haproxy restart loop due to too short timeout
Bug #1464650 reported by
Eric Jacobs
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fuel for OpenStack |
Fix Released
|
Critical
|
Sergii Golovatiuk | ||
6.1.x |
Fix Committed
|
Critical
|
Eric Jacobs | ||
7.0.x |
Fix Released
|
Critical
|
Eric Jacobs |
Bug Description
The timeout for p_haproxy-start is 30 seconds. In situations where haproxy takes more than 30 seconds to start, the following loop occurs:
1. haproxy starts to come up; binds vips and ports
2. traffic starts to flow on the vips
3. haproxy gets killed; vips go down
4. goto 1
This can cause deployment to fail during secondary controller setup, and causes major difficult to diagnose problems for the entire environment.
Recommended fix: increase timeout to 45 or 60 seconds.
tags: | added: 6.1rc2 |
Changed in fuel: | |
status: | Triaged → In Progress |
assignee: | Eric Jacobs (ejacobs) → Sergii Golovatiuk (sgolovatiuk) |
To post a comment you must log in.
Fix proposed to branch: master /review. openstack. org/191038
Review: https:/