[LMA] HAproxy doesn't recover when Pacemaker restarts the LMA collector after a crash
Bug #1516061 reported by
Simon Pasquier
This bug report is a duplicate of:
Bug #1514893: [LMA] Pacemaker fails to restart the collector when the watchdog check fails.
Edit
Remove
This bug affects 1 person
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| StackLight |
New
|
Undecided
|
guillaume thouvenin | ||
Bug Description
Steps to reproduce
===============
1. Kill -9 the hekad process on the controller owning the management VIP.
2. Verify that HAProxy has detected that the hekad backend is down.
3. Wait for 1 minute.
4. Verify that the hekad process has been restarted by Pacemaker but that HAProxy doesn't detect that it's up.
Expected result
=============
HAProxy should detect that the local hekad process is up again.
| Changed in lma-toolchain: | |
| assignee: | LMA-Toolchain Fuel Plugins (mos-lma-toolchain) → guillaume thouvenin (guillaume-thouvenin) |
Revision history for this message
| Simon Pasquier (simon-pasquier) wrote | #1 |
To post a comment you must log in.
Marking as a duplicate since I believe that this is the same issue as #1514893.