Bug #50308
mon: stretch state is inconsistently-maintained on peons, preventing proper failover in repeated netsplits
% Done:
0%
Source:
Tags:
Backport:
pacific
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(RADOS):
Monitor
Pull request ID:
Crash signature (v1):
Crash signature (v2):
Description
There's an issue where some in-memory monitor stretch state changes are only applied on the leader monitor, and so peons maintain a fragmented stretch state -- even if they happen to become the leader later on. This prevents them from changing to degraded stretch mode when they should and blocks cluster progress.
Related issues
History
#1 Updated by Greg Farnum almost 3 years ago
- Status changed from New to In Progress
- Backport set to pacific
#2 Updated by Greg Farnum almost 3 years ago
- Status changed from In Progress to Pending Backport
- Pull request ID set to 40835
#3 Updated by Backport Bot almost 3 years ago
- Copied to Backport #50344: pacific: mon: stretch state is inconsistently-maintained on peons, preventing proper failover in repeated netsplits added
#4 Updated by Greg Farnum almost 3 years ago
- Status changed from Pending Backport to Resolved