Re: pgsql: Add hooks for session start and session end, take two

On 10/1/19 10:28 AM, Michael Paquier wrote:
> On Tue, Oct 01, 2019 at 04:13:20PM +0900, Michael Paquier wrote:
>> On Tue, Oct 01, 2019 at 02:22:45PM +0900, Michael Paquier wrote:
>>> Actually, it makes little sense to allow parallel workers to log their
>>> activity in the module.  So if there are no objections, I would like
>>> to fix that by checking after IsParallelWorker() in the hooks of the
>>> module.
>> Fixed with 002962dc, and crake has just turned back to green.
> Even with that, there is still one failure with Msys and fairywren:
> ALTER DATABASE
> ============== running regression test queries        ==============
> test test_session_hooks           ... FAILED (test process exited with
> exit code 2)      493 ms
> ============== shutting down postmaster               ==============
>
> ======================
>  1 of 1 tests failed.
>  ======================
>
> The differences that caused some tests to fail can be viewed in the
> file
> "C:/tools/msys64/home/Administrator/bf/root/HEAD/pgsql.build/src/test/modules/test_session_hooks/regression.diffs".
> A copy of the test summary that you see above is saved in the file
> "C:/tools/msys64/home/Administrator/bf/root/HEAD/pgsql.build/src/test/modules/test_session_hooks/regression.out".
>
> Unfortunately it is a bit hard to grab what the problem actually is.
> Andrew, could it be possible to get more information from this animal?


I'll fix up the logging. Meanwhile, the log is showing:


\c :prevdb regress_sess_hook_usr1
\connect: FATAL:  SSPI authentication failed for user
"regress_sess_hook_usr1"


That's not surprising given the hba and ident file contents.


cheers


andrew


-- 
Andrew Dunstan                https://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services