On Sat, May 25, 2013 at 8:51 PM, Kristian Nielsen <knielsen@knielsen-hq.org> wrote:
yet). But from time to time we see failures of rpl.rpl_gtid_crash test with the output like this:
Hm, strange, it does not fail in our Buildbot, which runs hundreds of tests on different hosts...
Well, either our build environment differs, or we induced something with our patches...
that. It's possible that there's something wrong with us applying GTID patch on top of 10.0.1 and maybe I'm missing something crucial (we
off the top of your head what changes were done since 10.0.1 that are required for crash recovery to work properly?
I am wondering if you mean 10.0.2 here? (Since GTID was not included in 10.0.1, only 10.0.2).
No, I meant 10.0.1 -- we took it and applied all that you did in 10.0-mdev26 and then in 10.0-base, but only GTID-related patches and patches fixing flakiness of tests...
There was a change to rpl.rpl_gtid_crash.test to use --append_file instead of --remove_file/--write_file. Though these failures look like something else...
We already have the --append_file change.
I can't think of anything else right now, but if you send the files then there is a good chance I can say more.
OK. I'll try to catch log files next time this happens. Pavel