This PS sets the ownership of the co-ordination backend location
when using the default file backend. This does not make the service
functional but prevents fatal errors when loading it.
Change-Id: I69a0838469f889172fb061f0adea34ca11d9c8c3
This PS moves the cleaning net managment to a job.
Additionally the gate scripts are updated to enable single node
operation for development use.
Change-Id: Icb7015bcceaf93bc65f60399caf785fbdcf72413
Adds "helm-toolkit.utils.merge" which is a replacement for the
upstream sprig "merge" function which didn't quite do what we
wanted, specifically it didn't merge slices, it just overrode
one with the other.
Change-Id: I6402ffe043c75eb3a63797ef1be5b56f35b0adc4
When helm crashes/fails it can leave tmpcharts directories behind;
update the 'clean' target to remove these.
Change-Id: Ib0fb6e93262fbc6fd97dfdd491df46283dff4a9e
This PS extends the host targeting utility to create daemonsets
for storage focused charts. This PS supports both block device
and bind mount backed journals and OSDs.
Change-Id: Id90b197ba3e4f383eea7ce6137ed77b3ef4e5625
Previously, the ceph version was checked with "ceph -v".
However, this way is only check for the ceph client version.
So I change the way to check the ceph osd version.
Change-Id: I11f44186f671594e6e9e848f3fddb1d23b994b36
* Changed podManagementPolicy to parallel in order to allow recovery
from secondary or third master
After rebooting the whole cluster on purpose or after a power failure
a primary node the cluster can synchronize with is required. This is
usually done automatically by selecting the node with the highest
transaction id. The current implementation of the stateful set starts
the nodes in sequence, preventing the start of further nodes if the
process terminates with error state. Because of this, the cluster may
not come up if the first or second node are not in primary state.
* Elects first node started in primary state as bootstrap source.
* Display warnings and runs mysqld with wsrep-recover on crashed nodes
* Introduces FORCE_RECOVERY argument for crash recovery
In case the primary selection failed, the cluster bootstrap process
must be manually initiated from the most advanced node (highest
committed transaction id). This information is available from the
grastate.dat file in case of a clean shutdown. On crashed nodes an
InnoDB recovery is required to get the last committed transaction id.
start.sh will handle both cases and gives instructions on how to
recover the cluster on a hard failure. If FORCE_RECOVERY was set to
the name of a POD (mariadb-0, mariadb-1, ...), the bootstrap process
will be initiated from the specified node.
DocImpact
Closes-Bug: #1716461
Change-Id: I96a8cb52124f64920a7d9cf21a8924ede78ebf7b
This enables prometheus monitoring for rabbitmq and mariadb in
the armada gate, as the armada gate has deployed prometheus and
can be used to verify functionality
Change-Id: I219a93e934d19470f16fa319df8c3c0d86608416
This PS updates the multinode deploy scripts to show output in the
gates, stop proxying the kubedns service, and use a pvc to back mariadb.
Change-Id: I78caf3f15e4c5ca33eaa1e592e8df958f13be90b
Like a glance, gnocchi, ceph-admin-keyring script should be needed when
execute storage_init.sh in cinder.
Change-Id: I5ab1b474de7cc0a9f80642502d74d1d1f156dea5
Closes-bug: 1744207
This adds a check to openstack-helm for deploying the full osh +
lma stack with armada. It deploys a manifest that will install
a rabbitmq and memcached per openstack service
Depends-On: Id58f85f150054d82c047bd360258ebe9e571360f
Change-Id: Ie2cb29279a7e47570b7faee82828f7ec753ab73f
This PS moves the keystone endpoint jobs to be driven by
a helm-toolkit function providing greater consistency
to the charts in OSH, and reduced tech debt.
Change-Id: I06555171ce93a2b0a8798bc891a2990ea9626276
This PS moves the keystone service jobs to be driven by
a helm-toolkit function providing greater consistency
to the charts in OSH, and reduced tech debt.
Change-Id: I36e012854473fcb942d62391d182ee6a8406bb72
Remove 'connect_timeout' which defined in the '[client]' section which
surprisingly affects mysqldump by causing it to fail as it doesn't
understand this; we're not doing anything that depends on having this
value so this is safe.
This now means mysqldump works (as expected) and we can use the
packaged/pod version of mysqldump for backup purposes.
Change-Id: I9142969e4b8576b19353c063586dd11f91875f0b
LocalWords: mysqldump
This moves the prometheus exporters for services into a directory
in the chart themselves. This keeps the monitoring services
coupled to the services theyre meant to monitor. This changes the
chart directory structure to include a monitoring/ directory to
host any monitoring requirements (not just prometheus exporters)
Change-Id: Ic5f46a7e9a2a773a837ff399b8730782cf89f1de
The DNS-Helper chart was created to support early development of
openstack helm, it is no longer required, and causes more confusion
than help these days. This PS removes the chart.
Change-Id: I5decc612089ca205fa097f881b36045f247f057c
This PS updates the Keystone user job template to use default
values when not specified via the input dict.
Change-Id: I937e720ab3ab4ba70c33ea981d7693ebfd01c894
