Feature #12143
cephfs-data-scan: Tool for orchestrating multiple workers
Status:
New
Priority:
Normal
Assignee:
-
Category:
fsck/damage handling
Target version:
-
% Done:
0%
Source:
other
Tags:
Backport:
Reviewed:
Affected Versions:
Component(FS):
Labels (FS):
Pull request ID:
Description
To run at any kind of scale, this tool requires multiple workers executing across multiple clients.
It would be super-neat to have some kind of worker queue system running through RADOS, but probably much simpler to write some python that uses SSH to go and run a bunch of these processes in parallel.
Make sure to hook in progress information, detect and report any workers that failed, provide means to retry a set of failed workers.
History
#1 Updated by John Spray over 8 years ago
- Category changed from 66 to fsck/damage handling