[LU-897] recovery_mds_scale.sh does not use the test framework correctly. Created: 05/Dec/11  Updated: 27/Mar/12  Resolved: 19/Mar/12

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Major
Reporter: Chris Gearing (Inactive) Assignee: Chris Gearing (Inactive)
Resolution: Duplicate Votes: 0
Labels: None

Issue Links:
Duplicate
duplicates LU-734 improve recovery-*-scale tests to put... Resolved
Severity: 3
Rank (Obsolete): 6509

 Description   

recovery_mds_scale.sh does not use the test framework correctly. It should call run_test for example to start the test and return errors appropriately to the framework.

Without these compliance the test results cannot be correctly logged.

In addition recovery-mds-scale writes some log files to /tmp when they should be written to the log files and so recovery-mds-scale writes the client-job log files to /tmp when they should be written to the log files/, this needs to be corrected so that the logs are recorded as part of the test collateral and sent to Maloo for storage.



 Comments   
Comment by Jian Yu [ 19/Mar/12 ]

Hi Chris,

All of these works have been done in LU-734 and the patches have been landed on b2_2 and master branches.

Comment by Jian Yu [ 19/Mar/12 ]

Hello Chris,

From the latest failover test sessions on master and b2_2 branches, I saw the following issues in node-provisioning_1 stage:

16:20:24:ssh -q fat-intel-3vm1 "rsh -l root fat-intel-3vm7 'PATH=\$PATH:/sbin:/usr/sbin;cd /usr/lib64/lustre/tests; patch -l -N -t -b -p3 </home/autotest/lab/tools/fixups/LU-897' " 
16:20:25:patching file recovery-double-scale.sh
Hunk #1 FAILED at 10.
Hunk #2 FAILED at 95.
Hunk #3 FAILED at 112.
Hunk #4 FAILED at 163.
Hunk #5 FAILED at 250.
Hunk #6 FAILED at 276.
Hunk #7 FAILED at 330.
7 out of 7 hunks FAILED -- saving rejects to file recovery-double-scale.sh.rej
patching file recovery-mds-scale.sh
Hunk #1 FAILED at 7.
Hunk #2 FAILED at 63.
Hunk #3 FAILED at 168.
Hunk #4 FAILED at 191.
Hunk #5 FAILED at 218.
Hunk #6 FAILED at 239.
Hunk #7 FAILED at 270.
Hunk #8 FAILED at 284.
8 out of 8 hunks FAILED -- saving rejects to file recovery-mds-scale.sh.rej
patching file recovery-random-scale.sh
Hunk #1 FAILED at 23.
Hunk #2 FAILED at 175.
Hunk #3 FAILED at 198.
Hunk #4 FAILED at 238.
Hunk #5 FAILED at 248.
Hunk #6 FAILED at 271.
Hunk #7 FAILED at 294.
Hunk #8 FAILED at 315.
8 out of 8 hunks FAILED -- saving rejects to file recovery-random-scale.sh.rej
patching file run_IOR.sh
Hunk #1 FAILED at 1.
Hunk #2 succeeded at 32 (offset 3 lines).
Hunk #3 succeeded at 61 (offset 3 lines).
1 out of 3 hunks FAILED -- saving rejects to file run_IOR.sh.rej
patching file run_dbench.sh
Hunk #1 FAILED at 1.
Hunk #2 succeeded at 54 (offset 3 lines).
1 out of 2 hunks FAILED -- saving rejects to file run_dbench.sh.rej
patching file run_dd.sh
Hunk #1 FAILED at 1.
Hunk #2 succeeded at 55 (offset 3 lines).
1 out of 2 hunks FAILED -- saving rejects to file run_dd.sh.rej
patching file run_iozone.sh
Hunk #1 FAILED at 1.
Hunk #2 succeeded at 57 (offset 3 lines).
1 out of 2 hunks FAILED -- saving rejects to file run_iozone.sh.rej
patching file run_tar.sh
Hunk #1 FAILED at 1.
Hunk #2 succeeded at 62 (offset 3 lines).
1 out of 2 hunks FAILED -- saving rejects to file run_tar.sh.rej
patching file test-framework.sh
Hunk #1 FAILED at 1095.
Hunk #2 succeeded at 1129 (offset 10 lines).
Hunk #3 succeeded at 1139 (offset 10 lines).
Hunk #4 succeeded at 1175 (offset 10 lines).
Hunk #5 succeeded at 1213 (offset 10 lines).
1 out of 5 hunks FAILED -- saving rejects to file test-framework.sh.rej
16:20:28:ssh -q fat-intel-3vm1 "rsh -l root fat-intel-3vm7 'PATH=\$PATH:/sbin:/usr/sbin;cd /usr/lib64/lustre/tests; patch -l -N -t -b -p3 </home/autotest/lab/tools/fixups/LU-233' " 
16:20:28:patching file recovery-double-scale.sh
Hunk #1 FAILED at 244.
1 out of 1 hunk FAILED -- saving rejects to file recovery-double-scale.sh.rej
patching file recovery-mds-scale.sh
Hunk #1 FAILED at 162.
1 out of 1 hunk FAILED -- saving rejects to file recovery-mds-scale.sh.rej
patching file recovery-random-scale.sh
Hunk #1 FAILED at 169.
1 out of 1 hunk FAILED -- saving rejects to file recovery-random-scale.sh.rej
patching file test-framework.sh
Reversed (or previously applied) patch detected!  Skipping patch.
1 out of 1 hunk ignored -- saving rejects to file test-framework.sh.rej

https://maloo.whamcloud.com/test_sessions/e27e7fd4-715a-11e1-a89e-5254004bbbd3

The fixups covered by /home/autotest/lab/tools/fixups/

{LU-897,LU-233}

have been landed on b2_2 and master branches by the patch for LU-734. Although the patching operations were rejected, I saw some changes were still made into run_*.sh and test-framework.sh. Could you please disable those patching operations in autotest system? Thanks a lot.

Comment by Chris Gearing (Inactive) [ 20/Mar/12 ]

Fixups for LU-897 and LU-233 have been have been removed.

Comment by Jian Yu [ 20/Mar/12 ]

Fixups for LU-897 and LU-233 have been have been removed.

Thank you very much, Chris!

Generated at Sat Feb 10 01:11:28 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.