Actions
Bug #2393
closedobjecter: dropping messages (old connection being used)
Status:
Duplicate
Priority:
Urgent
Assignee:
-
Category:
-
Target version:
-
% Done:
0%
Source:
Community (dev)
Tags:
Backport:
Regression:
Severity:
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):
Description
2012-05-10 02:01:02.593957 7f9cda1b0700 0 -- [2607:f298:4:2243::5795]:0/1019484 send_message dropped message ping v1 because of no pipe on con 0x7f9cc459e660 2012-05-10 02:01:02.593970 7f9cda1b0700 1 -- [2607:f298:4:2243::5795]:0/1019484 --> [2607:f298:4:2243::7066]:6807/2644 -- ping v1 -- ?+0 0x7f9cbc3d2f40 con 0x7f9cc4144db0 2012-05-10 02:01:02.668326 7f9cdc2b5700 1 -- [2607:f298:4:2243::5795]:0/1019484 <== osd.290 [2607:f298:4:2243::7066]:6807/2644 46191 ==== watch-notify(c=1 v=1 i=740219 opcode=1) v1 ==== 543+0+0 (3452665825 0 0) 0x7f9cb43ac330 con 0x7f9cc4144db0 ... 2012-05-10 02:01:04.271134 7f9c9cff9700 0 -- [2607:f298:4:2243::5795]:0/1019484 send_message dropped message osd_op(client.92928.0:753509 .dir.93029.10938 [call rgw.bucket_prepare_op] 5.6182dfe9) v4 because of no pipe on con 0x7f9cc459e660
One point to note is that we do have an open connection to that osd (where we got the watch-notify msg from). However, we still try to send requests over the broken connection.
Actions