Project

General

Profile

Bug #16

mds restart vs dbench

Added by Sage Weil about 9 years ago. Updated over 2 years ago.

Status:
Resolved
Priority:
High
Assignee:
-
Category:
-
Target version:
-
Start date:
04/09/2010
Due date:
% Done:

0%

Source:
Tags:
Backport:
Regression:
Severity:
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
Labels (FS):
Pull request ID:

Description

dbench 1, restart mds (may take a few times), dbench will error out.

History

#1 Updated by Sage Weil about 9 years ago

  • Category set to 1

#2 Updated by Sage Weil almost 9 years ago

  • Priority changed from Normal to High

#3 Updated by Sage Weil almost 9 years ago

  • Status changed from New to Can't reproduce

this has apparently been fixed since april.. can't reproduce it now.

#4 Updated by Sage Weil almost 9 years ago

  • Status changed from Can't reproduce to In Progress

whoops, no, hit it.

CEPH_NUM_MDS=1 ./vstart.sh -d -n -x
while true ; do sleep 90 ; ./init-ceph restart mds ; done

and run 'dbench 1' under uml, eventually errors out with

   1    103369     0.95 MB/sec  execute 403 sec   
   1    103438     0.95 MB/sec  execute 404 sec   
(103497) unlink ./clients/client0/~dmtmp/WORDPRO/NEWS1_1B.PRN failed (No such file or directory) - expected NT_STATUS_OK
ERROR: child 0 failed
Child failed with status 1

#5 Updated by Sage Weil almost 9 years ago

aha, when teh rename request is replayed, the target(?) filename changes:

fatty:src 04:10 PM $ grep -- -- out/mds.a.4 | grep client4101:7166
10.07.09_15:55:21.198996 7f2e96d52710 -- 10.0.1.252:6802/25899 <== client4101 10.0.1.200:0/3430181140 3 ==== client_request(client4101:7166 rename #1000000000c/SALES1.XLS #1000000000c/SALES1.XLS RETRY=1 REPLAY) ==== 232+0+0 (1198761746 0 0) 0x24370e0
fatty:src 04:12 PM $ grep -- -- out/mds.a.5 | grep client4101:7166
10.07.09_15:55:00.697346 7fb475086710 -- 10.0.1.252:6802/25770 <== client4101 10.0.1.200:0/3430181140 3533 ==== client_request(client4101:7166 rename #1000000000c/SALES1.XLS #1000000000c/BEED0000) ==== 228+0+0 (3226454752 0 0) 0x7fb468244be0

#6 Updated by Sage Weil almost 9 years ago

  • Status changed from In Progress to Testing

hopefully fixed by commit:696bab43d66532768a9f85f71a86ddef2bff527f 'ceph: generate request paths only once'?? needs more testing.

#7 Updated by Sage Weil almost 9 years ago

  • Status changed from Testing to Resolved

also commit:8ea9cd5956c7c158fe93076930701927e13bfa03

#8 Updated by John Spray over 2 years ago

  • Project changed from Ceph to fs
  • Category deleted (1)

Bulk updating project=ceph category=mds bugs so that I can remove the MDS category from the Ceph project to avoid confusion.

Also available in: Atom PDF