Corosync somtimes hangs with 100% cpu load in epoll_wait for totem cpg
Bug #1677549 reported by
Vladimir Kuklin
This bug affects 1 person
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| Fuel for OpenStack |
In Progress
|
High
|
Dmitry Teselkin | ||
Bug Description
We have a set of cases when corosync hangs and pacemaker gets broken. Corosync hangs in epoll_wait endless loop eating 100% cpu while completely hanging the cluster. This reproduces not so often, but looks like an issue with totempg code. We need to build corosync 2.3.6 which has the fix enabled and ship it with newton repos.
Revision history for this message
| Fuel Devops McRobotson (fuel-devops-robot) wrote Related fix proposed to fuel-infra/jeepyb-config (master) | #1 |
| Changed in fuel: | |
| status: | Confirmed → In Progress |
Revision history for this message
| Fuel Devops McRobotson (fuel-devops-robot) wrote Related fix proposed to fuel-infra/zuul-layouts (master) | #2 |
Revision history for this message
| Fuel Devops McRobotson (fuel-devops-robot) wrote Related fix merged to fuel-infra/zuul-layouts (master) | #3 |
Revision history for this message
| Fuel Devops McRobotson (fuel-devops-robot) wrote Related fix proposed to packages/xenial/corosync (master) | #4 |
Revision history for this message
| Fuel Devops McRobotson (fuel-devops-robot) wrote Related fix proposed to packages/xenial/corosync (10.0/newton) | #5 |
To post a comment you must log in.
Related fix proposed to branch: master
Change author: Dmitry Teselkin <email address hidden>
Review: https:/