15119 2023-05-11 15:27:51 +0000 ptests randomly have no results 2023-06-22 08:15:48 +0000 1 1 10 QA/Testing Package Testing (ptest) ptest unspecified x86 Multiple RESOLVED FIXED AB-INT Medium+ normal 4.3 M3 1 alexandre.belloni ross.burton randy.macleod tgamblin --- --- No (bug/feature does not impact docs) oldest_to_newest 95522 0 alexandre.belloni 2023-05-11 15:27:51 +0000 https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/4829/steps/12/logs/stdio AssertionError: ptests which had no test results: ['gdbm'] 95557 1 ross.burton 2023-05-18 16:50:31 +0000 Happened once, and according to the log the test suite did nothing: START: ptest-runner 2023-05-06T18:34 BEGIN: /usr/lib/gdbm/ptest DURATION: 3 END: /usr/lib/gdbm/ptest 2023-05-06T18:34 STOP: ptest-runner TOTAL: 1 FAIL: 0 No errors, no output, nothing. I've got a small change to remove the sed which might make any errors clearer. Leaving open and we'll see if it happens again. 95561 2 alexandre.belloni 2023-05-18 20:05:48 +0000 libnl occurrence: https://autobuilder.yoctoproject.org/typhoon/#/builders/81/builds/5061/steps/12/logs/stdio cpio occurrence: https://autobuilder.yoctoproject.org/typhoon/#/builders/81/builds/5010/steps/12/logs/stdio 95563 3 ross.burton 2023-05-18 20:58:53 +0000 Ok that's weird and not the problem I thought it was. 95565 4 alexandre.belloni 2023-05-18 21:23:21 +0000 An other older cpio occurrence: https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/4820/steps/12/logs/stdio 95779 5 tgamblin 2023-06-21 19:39:17 +0000 I ran the gdbm ptests 3000 times using core-image-ptest-gdbm and didn't encounter that failure even once. My system wasn't under heavy load and the qemuparams were pretty minimal ("runqemu kvm nographic"). Makes me think that this is either another context-dependent failure or the issue has been fixed by change merged since the last reports. 95783 6 ross.burton 2023-06-22 08:15:48 +0000 https://git.yoctoproject.org/poky/commit/?h=master-next&id=55b0208c333e01778b25a1bdcb642ea6608e8f14 _should_ be the fix and suggests that it is a race. I'm going to provisionally close this and we can reopen if it happens again.