Project

General

Profile

Actions
Copy link

Support #12443

closed

The ceph cluster will be stuck after delete a 65535 pgs pool

Added by science luo almost 9 years ago. Updated almost 7 years ago.

Status:
Closed
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:

0%

Tags:
Reviewed:
Affected Versions:
Pull request ID:

Description

There is a strange phenomenon in my ceph cluster.
Everytime I delete a pool with large pg_num like 65535 pgs,my ceph cluster will be stuck.
Any operation like 'ceph -s' or 'ceph osd dump' etc. will be stuck.And my ceph nodes' ceph-mon COMMAND CPU reachs to 100
But if I reboot my ceph node , my ceph cluster gets right.
Is there any limits on pg_num each pool?
Thx for any hints.
PS:By the way, my ceph version was 0.87,3 mons on 3 nodes,and each node has 11 osds.

Actions
Copy link
 #1

Updated by Jason Dillaman almost 9 years ago

  • Project changed from rbd to Ceph
  • Affected Versions deleted (v0.87)
Actions
Copy link
 #2

Updated by science luo over 8 years ago

@Jason Borden Dillaman

Actions
Copy link
 #3

Updated by Greg Farnum almost 7 years ago

  • Tracker changed from Bug to Support
  • Status changed from New to Closed

This seems to just be about throttling PG deletes. I think we are more careful about that now?

Actions
Copy link

Also available in: Atom PDF