From 98b3984ec7048ce48dce9905485093be28993928 Mon Sep 17 00:00:00 2001
From: Pete Birley <pete@port.direct>
Date: Thu, 16 Aug 2018 11:11:09 -0500
Subject: [PATCH] Libvirt: Warn if hugetlb cgroup controller is enabled

This PS adds a warning if the hugetlb cgroup controller is enabled
as it prevents restarts of the libvirt pod without shutting down
kvm vms. The controller can be disabled by passing
`cgroup_disable=hugetlb` to the kernel params.

Change-Id: Idd26fd3c19191f3c3e0aeb6fa3efa9a4ec48ee54
Signed-off-by: Pete Birley <pete@port.direct>
---
 libvirt/templates/bin/_libvirt.sh.tpl | 26 ++++++++++++++++----------
 1 file changed, 16 insertions(+), 10 deletions(-)

diff --git a/libvirt/templates/bin/_libvirt.sh.tpl b/libvirt/templates/bin/_libvirt.sh.tpl
index 5a5e38dd..7d2c5b61 100644
--- a/libvirt/templates/bin/_libvirt.sh.tpl
+++ b/libvirt/templates/bin/_libvirt.sh.tpl
@@ -54,17 +54,23 @@ if [ 0"$hp_count" -gt 0 ]; then
   # Kubernetes 1.10.x introduced cgroup changes that caused the container's
   # hugepage byte limit quota to zero out. This workaround sets that pod limit
   # back to the total number of hugepage bytes available to the baremetal host.
+  if [ -d /sys/fs/cgroup/hugetlb ]; then
+    # NOTE(portdirect): Kubelet will always create pod specific cgroups for
+    # hugetables so if the hugetlb cgroup is enabled, when k8s removes the pod
+    # it will also remove the hugetlb cgroup for the pod, taking any qemu
+    # processes with it.
+    echo "WARN: As the hugetlb cgroup is enabled, it will not be possible to restart the libvirt pod via k8s, without killing VMs."
+    for limit in $(ls /sys/fs/cgroup/hugetlb/kubepods/hugetlb.*.limit_in_bytes); do
+      target="/sys/fs/cgroup/hugetlb/$(dirname $(awk -F: '($2~/hugetlb/){print $3}' /proc/self/cgroup))/$(basename $limit)"
+      # Ensure the write target for the hugepage limit for the pod exists
+      if [ ! -f "$target" ]; then
+        echo "ERROR: Could not find write target for hugepage limit: $target"
+      fi
 
-  for limit in $(ls /sys/fs/cgroup/hugetlb/kubepods/hugetlb.*.limit_in_bytes); do
-    target="/sys/fs/cgroup/hugetlb/$(dirname $(awk -F: '($2~/hugetlb/){print $3}' /proc/self/cgroup))/$(basename $limit)"
-    # Ensure the write target for the hugepage limit for the pod exists
-    if [ ! -f "$target" ]; then
-      echo "ERROR: Could not find write target for hugepage limit: $target"
-    fi
-
-    # Write hugetable limit for pod
-    echo "$(cat $limit)" > "$target"
-  done
+      # Write hugetable limit for pod
+      echo "$(cat $limit)" > "$target"
+    done
+  fi
 
   # Determine OS default hugepage size to use for the hugepage write test
   default_hp_kb="$(cat /proc/meminfo | grep Hugepagesize | tr -cd '[:digit:]')"