Bug 11234

Summary: improve test_bmap_long runtime
Product: [QA/Testing] Build Testing Reporter: Jussi Kukkonen <jku>
Component: generalAssignee: Leonardo Sandoval Gonzalez <leonardo.sandoval.gonzalez>
Status: RESOLVED NOTABUG QA Contact:
Severity: enhancement    
Priority: Medium+ CC: benjamin.esquivel, eduard.bartosh, joshuagloe
Version: unspecified   
Target Milestone: 2.4 M1   
Hardware: x86   
OS: Multiple   
Whiteboard:
OS type for building Yocto: --- Type of Regression: ---
Verified: Documentation change: No (bug/feature does not impact docs)
Bug Depends on:    
Bug Blocks: 10874    

Description Jussi Kukkonen 2017-03-24 09:32:47 UTC 
I'm filing this because oeqa.selftest.wic.Wic (test_bmap_long) seems to take too long on the autobuilder . I'm not a subject matter expert so if the runtime is actually reasonable, please close.

The test can take 15 minutes (~6% of the whole selftest time) to finish. Improving this would have clear effect on autobuilder usability.

Example https://autobuilder.yocto.io/builders/nightly-oe-selftest/builds/200/steps/Running%20oe-selftest/logs/stdio
Comment 1 Leonardo Sandoval Gonzalez 2017-04-26 21:55:14 UTC 
This is an interesting unit test: the range of unit test durations is pretty big, from less than a minute to several minutes. What can be causing such a variance? cpu load?

lsandov1@yctb05:~> for i in $(seq 860 880); do curl -s "https://autobuilder.yoctoproject.org/main/builders/nightly-oe-selftest/builds/$i/steps/Running%20oe-selftest/logs/stdio" | grep bmap_long; done
2017-03-28 16:44:08 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (79.092s)
2017-03-31 00:07:45 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (158.015s)
2017-03-31 08:35:30 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (85.140s)
2017-04-01 11:24:07 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (120.411s)
2017-04-03 23:46:48 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (58.434s)
2017-04-06 00:04:13 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (226.932s)
2017-04-09 13:51:27 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (381.894s)
2017-04-09 22:50:29 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (104.383s)
2017-04-10 00:39:18 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (126.834s)
2017-04-12 05:52:43 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (126.495s)
2017-04-13 01:53:57 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (414.603s)
2017-04-14 04:29:27 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (62.221s)
2017-04-14 18:53:03 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (81.530s)
2017-04-15 08:32:48 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (82.254s)
2017-04-19 01:25:09 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (354.505s)
2017-04-20 01:40:49 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (137.453s)
2017-04-20 13:48:56 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (121.345s)
2017-04-21 01:46:22 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (424.435s)
2017-04-21 11:24:24 -   test_bmap_long (oeqa.selftest.wic.Wic) ... OK (71.164s)
Comment 2 Leonardo Sandoval Gonzalez 2017-04-27 15:53:17 UTC 
This unit test (bmap_long) is the first to be executed on the test case, thus triggering bitbake wic-tools. High and low times are related to hot/cold sshare states, so times seen in comment 2 are expected. Resolving it as not-a-bug.