Actions
Feature #7347
openPer CRUSH type mon_osd_down_out_interval
Status:
New
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:
0%
Source:
Support
Tags:
Backport:
Reviewed:
Affected Versions:
Pull request ID:
Description
Ability to have different mon_osd_down_out_interval values for different CRUSH types.
For example, you could set:
interval for an osd: 5min
interval for an host: 15min
interval for a chassis: 60min
The rationale is to be able to define the mean time to repair on
different CRUSH types directly in Ceph. If the mean time to repair
threshold is over, data redistribution should be triggered.
No data to display
Actions