[LU-746] obdfilter-survey FAIL: test_1b ost4: hndls expected > 8, have 2 Created: 10/Oct/11 Updated: 16/Aug/16 Resolved: 16/Aug/16 |
|
| Status: | Closed |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | Lustre 1.8.8, Lustre 1.8.7 |
| Fix Version/s: | Lustre 2.2.0, Lustre 1.8.8 |
| Type: | Bug | Priority: | Minor |
| Reporter: | Jian Yu | Assignee: | Lai Siyao |
| Resolution: | Won't Fix | Votes: | 0 |
| Labels: | None | ||
| Environment: |
Lustre Branch: b1_8 |
||
| Issue Links: |
|
||||||||
| Severity: | 3 | ||||||||
| Rank (Obsolete): | 4749 | ||||||||
| Description |
|
obdfilter-survey test 1b failed as follows: Sat Oct 8 23:12:57 PDT 2011 Obdfilter-survey for case=disk from client-26vm1.lab.whamcloud.com ost 7 sz 7340032K rsz 1024K obj 7 thr 28 write 10.21 [ 0.00, 6.99] rewrite 16.39 [ 0.00, 22.96] read 28.63 [ 0.00, 25.96] starting run for test: write rsz: 1024 threads: 4 objects: 1 starting run for test: rewrite rsz: 1024 threads: 4 objects: 1 starting run for test: read rsz: 1024 threads: 4 objects: 1 R 1024 0 5000 0 0 137 38 2 3 R 1025 0 5000 0 0 123 2 16 17 R 1008 0 5000 0 0 198 11 2 3 R 1009 0 5000 0 0 172 2 16 17 R 1009 0 5000 0 0 2117 15 2 3 R 1010 0 5000 0 0 259 2 16 17 R 1010 0 5000 0 0 399 2 2 3 R 1011 0 5000 0 0 175 2 16 17 obdfilter-survey test_1b: @@@@@@ FAIL: ost4: hndls expected > 8, have 2 Dumping lctl log to /logdir/test_logs/2011-10-08/lustre-b1_8-el5-x86_64__134__-7f5c39000658/obdfilter-survey.test_1b.*.1318142207.log R 1008 0 5000 0 0 1845 3 2 3 R 1009 0 5000 0 0 233 2 16 17 obdfilter-survey test_1b: @@@@@@ FAIL: ost5: hndls expected > 8, have 3 Dumping lctl log to /logdir/test_logs/2011-10-08/lustre-b1_8-el5-x86_64__134__-7f5c39000658/obdfilter-survey.test_1b.*.1318142217.log R 1004 0 5000 0 0 1031 1 2 3 R 1005 0 5000 0 0 180 2 16 17 obdfilter-survey test_1b: @@@@@@ FAIL: ost6: hndls expected > 8, have 1 Dumping lctl log to /logdir/test_logs/2011-10-08/lustre-b1_8-el5-x86_64__134__-7f5c39000658/obdfilter-survey.test_1b.*.1318142222.log R 998 0 5000 0 0 2262 5 2 3 R 999 0 5000 0 0 101 2 16 17 obdfilter-survey test_1b: @@@@@@ FAIL: ost7: hndls expected > 8, have 5 Dumping lctl log to /logdir/test_logs/2011-10-08/lustre-b1_8-el5-x86_64__134__-7f5c39000658/obdfilter-survey.test_1b.*.1318142227.log obdfilter-survey test_1b: @@@@@@ FAIL: test_1b failed with 8 Dumping lctl log to /logdir/test_logs/2011-10-08/lustre-b1_8-el5-x86_64__134__-7f5c39000658/obdfilter-survey.test_1b.*.1318142232.log Resetting fail_loc on all nodes...done. FAIL (1499s) Maloo report: https://maloo.whamcloud.com/test_sets/d09bdfde-f24b-11e0-908b-52540025f9af |
| Comments |
| Comment by Peter Jones [ 13/Oct/11 ] |
|
Lai Could you please look into this possible 1.8.7 blocker as your to priority? Thanks Peter |
| Comment by Sarah Liu [ 14/Oct/11 ] |
|
passed on manual test:https://maloo.whamcloud.com/test_sets/b8cb27c6-f6a0-11e0-a451-52540025f9af |
| Comment by Lai Siyao [ 17/Oct/11 ] |
|
The last brw write ends at time 1318141924: 00002000:00100000:0:1318141924.703756:0:24425:0:(filter.c:198:filter_finish_transno()) wrote trans 21474842626 for client ECHO_UUID at #3: err = 0 The last object destroy ends at time 1318142186: 00002000:00020000:0:1318142186.737229:0:26178:0:(filter.c:1557:filter_destroy_internal()) destroying objid 3 ino 56 nlink 1 count 1 00002000:00100000:0:1318142186.740251:0:26178:0:(filter.c:198:filter_finish_transno()) wrote trans 12884908035 for client ECHO_UUID at #3: err = 0 jbd hndls checked at time 1318142207: 00000001:02000400:0:1318142207.957982:0:27184:0:(debug.c:535:libcfs_debug_mark_buffer()) DEBUG MARKER: obdfilter-survey test_1b: @@@@@@ FAIL: ost4: hndls expected > 8, have 2 obdfilter-survey.sh '1b' test checked jbd history data of the past 15 seconds, but unfortunately in this round of test, system is idle at that period [1318142192, 1318142207], so the stats collected is not the one wanted. I guess this is caused by slow VMs (and this can't be reproduced on physical machines manually by Sarah, see above). To fix this, we may need to make `check_jbd_values_facets` execute right after brw. |
| Comment by Lai Siyao [ 18/Oct/11 ] |
|
review is on http://review.whamcloud.com/#change,1534 |
| Comment by Peter Jones [ 03/Nov/11 ] |
|
Lai Would it be useful to port this patch to master? Peter |
| Comment by Lai Siyao [ 03/Nov/11 ] |
|
Peter, yes, I'll port it to master later. |
| Comment by Minh Diep [ 06/Feb/12 ] |
|
Can we get this patch to land? this bug is preventing lustre-review to have a good run |
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 07/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Build Master (Inactive) [ 08/Feb/12 ] |
|
Integrated in Result = SUCCESS
|
| Comment by Peter Jones [ 08/Feb/12 ] |
|
Landed for 2.2 |
| Comment by Build Master (Inactive) [ 17/Feb/12 ] |
|
Integrated in Result = FAILURE
|
| Comment by Build Master (Inactive) [ 17/Feb/12 ] |
|
Integrated in Result = FAILURE
|
| Comment by Build Master (Inactive) [ 17/Feb/12 ] |
|
Integrated in Result = ABORTED
|
| Comment by Jian Yu [ 16/May/12 ] |
|
Lustre Tag: v1_8_8_WC1_RC1 The issue still occurred: https://maloo.whamcloud.com/test_sets/ee85f612-9d5c-11e1-8587-52540035b04c |
| Comment by Jian Yu [ 14/Jan/13 ] |
|
Lustre Branch: b1_8 The obdfilter-survey test 2b failed as follows: obd survey finished in 957 seconds
CMD: client-28vm4 dv=\$(lctl get_param -n *.lustre-OST0000.mntdev);
if foo=\$(lvdisplay -c \$dv 2>/dev/null); then
echo dm-\${foo##*:};
else
echo \$(basename \$dv);
fi;
CMD: client-28vm4 cat /proc/fs/jbd*/dm-0*/history
R 2298 0 5001 0 0 1709 12 2 3
CMD: client-28vm4 dv=\$(lctl get_param -n *.lustre-OST0001.mntdev);
if foo=\$(lvdisplay -c \$dv 2>/dev/null); then
echo dm-\${foo##*:};
else
echo \$(basename \$dv);
fi;
CMD: client-28vm4 cat /proc/fs/jbd*/dm-1*/history
R 2327 0 5000 0 0 2071 19 2 3
CMD: client-28vm4 dv=\$(lctl get_param -n *.lustre-OST0002.mntdev);
if foo=\$(lvdisplay -c \$dv 2>/dev/null); then
echo dm-\${foo##*:};
else
echo \$(basename \$dv);
fi;
CMD: client-28vm4 cat /proc/fs/jbd*/dm-2*/history
R 2333 0 5004 0 0 2013 20 2 3
CMD: client-28vm4 dv=\$(lctl get_param -n *.lustre-OST0003.mntdev);
if foo=\$(lvdisplay -c \$dv 2>/dev/null); then
echo dm-\${foo##*:};
else
echo \$(basename \$dv);
fi;
CMD: client-28vm4 cat /proc/fs/jbd*/dm-3*/history
R 2331 0 5000 0 0 1119 15 2 3
CMD: client-28vm4 dv=\$(lctl get_param -n *.lustre-OST0004.mntdev);
if foo=\$(lvdisplay -c \$dv 2>/dev/null); then
echo dm-\${foo##*:};
else
echo \$(basename \$dv);
fi;
CMD: client-28vm4 cat /proc/fs/jbd*/dm-4*/history
R 2326 1 4999 0 0 2035 20 2 3
obdfilter-survey test_2b: @@@@@@ FAIL: ost5: run expected 5000, have 4999
Maloo report: https://maloo.whamcloud.com/test_sets/83456338-5daa-11e2-8199-52540035b04c |
| Comment by James A Simmons [ 16/Aug/16 ] |
|
Old ticket for unsupported version |