Actions
Bug #51270
openmon: stretch mode clusters do not sanely set default crush rules
Status:
Pending Backport
Priority:
Normal
Assignee:
Category:
Administration/Usability
Target version:
-
% Done:
0%
Source:
Tags:
backport_processed
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
If you do not specify a crush rule when creating a pool, the OSDMonitor picks the default one for you out of the config file.
However, this default is not automatically changed when you enter stretch mode, and it probably won't satisfy stretch peering requirements.
This is a problem especially when you turn on new kinds of daemons which auto-create pools ((RGW, new manager modules, etc) and need them to become available without fussing about the CRUSH rule themselves.
Since we don't actually record the crush rule used when invoking stretch mode, the simplest way to fix this is to just find the most-used CRUSH rule amongst the stretch-mode pools. Do that.
Updated by Greg Farnum almost 3 years ago
- Status changed from In Progress to Fix Under Review
- Backport set to pacific, octopus, nautilus
- Pull request ID set to 41921
Updated by Kefu Chai almost 3 years ago
- Status changed from Fix Under Review to Pending Backport
Updated by Backport Bot almost 3 years ago
- Copied to Backport #51288: nautilus: mon: stretch mode clusters do not sanely set default crush rules added
Updated by Backport Bot almost 3 years ago
- Copied to Backport #51289: octopus: mon: stretch mode clusters do not sanely set default crush rules added
Updated by Backport Bot almost 3 years ago
- Copied to Backport #51290: pacific: mon: stretch mode clusters do not sanely set default crush rules added
Updated by Greg Farnum almost 3 years ago
- Backport changed from pacific, octopus, nautilus to pacific
Accidentally requested backports to Octopus/Nautilus, so nuking those.
Actions