mirror/runc
1
0
Fork 0
mirror of https://github.com/opencontainers/runc.git synced 2026-07-11 06:03:57 +08:00
Code Issues Packages Projects Releases Wiki Activity

cgroupv1 freezer: thaw to increase freeze chances

Browse Source 
It appears that briefly thawing the cgroup while freezing
greatly increases its chances to freeze successfully.

The test case I used is doing runc exec in a look parallel with runc
pause/resume in another loop, and the failure to freeze rate reduced
from 40 to 0 per minute (tested inside a VM using a busybox container
running sleep 1h, doing about 1500 pause/resumes and 650 execs per
minute), with max retries being 150 (of 1000).

This is still a game of chances, so failures are possible.

Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
This commit is contained in:
Kir Kolyshkin
2021-02-04 15:06:11 -08:00
parent 19437f3a55
commit d1007b08a3
1 changed files with 13 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
libcontainer/cgroups/fs/freezer.go
+13 -3
View File
@@ -12,6 +12,7 @@ import (
"github.com/opencontainers/runc/libcontainer/cgroups"
"github.com/opencontainers/runc/libcontainer/cgroups/fscommon"
"github.com/opencontainers/runc/libcontainer/configs"
"github.com/sirupsen/logrus"
"golang.org/x/sys/unix"
)
@@ -38,13 +39,19 @@ func (s *FreezerGroup) Set(path string, cgroup *configs.Cgroup) error {
// (either via fork/clone or by writing new PIDs to
// cgroup.procs).
//
// The number of retries below is chosen to have a decent
// chance to succeed even in the worst case scenario (runc
// pause/unpause with parallel runc exec).
// The numbers below are chosen to have a decent chance to
// succeed even in the worst case scenario (runc pause/unpause
// with parallel runc exec).
//
// Adding any amount of sleep in between retries did not
// increase the chances of successful freeze.
for i := 0; i < 1000; i++ {
if i%50 == 49 {
// Briefly thawing the cgroup also helps.
_ = fscommon.WriteFile(path, "freezer.state", string(configs.Thawed))
time.Sleep(10 * time.Millisecond)
}
if err := fscommon.WriteFile(path, "freezer.state", string(configs.Frozen)); err != nil {
return err
}
@@ -58,6 +65,9 @@ func (s *FreezerGroup) Set(path string, cgroup *configs.Cgroup) error {
case "FREEZING":
continue
case string(configs.Frozen):
if i > 1 {
logrus.Debugf("frozen after %d retries", i)
}
return nil
default:
// should never happen