Red Hat Bugzilla – Bug 266361
Rebuild of mysql failed near testing of rpl_rotate_logs
Last modified: 2013-07-02 23:15:05 EDT
Description of problem: Rebuild of mysql 5.0.22 failed on rpl_rotate_logs test. Version-Release number of selected component (if applicable): mysql-5.0.22-2.1.0.1.src.rpm mysql-5.0.22-2.1.src.rpm How reproducible: Rebuild with command : rpmbuild --rebuild --target=x86_64 --rcfile /tmp/tps_rpmrc_a24300: mysql-5.0.22-2.1.0.1.src.rpm (for mysql ver. 5.0.22-2.1.0.1) Steps to Reproduce: 1. rebuild mysql Actual results: always FAILED : (from log) ... rpl_rotate_logs [ fail ] Errors are (from /usr/src/redhat/BUILD/mysql-5.0.22/mysql-test/var/log/mysqltest-time) : mysqltest: At line 28: query 'start slave' failed with wrong errno 1201: 'Could not initialize master info structure; more error messages can be found in the MySQL error log', instead of 1105... (the last lines may be the most important ones) Aborting: rpl_rotate_logs failed in default mode. To continue, re-run with '--force'. ... Expected results: rpl_rotate_logs [ pass ]
Yeah, the mysql regression tests are flaky in some environment-sensitive way that I've never been able to pin down --- I've found a few causes over the years, but there seems at least one left. It always works, or nearly always, on my own workstation, but fails annoyingly often when building in brew. I've sometimes thought that it's just a matter of timeouts being too low for when there's a lot of other activity slowing the machine. (Thinks for a bit ... ) I recall that there's some test that fails with a wrong-errno kind of problem if you try to run the build as root, but I didn't think it was this one. Anyway, unless you can give me a reproducible case, I'm going to have to close this CANTFIX.