func IbexCount(tx *gorm.DB) (int64, error) {
+	var cnt int64
+	err := tx.Count(&cnt).Error
+	return cnt, err
+}
+
+func tht(id int64) string {
+	return fmt.Sprintf("task_host_%d", id%100)
+}
+
+func TableRecordGets[T any](ctx *ctx.Context, table, where string, args ...interface{}) (lst T, err error) {
+	if config.C.IsCenter {
+		if where == "" || len(args) == 0 {
+			err = DB(ctx).Table(table).Find(&lst).Error
+		} else {
+			err = DB(ctx).Table(table).Where(where, args...).Find(&lst).Error
+		}
+		return
+	}
+
+	return poster.PostByUrlsWithResp[T](ctx, "/ibex/v1/table/record/list", map[string]interface{}{
+		"table": table,
+		"where": where,
+		"args":  args,
+	})
+}
+
+func TableRecordCount(ctx *ctx.Context, table, where string, args ...interface{}) (int64, error) {
+	if config.C.IsCenter {
+		if where == "" || len(args) == 0 {
+			return IbexCount(DB(ctx).Table(table))
+		}
+		return IbexCount(DB(ctx).Table(table).Where(where, args...))
+	}
+
+	return poster.PostByUrlsWithResp[int64](ctx, "/ibex/v1/table/record/count", map[string]interface{}{
+		"table": table,
+		"where": where,
+		"args":  args,
+	})
+}
+
+var IBEX_HOST_DOING = "ibex-host-doing"
+
+func CacheRecordGets[T any](ctx *ctx.Context) ([]T, error) {
+	lst := make([]T, 0)
+	values, _ := ctx.Redis.HVals(ctx.Ctx, IBEX_HOST_DOING).Result()
+	for _, val := range values {
+		t := new(T)
+		if err := json.Unmarshal([]byte(val), t); err != nil {
+			return nil, err
+		}
+		lst = append(lst, *t)
+	}
+	return lst, nil
+}

models/ibex_task_action.go

@@ -0,0 +1,112 @@
+package models
+
+import (
+	"fmt"
+	"time"
+
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+
+	"gorm.io/gorm"
+)
+
+type TaskAction struct {
+	Id     int64  `gorm:"column:id;primaryKey"`
+	Action string `gorm:"column:action;size:32;not null"`
+	Clock  int64  `gorm:"column:clock;not null;default:0"`
+}
+
+func (TaskAction) TableName() string {
+	return "task_action"
+}
+
+func TaskActionGet(ctx *ctx.Context, where string, args ...interface{}) (*TaskAction, error) {
+	var obj TaskAction
+	ret := DB(ctx).Where(where, args...).Find(&obj)
+	if ret.Error != nil {
+		return nil, ret.Error
+	}
+
+	if ret.RowsAffected == 0 {
+		return nil, nil
+	}
+
+	return &obj, nil
+}
+
+func TaskActionExistsIds(ctx *ctx.Context, ids []int64) ([]int64, error) {
+	if len(ids) == 0 {
+		return ids, nil
+	}
+
+	var ret []int64
+	err := DB(ctx).Model(&TaskAction{}).Where("id in ?", ids).Pluck("id", &ret).Error
+	return ret, err
+}
+
+func CancelWaitingHosts(ctx *ctx.Context, id int64) error {
+	return DB(ctx).Table(tht(id)).Where("id = ? and status = ?", id, "waiting").Update("status", "cancelled").Error
+}
+
+func StartTask(ctx *ctx.Context, id int64) error {
+	return DB(ctx).Model(&TaskScheduler{}).Where("id = ?", id).Update("scheduler", "").Error
+}
+
+func CancelTask(ctx *ctx.Context, id int64) error {
+	return CancelWaitingHosts(ctx, id)
+}
+
+func KillTask(ctx *ctx.Context, id int64) error {
+	if err := CancelWaitingHosts(ctx, id); err != nil {
+		return err
+	}
+
+	now := time.Now().Unix()
+
+	return DB(ctx).Transaction(func(tx *gorm.DB) error {
+		err := tx.Model(&TaskHostDoing{}).Where("id = ? and action <> ?", id, "kill").Updates(map[string]interface{}{
+			"clock":  now,
+			"action": "kill",
+		}).Error
+		if err != nil {
+			return err
+		}
+
+		return tx.Table(tht(id)).Where("id = ? and status = ?", id, "running").Update("status", "killing").Error
+	})
+}
+
+func (a *TaskAction) Update(ctx *ctx.Context, action string) error {
+	if !(action == "start" || action == "cancel" || action == "kill" || action == "pause") {
+		return fmt.Errorf("action invalid")
+	}
+
+	err := DB(ctx).Model(a).Updates(map[string]interface{}{
+		"action": action,
+		"clock":  time.Now().Unix(),
+	}).Error
+	if err != nil {
+		return err
+	}
+
+	if action == "start" {
+		return StartTask(ctx, a.Id)
+	}
+
+	if action == "cancel" {
+		return CancelTask(ctx, a.Id)
+	}
+
+	if action == "kill" {
+		return KillTask(ctx, a.Id)
+	}
+
+	return nil
+}
+
+// LongTaskIds two weeks ago
+func LongTaskIds(ctx *ctx.Context) ([]int64, error) {
+	clock := time.Now().Unix() - 604800*2
+	var ids []int64
+	err := DB(ctx).Model(&TaskAction{}).Where("clock < ?", clock).Pluck("id", &ids).Error
+	return ids, err
+}

models/ibex_task_host.go

@@ -0,0 +1,262 @@
+package models
+
+import (
+	"fmt"
+	"sync"
+	"time"
+
+	"github.com/ccfos/nightingale/v6/ibex/server/config"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"github.com/ccfos/nightingale/v6/pkg/poster"
+	"github.com/ccfos/nightingale/v6/storage"
+
+	"github.com/toolkits/pkg/logger"
+	"gorm.io/gorm"
+	"gorm.io/gorm/clause"
+)
+
+type TaskHost struct {
+	II     int64  `gorm:"column:ii;primaryKey;autoIncrement" json:"-"`
+	Id     int64  `gorm:"column:id;uniqueIndex:idx_id_host;not null" json:"id"`
+	Host   string `gorm:"column:host;uniqueIndex:idx_id_host;size:128;not null" json:"host"`
+	Status string `gorm:"column:status;size:32;not null" json:"status"`
+	Stdout string `gorm:"column:stdout;type:text" json:"stdout"`
+	Stderr string `gorm:"column:stderr;type:text" json:"stderr"`
+}
+
+func (taskHost *TaskHost) Upsert(ctx *ctx.Context) error {
+	return DB(ctx).Table(tht(taskHost.Id)).Clauses(clause.OnConflict{
+		Columns:   []clause.Column{{Name: "id"}, {Name: "host"}},
+		DoUpdates: clause.AssignmentColumns([]string{"status", "stdout", "stderr"}),
+	}).Create(taskHost).Error
+}
+
+func (taskHost *TaskHost) Create(ctx *ctx.Context) error {
+	if config.C.IsCenter {
+		return DB(ctx).Table(tht(taskHost.Id)).Create(taskHost).Error
+	}
+	return poster.PostByUrls(ctx, "/ibex/v1/task/host", taskHost)
+}
+
+func TaskHostUpserts(ctx *ctx.Context, lst []TaskHost) (map[string]error, error) {
+	if len(lst) == 0 {
+		return nil, fmt.Errorf("empty list")
+	}
+
+	if !config.C.IsCenter {
+		return poster.PostByUrlsWithResp[map[string]error](ctx, "/ibex/v1/task/hosts/upsert", lst)
+	}
+
+	errs := make(map[string]error, 0)
+	for _, taskHost := range lst {
+		if err := taskHost.Upsert(ctx); err != nil {
+			errs[fmt.Sprintf("%d:%s", taskHost.Id, taskHost.Host)] = err
+		}
+	}
+	return errs, nil
+}
+
+func TaskHostGet(ctx *ctx.Context, id int64, host string) (*TaskHost, error) {
+	var ret []*TaskHost
+	err := DB(ctx).Table(tht(id)).Where("id=? and host=?", id, host).Find(&ret).Error
+	if err != nil {
+		return nil, err
+	}
+
+	if len(ret) == 0 {
+		return nil, nil
+	}
+
+	return ret[0], nil
+}
+
+func MarkDoneStatus(ctx *ctx.Context, id, clock int64, host, status, stdout, stderr string, edgeAlertTriggered ...bool) error {
+	if len(edgeAlertTriggered) > 0 && edgeAlertTriggered[0] {
+		return CacheMarkDone(ctx, TaskHost{
+			Id:     id,
+			Host:   host,
+			Status: status,
+			Stdout: stdout,
+			Stderr: stderr,
+		})
+	}
+
+	if !config.C.IsCenter {
+		return poster.PostByUrls(ctx, "/ibex/v1/mark/done", map[string]interface{}{
+			"id":     id,
+			"clock":  clock,
+			"host":   host,
+			"status": status,
+			"stdout": stdout,
+			"stderr": stderr,
+		})
+	}
+
+	count, err := TableRecordCount(ctx, TaskHostDoing{}.TableName(), "id=? and host=? and clock=?", id, host, clock)
+	if err != nil {
+		return err
+	}
+
+	if count == 0 {
+		// 如果是timeout了，后来任务执行完成之后，结果又上来了，stdout和stderr最好还是存库，让用户看到
+		count, err = TableRecordCount(ctx, tht(id), "id=? and host=? and status=?", id, host, "timeout")
+		if err != nil {
+			return err
+		}
+
+		if count == 1 {
+			return DB(ctx).Table(tht(id)).Where("id=? and host=?", id, host).Updates(map[string]interface{}{
+				"status": status,
+				"stdout": stdout,
+				"stderr": stderr,
+			}).Error
+		}
+		return nil
+	}
+
+	return DB(ctx).Transaction(func(tx *gorm.DB) error {
+		err = tx.Table(tht(id)).Where("id=? and host=?", id, host).Updates(map[string]interface{}{
+			"status": status,
+			"stdout": stdout,
+			"stderr": stderr,
+		}).Error
+		if err != nil {
+			return err
+		}
+
+		if err = tx.Where("id=? and host=?", id, host).Delete(&TaskHostDoing{}).Error; err != nil {
+			return err
+		}
+
+		return nil
+	})
+}
+
+func RealTimeUpdateOutput(ctx *ctx.Context, id int64, host, stdout, stderr string) error {
+	return DB(ctx).Transaction(func(tx *gorm.DB) error {
+		err := tx.Table(tht(id)).Where("id=? and host=?", id, host).Updates(map[string]interface{}{
+			"stdout": stdout,
+			"stderr": stderr,
+		}).Error
+		if err != nil {
+			return err
+		}
+
+		return nil
+	})
+}
+
+func CacheMarkDone(ctx *ctx.Context, taskHost TaskHost) error {
+	if err := ctx.Redis.HDel(ctx.Ctx, IBEX_HOST_DOING, hostDoingCacheKey(taskHost.Id, taskHost.Host)).Err(); err != nil {
+		return err
+	}
+	TaskHostCachePush(taskHost)
+
+	return nil
+}
+
+func WaitingHostList(ctx *ctx.Context, id int64, limit ...int) ([]TaskHost, error) {
+	var hosts []TaskHost
+	session := DB(ctx).Table(tht(id)).Where("id = ? and status = 'waiting'", id).Order("ii")
+	if len(limit) > 0 {
+		session = session.Limit(limit[0])
+	}
+	err := session.Find(&hosts).Error
+	return hosts, err
+}
+
+func WaitingHostCount(ctx *ctx.Context, id int64) (int64, error) {
+	return TableRecordCount(ctx, tht(id), "id=? and status='waiting'", id)
+}
+
+func UnexpectedHostCount(ctx *ctx.Context, id int64) (int64, error) {
+	return TableRecordCount(ctx, tht(id), "id=? and status in ('failed', 'timeout', 'killfailed')", id)
+}
+
+func IngStatusHostCount(ctx *ctx.Context, id int64) (int64, error) {
+	return TableRecordCount(ctx, tht(id), "id=? and status in ('waiting', 'running', 'killing')", id)
+}
+
+func RunWaitingHosts(ctx *ctx.Context, taskHosts []TaskHost) error {
+	count := len(taskHosts)
+	if count == 0 {
+		return nil
+	}
+
+	now := time.Now().Unix()
+
+	return DB(ctx).Transaction(func(tx *gorm.DB) error {
+		for i := 0; i < count; i++ {
+			if err := tx.Table(tht(taskHosts[i].Id)).Where("id=? and host=?", taskHosts[i].Id, taskHosts[i].Host).Update("status", "running").Error; err != nil {
+				return err
+			}
+			err := tx.Create(&TaskHostDoing{Id: taskHosts[i].Id, Host: taskHosts[i].Host, Clock: now, Action: "start"}).Error
+			if err != nil {
+				return err
+			}
+		}
+
+		return nil
+	})
+}
+
+func TaskHostStatus(ctx *ctx.Context, id int64) ([]TaskHost, error) {
+	var ret []TaskHost
+	err := DB(ctx).Table(tht(id)).Select("id", "host", "status").Where("id=?", id).Order("ii").Find(&ret).Error
+	return ret, err
+}
+
+func TaskHostGets(ctx *ctx.Context, id int64) ([]TaskHost, error) {
+	var ret []TaskHost
+	err := DB(ctx).Table(tht(id)).Where("id=?", id).Order("ii").Find(&ret).Error
+	return ret, err
+}
+
+var (
+	taskHostCache = make([]TaskHost, 0, 128)
+	taskHostLock  sync.RWMutex
+)
+
+func TaskHostCachePush(taskHost TaskHost) {
+	taskHostLock.Lock()
+	defer taskHostLock.Unlock()
+
+	taskHostCache = append(taskHostCache, taskHost)
+}
+
+func TaskHostCachePopAll() []TaskHost {
+	taskHostLock.Lock()
+	defer taskHostLock.Unlock()
+
+	all := taskHostCache
+	taskHostCache = make([]TaskHost, 0, 128)
+
+	return all
+}
+
+func ReportCacheResult(ctx *ctx.Context) error {
+	result := TaskHostCachePopAll()
+	reports := make([]TaskHost, 0)
+	for _, th := range result {
+		// id大于redis初始id，说明是edge与center失联时，本地告警规则触发的自愈脚本生成的id
+		// 为了防止不同边缘机房生成的脚本任务id相同，不上报结果至数据库
+		if th.Id >= storage.IDINITIAL {
+			logger.Infof("task[%d] host[%s] done, result:[%v]", th.Id, th.Host, th)
+		} else {
+			reports = append(reports, th)
+		}
+	}
+
+	if len(reports) == 0 {
+		return nil
+	}
+
+	errs, err := TaskHostUpserts(ctx, reports)
+	if err != nil {
+		return err
+	}
+	for key, err := range errs {
+		logger.Warningf("report task_host_cache[%s] result error: %v", key, err)
+	}
+	return nil
+}

models/ibex_task_host_doing.go

@@ -0,0 +1,65 @@
+package models
+
+import (
+	"encoding/json"
+	"fmt"
+	"sync"
+)
+
+type TaskHostDoing struct {
+	Id             int64  `gorm:"column:id;index"`
+	Host           string `gorm:"column:host;size:128;not null;index"`
+	Clock          int64  `gorm:"column:clock;not null;default:0"`
+	Action         string `gorm:"column:action;size:16;not null"`
+	AlertTriggered bool   `gorm:"-"`
+}
+
+func (TaskHostDoing) TableName() string {
+	return "task_host_doing"
+}
+
+func (doing *TaskHostDoing) MarshalBinary() ([]byte, error) {
+	return json.Marshal(doing)
+}
+
+func (doing *TaskHostDoing) UnmarshalBinary(data []byte) error {
+	return json.Unmarshal(data, doing)
+}
+
+func hostDoingCacheKey(id int64, host string) string {
+	return fmt.Sprintf("%s:%d", host, id)
+}
+
+var (
+	doingLock sync.RWMutex
+	doingMaps map[string][]TaskHostDoing
+)
+
+func SetDoingCache(v map[string][]TaskHostDoing) {
+	doingLock.Lock()
+	doingMaps = v
+	doingLock.Unlock()
+}
+
+func GetDoingCache(host string) []TaskHostDoing {
+	doingLock.RLock()
+	defer doingLock.RUnlock()
+
+	return doingMaps[host]
+}
+
+func CheckExistAndEdgeAlertTriggered(host string, id int64) (exist, isAlertTriggered bool) {
+	doingLock.RLock()
+	defer doingLock.RUnlock()
+
+	doings := doingMaps[host]
+	for _, doing := range doings {
+		if doing.Id == id {
+			exist = true
+			isAlertTriggered = doing.AlertTriggered
+			return
+		}
+	}
+
+	return false, false
+}

models/ibex_task_meta.go

@@ -0,0 +1,364 @@
+package models
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"strings"
+	"time"
+
+	"github.com/ccfos/nightingale/v6/ibex/server/config"
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+	"github.com/ccfos/nightingale/v6/pkg/poster"
+	"github.com/ccfos/nightingale/v6/storage"
+
+	"github.com/toolkits/pkg/str"
+	"gorm.io/gorm"
+)
+
+type TaskMeta struct {
+	Id        int64     `gorm:"column:id;primaryKey;autoIncrement" json:"id"`
+	Title     string    `gorm:"column:title;size:255;not null;default:''" json:"title"`
+	Account   string    `gorm:"column:account;size:64;not null" json:"account"`
+	Batch     int       `gorm:"column:batch;not null;default:0" json:"batch"`
+	Tolerance int       `gorm:"column:tolerance;not null;default:0" json:"tolerance"`
+	Timeout   int       `gorm:"column:timeout;not null;default:0" json:"timeout"`
+	Pause     string    `gorm:"column:pause;size:255;not null;default:''" json:"pause"`
+	Script    string    `gorm:"column:script;type:text;not null" json:"script"`
+	Args      string    `gorm:"column:args;size:512;not null;default:''" json:"args"`
+	Stdin     string    `gorm:"column:stdin;size:1024;not null;default:''" json:"stdin"`
+	Creator   string    `gorm:"column:creator;size:64;not null;default:'';index" json:"creator"`
+	Created   time.Time `gorm:"column:created;not null;default:CURRENT_TIMESTAMP;type:timestamp;index" json:"created"`
+	Done      bool      `json:"done" gorm:"-"`
+}
+
+func (TaskMeta) TableName() string {
+	return "task_meta"
+}
+
+func (taskMeta *TaskMeta) MarshalBinary() ([]byte, error) {
+	return json.Marshal(taskMeta)
+}
+
+func (taskMeta *TaskMeta) UnmarshalBinary(data []byte) error {
+	return json.Unmarshal(data, taskMeta)
+}
+
+func (taskMeta *TaskMeta) Create(ctx *ctx.Context) error {
+	if config.C.IsCenter {
+		return DB(ctx).Create(taskMeta).Error
+	}
+
+	id, err := poster.PostByUrlsWithResp[int64](ctx, "/ibex/v1/task/meta", taskMeta)
+	if err == nil {
+		taskMeta.Id = id
+	}
+
+	return err
+}
+
+func taskMetaCacheKey(id int64) string {
+	return fmt.Sprintf("task:meta:%d", id)
+}
+
+func TaskMetaGet(ctx *ctx.Context, where string, args ...interface{}) (*TaskMeta, error) {
+	lst, err := TableRecordGets[[]*TaskMeta](ctx, TaskMeta{}.TableName(), where, args...)
+	if err != nil {
+		return nil, err
+	}
+
+	if len(lst) == 0 {
+		return nil, nil
+	}
+
+	return lst[0], nil
+}
+
+// TaskMetaGet 根据ID获取任务元信息，会用到缓存
+func TaskMetaGetByID(ctx *ctx.Context, id int64) (*TaskMeta, error) {
+	meta, err := TaskMetaCacheGet(ctx, id)
+	if err == nil {
+		return meta, nil
+	}
+
+	meta, err = TaskMetaGet(ctx, "id=?", id)
+	if err != nil {
+		return nil, err
+	}
+
+	if meta == nil {
+		return nil, nil
+	}
+
+	_, err = ctx.Redis.Set(context.Background(), taskMetaCacheKey(id), meta, storage.DEFAULT).Result()
+
+	return meta, err
+}
+
+func TaskMetaCacheGet(ctx *ctx.Context, id int64) (*TaskMeta, error) {
+	res := ctx.Redis.Get(context.Background(), taskMetaCacheKey(id))
+	meta := new(TaskMeta)
+	err := res.Scan(meta)
+	return meta, err
+}
+
+func (m *TaskMeta) CleanFields() error {
+	if m.Batch < 0 {
+		return fmt.Errorf("arg(batch) should be nonnegative")
+	}
+
+	if m.Tolerance < 0 {
+		return fmt.Errorf("arg(tolerance) should be nonnegative")
+	}
+
+	if m.Timeout < 0 {
+		return fmt.Errorf("arg(timeout) should be nonnegative")
+	}
+
+	if m.Timeout > 3600*24*5 {
+		return fmt.Errorf("arg(timeout) longer than five days")
+	}
+
+	if m.Timeout == 0 {
+		m.Timeout = 30
+	}
+
+	m.Pause = strings.Replace(m.Pause, "，", ",", -1)
+	m.Pause = strings.Replace(m.Pause, " ", "", -1)
+	m.Args = strings.Replace(m.Args, "，", ",", -1)
+
+	if m.Title == "" {
+		return fmt.Errorf("arg(title) is required")
+	}
+
+	if str.Dangerous(m.Title) {
+		return fmt.Errorf("arg(title) is dangerous")
+	}
+
+	if m.Script == "" {
+		return fmt.Errorf("arg(script) is required")
+	}
+
+	if str.Dangerous(m.Args) {
+		return fmt.Errorf("arg(args) is dangerous")
+	}
+
+	if str.Dangerous(m.Pause) {
+		return fmt.Errorf("arg(pause) is dangerous")
+	}
+
+	return nil
+}
+
+func (m *TaskMeta) HandleFH(fh string) {
+	i := strings.Index(m.Title, " FH: ")
+	if i > 0 {
+		m.Title = m.Title[:i]
+	}
+	m.Title = m.Title + " FH: " + fh
+}
+
+func (taskMeta *TaskMeta) Cache(ctx *ctx.Context, host string) error {
+
+	tx := ctx.Redis.TxPipeline()
+	tx.Set(ctx.Ctx, taskMetaCacheKey(taskMeta.Id), taskMeta, storage.DEFAULT)
+	tx.HSet(ctx.Ctx, IBEX_HOST_DOING, hostDoingCacheKey(taskMeta.Id, host), &TaskHostDoing{
+		Id:     taskMeta.Id,
+		Host:   host,
+		Clock:  time.Now().Unix(),
+		Action: "start",
+	})
+
+	_, err := tx.Exec(ctx.Ctx)
+
+	return err
+}
+
+func (taskMeta *TaskMeta) Save(ctx *ctx.Context, hosts []string, action string) error {
+	return DB(ctx).Transaction(func(tx *gorm.DB) error {
+		if err := tx.Create(taskMeta).Error; err != nil {
+			return err
+		}
+
+		id := taskMeta.Id
+
+		if err := tx.Create(&TaskScheduler{Id: id}).Error; err != nil {
+			return err
+		}
+
+		if err := tx.Create(&TaskAction{Id: id, Action: action, Clock: time.Now().Unix()}).Error; err != nil {
+			return err
+		}
+
+		for i := 0; i < len(hosts); i++ {
+			host := strings.TrimSpace(hosts[i])
+			if host == "" {
+				continue
+			}
+
+			err := tx.Exec("INSERT INTO "+tht(id)+" (id, host, status) VALUES (?, ?, ?)", id, host, "waiting").Error
+			if err != nil {
+				return err
+			}
+		}
+
+		return nil
+	})
+}
+
+func (m *TaskMeta) Action(ctx *ctx.Context) (*TaskAction, error) {
+	return TaskActionGet(ctx, "id=?", m.Id)
+}
+
+func (m *TaskMeta) Hosts(ctx *ctx.Context) ([]TaskHost, error) {
+	var ret []TaskHost
+	err := DB(ctx).Table(tht(m.Id)).Where("id=?", m.Id).Select("id", "host", "status").Order("ii").Find(&ret).Error
+	return ret, err
+}
+
+func (m *TaskMeta) KillHost(ctx *ctx.Context, host string) error {
+	bean, err := TaskHostGet(ctx, m.Id, host)
+	if err != nil {
+		return err
+	}
+
+	if bean == nil {
+		return fmt.Errorf("no such host")
+	}
+
+	if !(bean.Status == "running" || bean.Status == "timeout") {
+		return fmt.Errorf("current status cannot kill")
+	}
+
+	if err := redoHost(ctx, m.Id, host, "kill"); err != nil {
+		return err
+	}
+
+	return statusSet(ctx, m.Id, host, "killing")
+}
+
+func (m *TaskMeta) IgnoreHost(ctx *ctx.Context, host string) error {
+	return statusSet(ctx, m.Id, host, "ignored")
+}
+
+func (m *TaskMeta) RedoHost(ctx *ctx.Context, host string) error {
+	bean, err := TaskHostGet(ctx, m.Id, host)
+	if err != nil {
+		return err
+	}
+
+	if bean == nil {
+		return fmt.Errorf("no such host")
+	}
+
+	if err := redoHost(ctx, m.Id, host, "start"); err != nil {
+		return err
+	}
+
+	return statusSet(ctx, m.Id, host, "running")
+}
+
+func statusSet(ctx *ctx.Context, id int64, host, status string) error {
+	return DB(ctx).Table(tht(id)).Where("id=? and host=?", id, host).Update("status", status).Error
+}
+
+func redoHost(ctx *ctx.Context, id int64, host, action string) error {
+	count, err := IbexCount(DB(ctx).Model(&TaskHostDoing{}).Where("id=? and host=?", id, host))
+	if err != nil {
+		return err
+	}
+
+	now := time.Now().Unix()
+	if count == 0 {
+		err = DB(ctx).Table("task_host_doing").Create(map[string]interface{}{
+			"id":     id,
+			"host":   host,
+			"clock":  now,
+			"action": action,
+		}).Error
+	} else {
+		err = DB(ctx).Table("task_host_doing").Where("id=? and host=? and action <> ?", id, host, action).Updates(map[string]interface{}{
+			"clock":  now,
+			"action": action,
+		}).Error
+	}
+	return err
+}
+
+func (m *TaskMeta) HostStrs(ctx *ctx.Context) ([]string, error) {
+	var ret []string
+	err := DB(ctx).Table(tht(m.Id)).Where("id=?", m.Id).Order("ii").Pluck("host", &ret).Error
+	return ret, err
+}
+
+func (m *TaskMeta) Stdouts(ctx *ctx.Context) ([]TaskHost, error) {
+	var ret []TaskHost
+	err := DB(ctx).Table(tht(m.Id)).Where("id=?", m.Id).Select("id", "host", "status", "stdout").Order("ii").Find(&ret).Error
+	return ret, err
+}
+
+func (m *TaskMeta) Stderrs(ctx *ctx.Context) ([]TaskHost, error) {
+	var ret []TaskHost
+	err := DB(ctx).Table(tht(m.Id)).Where("id=?", m.Id).Select("id", "host", "status", "stderr").Order("ii").Find(&ret).Error
+	return ret, err
+}
+
+func TaskMetaTotal(ctx *ctx.Context, creator, query string, before time.Time) (int64, error) {
+	session := DB(ctx).Model(&TaskMeta{})
+
+	session = session.Where("created > '" + before.Format("2006-01-02 15:04:05") + "'")
+
+	if creator != "" {
+		session = session.Where("creator = ?", creator)
+	}
+
+	if query != "" {
+		// q1 q2 -q3
+		arr := strings.Fields(query)
+		for i := 0; i < len(arr); i++ {
+			if arr[i] == "" {
+				continue
+			}
+			if strings.HasPrefix(arr[i], "-") {
+				q := "%" + arr[i][1:] + "%"
+				session = session.Where("title not like ?", q)
+			} else {
+				q := "%" + arr[i] + "%"
+				session = session.Where("title like ?", q)
+			}
+		}
+	}
+
+	return IbexCount(session)
+}
+
+func TaskMetaGets(ctx *ctx.Context, creator, query string, before time.Time, limit, offset int) ([]TaskMeta, error) {
+	session := DB(ctx).Model(&TaskMeta{}).Order("created desc").Limit(limit).Offset(offset)
+
+	session = session.Where("created > '" + before.Format("2006-01-02 15:04:05") + "'")
+
+	if creator != "" {
+		session = session.Where("creator = ?", creator)
+	}
+
+	if query != "" {
+		// q1 q2 -q3
+		arr := strings.Fields(query)
+		for i := 0; i < len(arr); i++ {
+			if arr[i] == "" {
+				continue
+			}
+			if strings.HasPrefix(arr[i], "-") {
+				q := "%" + arr[i][1:] + "%"
+				session = session.Where("title not like ?", q)
+			} else {
+				q := "%" + arr[i] + "%"
+				session = session.Where("title like ?", q)
+			}
+		}
+	}
+
+	var objs []TaskMeta
+	err := session.Find(&objs).Error
+	return objs, err
+}

models/ibex_task_scheduler.go

@@ -0,0 +1,47 @@
+package models
+
+import (
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+
+	"gorm.io/gorm"
+)
+
+type TaskScheduler struct {
+	Id        int64  `gorm:"column:id;primaryKey"`
+	Scheduler string `gorm:"column:scheduler;size:128;not null;default:''"`
+}
+
+func (TaskScheduler) TableName() string {
+	return "task_scheduler"
+}
+
+func TasksOfScheduler(ctx *ctx.Context, scheduler string) ([]int64, error) {
+	var ids []int64
+	err := DB(ctx).Model(&TaskScheduler{}).Where("scheduler = ?", scheduler).Pluck("id", &ids).Error
+	return ids, err
+}
+
+func TakeOverTask(ctx *ctx.Context, id int64, pre, current string) (bool, error) {
+	ret := DB(ctx).Model(&TaskScheduler{}).Where("id = ? and scheduler = ?", id, pre).Update("scheduler", current)
+	if ret.Error != nil {
+		return false, ret.Error
+	}
+
+	return ret.RowsAffected > 0, nil
+}
+
+func OrphanTaskIds(ctx *ctx.Context) ([]int64, error) {
+	var ids []int64
+	err := DB(ctx).Model(&TaskScheduler{}).Where("scheduler = ''").Pluck("id", &ids).Error
+	return ids, err
+}
+
+func CleanDoneTask(ctx *ctx.Context, id int64) error {
+	return DB(ctx).Transaction(func(tx *gorm.DB) error {
+		if err := tx.Where("id = ?", id).Delete(&TaskScheduler{}).Error; err != nil {
+			return err
+		}
+
+		return tx.Where("id = ?", id).Delete(&TaskAction{}).Error
+	})
+}

models/ibex_task_scheduler_health.go

@@ -0,0 +1,47 @@
+package models
+
+import (
+	"github.com/ccfos/nightingale/v6/pkg/ctx"
+
+	"time"
+)
+
+type TaskSchedulerHealth struct {
+	Scheduler string `gorm:"column:scheduler;uniqueIndex;size:128;not null"`
+	Clock     int64  `gorm:"column:clock;not null;index"`
+}
+
+func (TaskSchedulerHealth) TableName() string {
+	return "task_scheduler_health"
+}
+
+func TaskSchedulerHeartbeat(ctx *ctx.Context, scheduler string) error {
+	var cnt int64
+	err := DB(ctx).Model(&TaskSchedulerHealth{}).Where("scheduler = ?", scheduler).Count(&cnt).Error
+	if err != nil {
+		return err
+	}
+
+	if cnt == 0 {
+		ret := DB(ctx).Create(&TaskSchedulerHealth{
+			Scheduler: scheduler,
+			Clock:     time.Now().Unix(),
+		})
+		err = ret.Error
+	} else {
+		err = DB(ctx).Model(&TaskSchedulerHealth{}).Where("scheduler = ?", scheduler).Update("clock", time.Now().Unix()).Error
+	}
+
+	return err
+}
+
+func DeadTaskSchedulers(ctx *ctx.Context) ([]string, error) {
+	clock := time.Now().Unix() - 10
+	var arr []string
+	err := DB(ctx).Model(&TaskSchedulerHealth{}).Where("clock < ?", clock).Pluck("scheduler", &arr).Error
+	return arr, err
+}
+
+func DelDeadTaskScheduler(ctx *ctx.Context, scheduler string) error {
+	return DB(ctx).Where("scheduler = ?", scheduler).Delete(&TaskSchedulerHealth{}).Error
+}

models/migrate/migrate.go

@@ -2,11 +2,9 @@ package migrate
 
 import (
 	"fmt"
-
 	"github.com/ccfos/nightingale/v6/models"
 	"github.com/ccfos/nightingale/v6/pkg/ormx"
 
-	imodels "github.com/flashcatcloud/ibex/src/models"
 	"github.com/toolkits/pkg/logger"
 	"gorm.io/driver/mysql"
 	"gorm.io/gorm"
@@ -28,7 +26,7 @@ func MigrateIbexTables(db *gorm.DB) {
 		db = db.Set("gorm:table_options", tableOptions)
 	}
 
-	dts := []interface{}{&imodels.TaskMeta{}, &imodels.TaskScheduler{}, &imodels.TaskSchedulerHealth{}, &imodels.TaskHostDoing{}, &imodels.TaskAction{}}
+	dts := []interface{}{&models.TaskMeta{}, &models.TaskScheduler{}, &models.TaskSchedulerHealth{}, &models.TaskHostDoing{}, &models.TaskAction{}}
 	for _, dt := range dts {
 		err := db.AutoMigrate(dt)
 		if err != nil {
@@ -38,7 +36,7 @@ func MigrateIbexTables(db *gorm.DB) {
 
 	for i := 0; i < 100; i++ {
 		tableName := fmt.Sprintf("task_host_%d", i)
-		err := db.Table(tableName).AutoMigrate(&imodels.TaskHost{})
+		err := db.Table(tableName).AutoMigrate(&models.TaskHost{})
 		if err != nil {
 			logger.Errorf("failed to migrate table:%s %v", tableName, err)
 		}
@@ -58,8 +56,7 @@ func MigrateTables(db *gorm.DB) error {
 	dts := []interface{}{&RecordingRule{}, &AlertRule{}, &AlertSubscribe{}, &AlertMute{},
 		&TaskRecord{}, &ChartShare{}, &Target{}, &Configs{}, &Datasource{}, &NotifyTpl{},
 		&Board{}, &BoardBusigroup{}, &Users{}, &SsoConfig{}, &models.BuiltinMetric{},
-		&models.MetricFilter{}, &models.BuiltinComponent{}, &models.NotificaitonRecord{},
-		&models.TargetBusiGroup{}}
+		&models.MetricFilter{}, &models.BuiltinComponent{}, &models.NotificaitonRecord{}}
 
 	if !columnHasIndex(db, &AlertHisEvent{}, "original_tags") ||
 		!columnHasIndex(db, &AlertCurEvent{}, "original_tags") {
@@ -89,7 +86,7 @@ func MigrateTables(db *gorm.DB) error {
 	for _, dt := range dts {
 		err := db.AutoMigrate(dt)
 		if err != nil {
-			logger.Errorf("failed to migrate table:%v %v", dt, err)
+			logger.Errorf("failed to migrate table: %v", err)
 		}
 	}
 
@@ -277,6 +274,5 @@ type SsoConfig struct {
 }
 
 type BuiltinPayloads struct {
-	UUID        int64 `json:"uuid" gorm:"type:bigint;not null;index:idx_uuid;comment:'uuid of payload'"`
-	ComponentID int64 `json:"component_id" gorm:"type:bigint;index:idx_component,sort:asc;not null;default:0;comment:'component_id of payload'"`
+	UUID int64 `json:"uuid" gorm:"type:bigint;not null;index:idx_uuid;comment:'uuid of payload'"`
 }

models/target.go

@@ -1,14 +1,12 @@
 package models
 
 import (
-	"log"
 	"sort"
 	"strings"
 	"time"
 
 	"github.com/ccfos/nightingale/v6/pkg/ctx"
 	"github.com/ccfos/nightingale/v6/pkg/poster"
-	"golang.org/x/exp/slices"
 
 	"github.com/pkg/errors"
 	"github.com/toolkits/pkg/container/set"
@@ -16,12 +14,10 @@ import (
 	"gorm.io/gorm"
 )
 
-type TargetDeleteHookFunc func(ctx *ctx.Context, idents []string) error
-
 type Target struct {
 	Id           int64             `json:"id" gorm:"primaryKey"`
 	GroupId      int64             `json:"group_id"`
-	GroupObjs    []*BusiGroup      `json:"group_objs" gorm:"-"`
+	GroupObj     *BusiGroup        `json:"group_obj" gorm:"-"`
 	Ident        string            `json:"ident"`
 	Note         string            `json:"note"`
 	Tags         string            `json:"-"` // user tags
@@ -42,7 +38,6 @@ type Target struct {
 	CpuUtil    float64 `json:"cpu_util" gorm:"-"`
 	Arch       string  `json:"arch" gorm:"-"`
 	RemoteAddr string  `json:"remote_addr" gorm:"-"`
-	GroupIds   []int64 `json:"group_ids" gorm:"-"`
 }
 
 func (t *Target) TableName() string {
@@ -50,49 +45,26 @@ func (t *Target) TableName() string {
 }
 
 func (t *Target) FillGroup(ctx *ctx.Context, cache map[int64]*BusiGroup) error {
-	var err error
-	if len(t.GroupIds) == 0 {
-		t.GroupIds, err = TargetGroupIdsGetByIdent(ctx, t.Ident)
-		if err != nil {
-			return errors.WithMessage(err, "failed to get target gids")
-		}
-		t.GroupObjs = make([]*BusiGroup, 0, len(t.GroupIds))
+	if t.GroupId <= 0 {
+		return nil
 	}
 
-	for _, gid := range t.GroupIds {
-		bg, has := cache[gid]
-		if has && bg != nil {
-			t.GroupObjs = append(t.GroupObjs, bg)
-			continue
-		}
-
-		bg, err := BusiGroupGetById(ctx, gid)
-		if err != nil {
-			return errors.WithMessage(err, "failed to get busi group")
-		}
-
-		if bg == nil {
-			continue
-		}
-
-		t.GroupObjs = append(t.GroupObjs, bg)
-		cache[gid] = bg
+	bg, has := cache[t.GroupId]
+	if has {
+		t.GroupObj = bg
+		return nil
 	}
 
+	bg, err := BusiGroupGetById(ctx, t.GroupId)
+	if err != nil {
+		return errors.WithMessage(err, "failed to get busi group")
+	}
+
+	t.GroupObj = bg
+	cache[t.GroupId] = bg
 	return nil
 }
 
-func (t *Target) MatchGroupId(gid ...int64) bool {
-	for _, tgId := range t.GroupIds {
-		for _, id := range gid {
-			if tgId == id {
-				return true
-			}
-		}
-	}
-	return false
-}
-
 func (t *Target) AfterFind(tx *gorm.DB) (err error) {
 	delta := time.Now().Unix() - t.UpdateAt
 	if delta < 60 {
@@ -119,39 +91,19 @@ func TargetStatistics(ctx *ctx.Context) (*Statistics, error) {
 	return stats[0], nil
 }
 
-func TargetDel(ctx *ctx.Context, idents []string, deleteHook TargetDeleteHookFunc) error {
+func TargetDel(ctx *ctx.Context, idents []string) error {
 	if len(idents) == 0 {
 		panic("idents empty")
 	}
-
-	return DB(ctx).Transaction(func(tx *gorm.DB) error {
-		txErr := tx.Where("ident in ?", idents).Delete(new(Target)).Error
-		if txErr != nil {
-			return txErr
-		}
-		txErr = deleteHook(ctx, idents)
-		if txErr != nil {
-			return txErr
-		}
-		return nil
-	})
+	return DB(ctx).Where("ident in ?", idents).Delete(new(Target)).Error
 }
 
 type BuildTargetWhereOption func(session *gorm.DB) *gorm.DB
 
 func BuildTargetWhereWithBgids(bgids []int64) BuildTargetWhereOption {
 	return func(session *gorm.DB) *gorm.DB {
-		if len(bgids) == 1 && bgids[0] == 0 {
-			session = session.Joins("left join target_busi_group on target.ident = " +
-				"target_busi_group.target_ident").Where("target_busi_group.target_ident is null")
-		} else if len(bgids) > 0 {
-			if slices.Contains(bgids, 0) {
-				session = session.Joins("left join target_busi_group on target.ident = target_busi_group.target_ident").
-					Where("target_busi_group.target_ident is null OR target_busi_group.group_id in (?)", bgids)
-			} else {
-				session = session.Joins("join target_busi_group on target.ident = "+
-					"target_busi_group.target_ident").Where("target_busi_group.group_id in (?)", bgids)
-			}
+		if len(bgids) > 0 {
+			session = session.Where("group_id in (?)", bgids)
 		}
 		return session
 	}
@@ -166,22 +118,13 @@ func BuildTargetWhereWithDsIds(dsIds []int64) BuildTargetWhereOption {
 	}
 }
 
-func BuildTargetWhereWithHosts(hosts []string) BuildTargetWhereOption {
-	return func(session *gorm.DB) *gorm.DB {
-		if len(hosts) > 0 {
-			session = session.Where("ident in (?) or host_ip in (?)", hosts, hosts)
-		}
-		return session
-	}
-}
-
 func BuildTargetWhereWithQuery(query string) BuildTargetWhereOption {
 	return func(session *gorm.DB) *gorm.DB {
 		if query != "" {
 			arr := strings.Fields(query)
 			for i := 0; i < len(arr); i++ {
 				q := "%" + arr[i] + "%"
-				session = session.Where("ident like ? or host_ip like ? or note like ? or tags like ? or host_tags like ? or os like ?", q, q, q, q, q, q)
+				session = session.Where("ident like ? or note like ? or tags like ? or host_tags like ? or os like ?", q, q, q, q, q)
 			}
 		}
 		return session
@@ -191,18 +134,18 @@ func BuildTargetWhereWithQuery(query string) BuildTargetWhereOption {
 func BuildTargetWhereWithDowntime(downtime int64) BuildTargetWhereOption {
 	return func(session *gorm.DB) *gorm.DB {
 		if downtime > 0 {
-			session = session.Where("target.update_at < ?", time.Now().Unix()-downtime)
+			session = session.Where("update_at < ?", time.Now().Unix()-downtime)
 		}
 		return session
 	}
 }
 
 func buildTargetWhere(ctx *ctx.Context, options ...BuildTargetWhereOption) *gorm.DB {
-	sub := DB(ctx).Model(&Target{}).Distinct("target.ident")
+	session := DB(ctx).Model(&Target{})
 	for _, opt := range options {
-		sub = opt(sub)
+		session = opt(session)
 	}
-	return DB(ctx).Model(&Target{}).Where("ident in (?)", sub)
+	return session
 }
 
 func TargetTotal(ctx *ctx.Context, options ...BuildTargetWhereOption) (int64, error) {
@@ -260,18 +203,15 @@ func MissTargetCountByFilter(ctx *ctx.Context, query []map[string]interface{}, t
 }
 
 func TargetFilterQueryBuild(ctx *ctx.Context, query []map[string]interface{}, limit, offset int) *gorm.DB {
-	sub := DB(ctx).Model(&Target{}).Distinct("target.ident").Joins("left join " +
-		"target_busi_group on target.ident = target_busi_group.target_ident")
+	session := DB(ctx).Model(&Target{})
 	for _, q := range query {
 		tx := DB(ctx).Model(&Target{})
 		for k, v := range q {
 			tx = tx.Or(k, v)
 		}
-		sub = sub.Where(tx)
+		session = session.Where(tx)
 	}
 
-	session := DB(ctx).Model(&Target{}).Where("ident in (?)", sub)
-
 	if limit > 0 {
 		session = session.Limit(limit).Offset(offset)
 	}
@@ -287,20 +227,9 @@ func TargetGetsAll(ctx *ctx.Context) ([]*Target, error) {
 
 	var lst []*Target
 	err := DB(ctx).Model(&Target{}).Find(&lst).Error
-	if err != nil {
-		return lst, err
-	}
-
-	tgs, err := TargetBusiGroupsGetAll(ctx)
-	if err != nil {
-		return lst, err
-	}
-
 	for i := 0; i < len(lst); i++ {
 		lst[i].FillTagsMap()
-		lst[i].GroupIds = tgs[lst[i].Ident]
 	}
-
 	return lst, err
 }
 
@@ -558,73 +487,3 @@ func IdentsFilter(ctx *ctx.Context, idents []string, where string, args ...inter
 func (m *Target) UpdateFieldsMap(ctx *ctx.Context, fields map[string]interface{}) error {
 	return DB(ctx).Model(m).Updates(fields).Error
 }
-
-func MigrateBg(ctx *ctx.Context, bgLabelKey string) {
-	// 1. 判断是否已经完成迁移
-	var maxGroupId int64
-	if err := DB(ctx).Model(&Target{}).Select("MAX(group_id)").Scan(&maxGroupId).Error; err != nil {
-		log.Println("failed to get max group_id from target table, err:", err)
-		return
-	}
-
-	if maxGroupId == 0 {
-		log.Println("migration bgid has been completed.")
-		return
-	}
-
-	err := DoMigrateBg(ctx, bgLabelKey)
-	if err != nil {
-		log.Println("failed to migrate bgid, err:", err)
-		return
-	}
-
-	log.Println("migration bgid has been completed")
-}
-
-func DoMigrateBg(ctx *ctx.Context, bgLabelKey string) error {
-	// 2. 获取全量 target
-	targets, err := TargetGetsAll(ctx)
-	if err != nil {
-		return err
-	}
-
-	// 3. 获取全量 busi_group
-	bgs, err := BusiGroupGetAll(ctx)
-	if err != nil {
-		return err
-	}
-
-	bgById := make(map[int64]*BusiGroup, len(bgs))
-	for _, bg := range bgs {
-		bgById[bg.Id] = bg
-	}
-
-	// 4. 如果某 busi_group 有 label，将其存至对应的 target tags 中
-	for _, t := range targets {
-		if t.GroupId == 0 {
-			continue
-		}
-		err := DB(ctx).Transaction(func(tx *gorm.DB) error {
-			// 4.1 将 group_id 迁移至关联表
-			if err := TargetBindBgids(ctx, []string{t.Ident}, []int64{t.GroupId}); err != nil {
-				return err
-			}
-			if err := TargetUpdateBgid(ctx, []string{t.Ident}, 0, false); err != nil {
-				return err
-			}
-
-			// 4.2 判断该机器是否需要新增 tag
-			if bg, ok := bgById[t.GroupId]; !ok || bg.LabelEnable == 0 ||
-				strings.Contains(t.Tags, bgLabelKey+"=") {
-				return nil
-			} else {
-				return t.AddTags(ctx, []string{bgLabelKey + "=" + bg.LabelValue})
-			}
-		})
-		if err != nil {
-			log.Printf("failed to migrate %v bg, err: %v\n", t.Ident, err)
-			continue
-		}
-	}
-	return nil
-}

models/target_busi_group.go

@@ -1,158 +0,0 @@
-package models
-
-import (
-	"time"
-
-	"github.com/ccfos/nightingale/v6/pkg/ctx"
-
-	"gorm.io/gorm"
-	"gorm.io/gorm/clause"
-)
-
-type TargetBusiGroup struct {
-	Id          int64  `json:"id" gorm:"primaryKey;type:bigint;autoIncrement"`
-	TargetIdent string `json:"target_ident" gorm:"type:varchar(191);not null;index:idx_target_group,unique,priority:1"`
-	GroupId     int64  `json:"group_id" gorm:"type:bigint;not null;index:idx_target_group,unique,priority:2"`
-	UpdateAt    int64  `json:"update_at" gorm:"type:bigint;not null"`
-}
-
-func (t *TargetBusiGroup) TableName() string {
-	return "target_busi_group"
-}
-
-func TargetBusiGroupsGetAll(ctx *ctx.Context) (map[string][]int64, error) {
-	var lst []*TargetBusiGroup
-	err := DB(ctx).Find(&lst).Error
-	if err != nil {
-		return nil, err
-	}
-	tgs := make(map[string][]int64)
-	for _, tg := range lst {
-		tgs[tg.TargetIdent] = append(tgs[tg.TargetIdent], tg.GroupId)
-	}
-	return tgs, nil
-}
-
-func TargetGroupIdsGetByIdent(ctx *ctx.Context, ident string) ([]int64, error) {
-	var lst []*TargetBusiGroup
-	err := DB(ctx).Where("target_ident = ?", ident).Find(&lst).Error
-	if err != nil {
-		return nil, err
-	}
-	groupIds := make([]int64, 0, len(lst))
-	for _, tg := range lst {
-		groupIds = append(groupIds, tg.GroupId)
-	}
-	return groupIds, nil
-}
-
-func TargetGroupIdsGetByIdents(ctx *ctx.Context, idents []string) ([]int64, error) {
-	var groupIds []int64
-	err := DB(ctx).Model(&TargetBusiGroup{}).
-		Where("target_ident IN ?", idents).
-		Distinct().
-		Pluck("group_id", &groupIds).
-		Error
-	if err != nil {
-		return nil, err
-	}
-
-	return groupIds, nil
-}
-
-func TargetBindBgids(ctx *ctx.Context, idents []string, bgids []int64) error {
-	lst := make([]TargetBusiGroup, 0, len(bgids)*len(idents))
-	updateAt := time.Now().Unix()
-	for _, bgid := range bgids {
-		for _, ident := range idents {
-			cur := TargetBusiGroup{
-				TargetIdent: ident,
-				GroupId:     bgid,
-				UpdateAt:    updateAt,
-			}
-			lst = append(lst, cur)
-		}
-	}
-
-	var cl clause.Expression = clause.Insert{Modifier: "ignore"}
-	switch DB(ctx).Dialector.Name() {
-	case "sqlite":
-		cl = clause.Insert{Modifier: "or ignore"}
-	case "postgres":
-		cl = clause.OnConflict{DoNothing: true}
-	}
-	return DB(ctx).Clauses(cl).CreateInBatches(&lst, 10).Error
-}
-
-func TargetUnbindBgids(ctx *ctx.Context, idents []string, bgids []int64) error {
-	return DB(ctx).Where("target_ident in ? and group_id in ?",
-		idents, bgids).Delete(&TargetBusiGroup{}).Error
-}
-
-func TargetDeleteBgids(ctx *ctx.Context, idents []string) error {
-	return DB(ctx).Where("target_ident in ?", idents).Delete(&TargetBusiGroup{}).Error
-}
-
-func TargetOverrideBgids(ctx *ctx.Context, idents []string, bgids []int64) error {
-	return DB(ctx).Transaction(func(tx *gorm.DB) error {
-		// 先删除旧的关联
-		if err := tx.Where("target_ident IN ?", idents).Delete(&TargetBusiGroup{}).Error; err != nil {
-			return err
-		}
-
-		// 准备新的关联数据
-		lst := make([]TargetBusiGroup, 0, len(bgids)*len(idents))
-		updateAt := time.Now().Unix()
-		for _, ident := range idents {
-			for _, bgid := range bgids {
-				cur := TargetBusiGroup{
-					TargetIdent: ident,
-					GroupId:     bgid,
-					UpdateAt:    updateAt,
-				}
-				lst = append(lst, cur)
-			}
-		}
-
-		if len(lst) == 0 {
-			return nil
-		}
-
-		// 添加新的关联
-		var cl clause.Expression = clause.Insert{Modifier: "ignore"}
-		switch tx.Dialector.Name() {
-		case "sqlite":
-			cl = clause.Insert{Modifier: "or ignore"}
-		case "postgres":
-			cl = clause.OnConflict{DoNothing: true}
-		}
-		return tx.Clauses(cl).CreateInBatches(&lst, 10).Error
-	})
-}
-
-func SeparateTargetIdents(ctx *ctx.Context, idents []string) (existing, nonExisting []string, err error) {
-	existingMap := make(map[string]bool)
-
-	// 查询已存在的 idents 并直接填充 map
-	err = DB(ctx).Model(&TargetBusiGroup{}).
-		Where("target_ident IN ?", idents).
-		Distinct().
-		Pluck("target_ident", &existing).
-		Error
-	if err != nil {
-		return nil, nil, err
-	}
-
-	for _, ident := range existing {
-		existingMap[ident] = true
-	}
-
-	// 分离不存在的 idents
-	for _, ident := range idents {
-		if !existingMap[ident] {
-			nonExisting = append(nonExisting, ident)
-		}
-	}
-
-	return
-}

models/user.go

@@ -296,34 +296,6 @@ func UserGet(ctx *ctx.Context, where string, args ...interface{}) (*User, error)
 	return lst[0], nil
 }
 
-func UsersGet(ctx *ctx.Context, where string, args ...interface{}) ([]*User, error) {
-	var lst []*User
-	err := DB(ctx).Where(where, args...).Find(&lst).Error
-	if err != nil {
-		return nil, err
-	}
-
-	for _, user := range lst {
-		user.RolesLst = strings.Fields(user.Roles)
-		user.Admin = user.IsAdmin()
-	}
-
-	return lst, nil
-}
-
-func UserMapGet(ctx *ctx.Context, where string, args ...interface{}) map[string]*User {
-	lst, err := UsersGet(ctx, where, args...)
-	if err != nil {
-		logger.Errorf("UsersGet err: %v", err)
-		return nil
-	}
-	um := make(map[string]*User, len(lst))
-	for _, user := range lst {
-		um[user.Username] = user
-	}
-	return um
-}
-
 func UserGetByUsername(ctx *ctx.Context, username string) (*User, error) {
 	return UserGet(ctx, "username=?", username)
 }
@@ -732,10 +704,7 @@ func (u *User) NopriIdents(ctx *ctx.Context, idents []string) ([]string, error)
 	}
 
 	var allowedIdents []string
-	sub := DB(ctx).Model(&Target{}).Distinct("target.ident").
-		Joins("join target_busi_group on target.ident = target_busi_group.target_ident").
-		Where("target_busi_group.group_id in (?)", bgids)
-	err = DB(ctx).Model(&Target{}).Where("ident in (?)", sub).Pluck("ident", &allowedIdents).Error
+	err = DB(ctx).Model(&Target{}).Where("group_id in ?", bgids).Pluck("ident", &allowedIdents).Error
 	if err != nil {
 		return []string{}, err
 	}
@@ -767,11 +736,7 @@ func (u *User) BusiGroups(ctx *ctx.Context, limit int, query string, all ...bool
 				return lst, nil
 			}
 
-			t.GroupIds, err = TargetGroupIdsGetByIdent(ctx, t.Ident)
-			if err != nil {
-				return nil, err
-			}
-			err = DB(ctx).Order("name").Limit(limit).Where("id in ?", t.GroupIds).Find(&lst).Error
+			err = DB(ctx).Order("name").Limit(limit).Where("id=?", t.GroupId).Find(&lst).Error
 		}
 
 		return lst, err
@@ -803,12 +768,8 @@ func (u *User) BusiGroups(ctx *ctx.Context, limit int, query string, all ...bool
 			return lst, err
 		}
 
-		t.GroupIds, err = TargetGroupIdsGetByIdent(ctx, t.Ident)
-		if err != nil {
-			return nil, err
-		}
-		if t != nil && t.MatchGroupId(busiGroupIds...) {
-			err = DB(ctx).Order("name").Limit(limit).Where("id in ?", t.GroupIds).Find(&lst).Error
+		if t != nil && slice.ContainsInt64(busiGroupIds, t.GroupId) {
+			err = DB(ctx).Order("name").Limit(limit).Where("id=?", t.GroupId).Find(&lst).Error
 		}
 	}
 

pkg/ctx/ctx.go

@@ -2,20 +2,21 @@ package ctx
 
 import (
 	"context"
+	"github.com/ccfos/nightingale/v6/storage"
 
 	"github.com/ccfos/nightingale/v6/conf"
-
 	"gorm.io/gorm"
 )
 
 type Context struct {
 	DB        *gorm.DB
+	Redis     storage.Redis
 	CenterApi conf.CenterApi
 	Ctx       context.Context
 	IsCenter  bool
 }
 
-func NewContext(ctx context.Context, db *gorm.DB, isCenter bool, centerApis ...conf.CenterApi) *Context {
+func NewContext(ctx context.Context, db *gorm.DB, redis storage.Redis, isCenter bool, centerApis ...conf.CenterApi) *Context {
 	var api conf.CenterApi
 	if len(centerApis) > 0 {
 		api = centerApis[0]
@@ -24,6 +25,7 @@ func NewContext(ctx context.Context, db *gorm.DB, isCenter bool, centerApis ...c
 	return &Context{
 		Ctx:       ctx,
 		DB:        db,
+		Redis:     redis,
 		CenterApi: api,
 		IsCenter:  isCenter,
 	}

pkg/httpx/httpx.go

@@ -120,7 +120,7 @@ func GinEngine(mode string, cfg Config) *gin.Engine {
 	return r
 }
 
-func Init(cfg Config, handler http.Handler) func() {
+func Init(cfg Config, ctx context.Context, handler http.Handler) func() {
 	addr := fmt.Sprintf("%s:%d", cfg.Host, cfg.Port)
 	srv := &http.Server{
 		Addr:         addr,
@@ -146,7 +146,7 @@ func Init(cfg Config, handler http.Handler) func() {
 	}()
 
 	return func() {
-		ctx, cancel := context.WithTimeout(context.Background(), time.Second*time.Duration(cfg.ShutdownTimeout))
+		ctx, cancel := context.WithTimeout(ctx, time.Second*time.Duration(cfg.ShutdownTimeout))
 		defer cancel()
 
 		srv.SetKeepAlivesEnabled(false)

pkg/i18nx/var.go

@@ -104,56 +104,6 @@ var I18N = `
     "builtin metric already exists":"内置指标已存在",
     "AlertRule already exists":"告警规则已存在",
     "This functionality has not been enabled. Please contact the system administrator to activate it.":"此功能尚未启用。请联系系统管理员启用"
-  },
-  "ja_JP": {
-    "Username or password invalid": "ユーザー名またはパスワードが無効です",
-    "incorrect verification code": "認証コードが正しくありません",
-    "roles empty": "役割を空にすることはできません",
-    "Username already exists": "このユーザー名は既に存在します。別のユーザー名を使用してください",
-    "failed to count user-groups": "データの検証に失敗しました。もう一度お試しください",
-    "UserGroup already exists": "グループ名は既に存在します。別の名前を使用してください",
-    "members empty": "メンバーを空にすることはできません",
-    "At least one team have rw permission": "少なくとも1つのチームに読み書き権限が必要です",
-    "Failed to create BusiGroup(%s)": "[%s]の作成に失敗しました。もう一度お試しください",
-    "business group id invalid": "ビジネスグループIDが正しくありません",
-    "idents empty": "監視対象を空にすることはできません",
-    "invalid tag(%s)": "タグ[%s]が無効です",
-    "invalid tagkey(%s): cannot contains . ": "タグキー[%s]にドット(.)を含めることはできません",
-    "invalid tagkey(%s): cannot contains _ ": "タグキー[%s]にアンダースコア(_)を含めることはできません",
-    "invalid tagkey(%s)": "タグキー[%s]が無効です",
-    "duplicate tagkey(%s)": "タグキー(%s)が重複しています",
-    "name is empty": "名前を空にすることはできません",
-    "Ident duplicate": "ダッシュボードの一意の識別子が既に存在します",
-    "No such dashboard": "ダッシュボードが存在しません",
-    "Name has invalid characters": "名前に無効な文字が含まれています",
-    "Name is blank": "名前を空白にすることはできません",
-    "forbidden": "権限がありません",
-    "builtin alerts is empty, file: %s": "ビルトインアラートテンプレートが空です %s",
-    "input json is empty": "提出内容を空にすることはできません",
-    "fields empty": "選択フィールドを空にすることはできません",
-    "No such AlertRule": "そのようなアラートルールはありません",
-    "GroupId(%d) invalid": "ビジネスグループIDが無効です",
-    "No such recording rule": "そのような記録ルールはありません",
-    "tags is blank": "タグを空白にすることはできません",
-    "oops... etime(%d) <= btime(%d)": "開始時間は終了時間より大きくすることはできません",
-    "group_id invalid": "ビジネスグループが無効です",
-    "No such AlertMute": "そのようなアラートミュートルールはありません",
-    "rule_id and tags are both blank": "アラートルールとタグを同時に空にすることはできません",
-    "rule is blank": "ルールを空にすることはできません",
-    "rule invalid": "ルールが無効です。正しいかどうか確認してください",
-    "unsupported field: %s": "フィールド %s はサポートされていません",
-    "arg(batch) should be nonnegative": "batchは負の数にできません",
-    "arg(tolerance) should be nonnegative": "toleranceは負の数にできません",
-    "arg(timeout) should be nonnegative": "timeoutは負の数にできません",
-    "arg(timeout) longer than five days": "timeoutは5日を超えることはできません",
-    "arg(title) is required": "titleは必須項目です",
-    "created task.id is zero": "作成されたタスクIDがゼロです",
-    "invalid ibex address: %s": "ibex %s のアドレスが無効です",
-    "url path invalid": "URLパスが無効です",
-    "no such server": "そのようなインスタンスはありません",
-    "admin role can not be modified": "管理者ロールは変更できません",
-    "builtin payload already exists": "ビルトインテンプレートは既に存在します",
-    "This functionality has not been enabled. Please contact the system administrator to activate it.": "この機能はまだ有効になっていません。システム管理者に連絡して有効にしてください"
   }
 }
 `

pkg/ldapx/ldapx.go

@@ -43,7 +43,6 @@ type SsoClient struct {
 	Host            string
 	Port            int
 	BaseDn          string
-	BaseDns         []string
 	BindUser        string
 	BindPass        string
 	SyncAdd         bool
@@ -132,8 +131,6 @@ func (s *SsoClient) Reload(cf Config) {
 	if s.SyncInterval > 0 {
 		s.Ticker.Reset(s.SyncInterval * time.Second)
 	}
-
-	s.BaseDns = strings.Split(s.BaseDn, "|")
 }
 
 func (s *SsoClient) Copy() *SsoClient {
@@ -161,40 +158,27 @@ func (s *SsoClient) LoginCheck(user, pass string) (*ldap.SearchResult, error) {
 	}
 	defer conn.Close()
 
-	srs, err := lc.ldapReq(conn, lc.AuthFilter, user)
+	sr, err := lc.ldapReq(conn, lc.AuthFilter, user)
 	if err != nil {
 		return nil, fmt.Errorf("ldap.error: ldap search fail: %v", err)
 	}
 
-	var sr *ldap.SearchResult
-
-	for i := range srs {
-		if srs[i] == nil || len(srs[i].Entries) == 0 {
-			continue
-		}
-
-		// 多个 dn 中，账号的唯一性由 LDAP 保证
-		if len(srs[i].Entries) > 1 {
-			return nil, fmt.Errorf("ldap.error: search user(%s), multi entries found", user)
-		}
-
-		sr = srs[i]
-
-		if err := conn.Bind(srs[i].Entries[0].DN, pass); err != nil {
-			return nil, fmt.Errorf("username or password invalid")
-		}
-
-		for _, info := range srs[i].Entries[0].Attributes {
-			logger.Infof("ldap.info: user(%s) info: %+v", user, info)
-		}
-
-		break
+	if len(sr.Entries) == 0 {
+		return nil, fmt.Errorf("username or password invalid")
 	}
 
-	if sr == nil {
+	if len(sr.Entries) > 1 {
+		return nil, fmt.Errorf("ldap.error: search user(%s), multi entries found", user)
+	}
+
+	if err := conn.Bind(sr.Entries[0].DN, pass); err != nil {
 		return nil, fmt.Errorf("username or password invalid")
 	}
 
+	for _, info := range sr.Entries[0].Attributes {
+		logger.Infof("ldap.info: user(%s) info: %+v", user, info)
+	}
+
 	return sr, nil
 }
 
@@ -234,26 +218,21 @@ func (s *SsoClient) newLdapConn() (*ldap.Conn, error) {
 	return conn, nil
 }
 
-func (s *SsoClient) ldapReq(conn *ldap.Conn, filter string, values ...interface{}) ([]*ldap.SearchResult, error) {
-	srs := make([]*ldap.SearchResult, 0, len(s.BaseDns))
+func (s *SsoClient) ldapReq(conn *ldap.Conn, filter string, values ...interface{}) (*ldap.SearchResult, error) {
+	searchRequest := ldap.NewSearchRequest(
+		s.BaseDn, // The base dn to search
+		ldap.ScopeWholeSubtree, ldap.NeverDerefAliases, 0, 0, false,
+		fmt.Sprintf(filter, values...), // The filter to apply
+		s.genLdapAttributeSearchList(), // A list attributes to retrieve
+		nil,
+	)
 
-	for i := range s.BaseDns {
-		searchRequest := ldap.NewSearchRequest(
-			strings.TrimSpace(s.BaseDns[i]), // The base dn to search
-			ldap.ScopeWholeSubtree, ldap.NeverDerefAliases, 0, 0, false,
-			fmt.Sprintf(filter, values...), // The filter to apply
-			s.genLdapAttributeSearchList(), // A list attributes to retrieve
-			nil,
-		)
-		sr, err := conn.Search(searchRequest)
-		if err != nil {
-			logger.Errorf("ldap.error: ldap search fail: %v", err)
-			continue
-		}
-		srs = append(srs, sr)
+	sr, err := conn.Search(searchRequest)
+	if err != nil {
+		return nil, fmt.Errorf("ldap.error: ldap search fail: %v", err)
 	}
 
-	return srs, nil
+	return sr, nil
 }
 
 // GetUserRolesAndTeams Gets the roles and teams of the user
@@ -323,7 +302,6 @@ func LdapLogin(ctx *ctx.Context, username, pass string, defaultRoles []string, d
 	if err != nil {
 		return nil, err
 	}
-
 	// copy attributes from ldap
 	ldap.RLock()
 	attrs := ldap.Attributes

pkg/ldapx/user_sync.go

@@ -82,35 +82,29 @@ func (s *SsoClient) UserGetAll() (map[string]*models.User, error) {
 	}
 	defer conn.Close()
 
-	srs, err := lc.ldapReq(conn, lc.UserFilter)
+	sr, err := lc.ldapReq(conn, lc.UserFilter)
 	if err != nil {
 		return nil, fmt.Errorf("ldap.error: ldap search fail: %v", err)
 	}
 
-	res := make(map[string]*models.User)
+	res := make(map[string]*models.User, len(sr.Entries))
+	for _, entry := range sr.Entries {
+		attrs := lc.Attributes
+		username := entry.GetAttributeValue(attrs.Username)
+		nickname := entry.GetAttributeValue(attrs.Nickname)
+		email := entry.GetAttributeValue(attrs.Email)
+		phone := entry.GetAttributeValue(attrs.Phone)
 
-	for i := range srs {
-		if srs[i] == nil {
-			continue
+		// Gets the roles and teams for this entry
+		roleTeamMapping := lc.GetUserRolesAndTeams(entry)
+		if len(roleTeamMapping.Roles) == 0 {
+			// No role mapping is configured, the configured default role is used
+			roleTeamMapping.Roles = lc.DefaultRoles
 		}
-		for _, entry := range srs[i].Entries {
-			attrs := lc.Attributes
-			username := entry.GetAttributeValue(attrs.Username)
-			nickname := entry.GetAttributeValue(attrs.Nickname)
-			email := entry.GetAttributeValue(attrs.Email)
-			phone := entry.GetAttributeValue(attrs.Phone)
+		user := new(models.User)
+		user.FullSsoFieldsWithTeams("ldap", username, nickname, phone, email, roleTeamMapping.Roles, roleTeamMapping.Teams)
 
-			// Gets the roles and teams for this entry
-			roleTeamMapping := lc.GetUserRolesAndTeams(entry)
-			if len(roleTeamMapping.Roles) == 0 {
-				// No role mapping is configured, the configured default role is used
-				roleTeamMapping.Roles = lc.DefaultRoles
-			}
-			user := new(models.User)
-			user.FullSsoFieldsWithTeams("ldap", username, nickname, phone, email, roleTeamMapping.Roles, roleTeamMapping.Teams)
-
-			res[entry.GetAttributeValue(attrs.Username)] = user
-		}
+		res[entry.GetAttributeValue(attrs.Username)] = user
 	}
 
 	return res, nil
@@ -178,20 +172,13 @@ func (s *SsoClient) UserExist(username string) (bool, error) {
 	}
 	defer conn.Close()
 
-	srs, err := lc.ldapReq(conn, "(&(%s=%s))", lc.Attributes.Username, username)
+	sr, err := lc.ldapReq(conn, "(&(%s=%s))", lc.Attributes.Username, username)
 	if err != nil {
 		return false, err
 	}
 
-	for i := range srs {
-		if srs[i] == nil {
-			continue
-		}
-
-		if len(srs[i].Entries) > 0 {
-			return true, nil
-		}
-
+	if len(sr.Entries) > 0 {
+		return true, nil
 	}
 
 	return false, nil