Project

General

Profile

Actions
Copy link

Bug #199

closed

OSD crash when rebalancing data

Added by Wido den Hollander almost 14 years ago. Updated over 13 years ago.

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

0%

Source:
Tags:
Backport:
Regression:
Severity:
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

Today i wanted to expand my number of OSD's from 5 to 7.

After i loaded my new crushmap the rebalancing of data started, but while that happened 3 of the 7 OSD's went down.

Attached you will find the traces if the 3 OSD's, my ceph.conf and the old and new OSD map.

The logs can be found at (to large to upload) at:
OSD4: http://zooi.widodh.nl/ceph/ceph08.1645.gz
OSD5: http://zooi.widodh.nl/ceph/ceph05.2526.gz
OSD6: http://zooi.widodh.nl/ceph/ceph05.2570.gz

Restarting the OSD's had no effect, they crashed again.

I followed this wiki page: http://ceph.newdream.net/wiki/OSD_cluster_expansion/contraction

Files

Download all files
ceph.conf (812 Bytes) ceph.conf Wido den Hollander, 06/14/2010 06:38 AM
crush_new.txt (1 KB) crush_new.txt Wido den Hollander, 06/14/2010 06:38 AM
crush_new.txt (1 KB) crush_new.txt Wido den Hollander, 06/14/2010 06:38 AM
osd4_strace.txt (1.66 KB) osd4_strace.txt Wido den Hollander, 06/14/2010 06:38 AM
osd5_strace.txt (1.84 KB) osd5_strace.txt Wido den Hollander, 06/14/2010 06:38 AM
osd6_strace.txt (1.8 KB) osd6_strace.txt Wido den Hollander, 06/14/2010 06:38 AM
Actions
Copy link
 #1

Updated by Sage Weil almost 14 years ago

  • Status changed from New to Closed

I can't verify because the stack traces have no symbols, but I'm going to guess this is the same crush map update issue I fixed yesterday and close this out. If you see it again, please reopen!

Actions
Copy link

Also available in: Atom PDF