Re: 2 new contributions to pgdg

Dear Christoph,

This was an actual bug only apparent when testing on different cluster
timezone configurations. I have fixed it and drafted a new release 1.5.2.
Thanks for all your help with this. I hope this one passes! It does on my
own machine and the VM I spun up on which I was able to reproduce your test
failure.

Thanks,

On Wed, Nov 28, 2018 at 10:08 AM Jeremy Finzel <finzelj(at)gmail(dot)com> wrote:

> Dear Christoph,
>
> Thank you for the full log. I can't reproduce this issue on several
> machines ... is it possible for you to run this with log_min_messages =
> DEBUG and log_min_duration_statement = 0 and provide output?
>
> Thanks,
>
> On Wed, Nov 28, 2018 at 8:44 AM Christoph Berg <myon(at)debian(dot)org> wrote:
>
>> Re: To Jeremy Finzel 2018-11-28 <20181128144419(dot)GB7747(at)msg(dot)df7cb(dot)de>
>> > Re: Jeremy Finzel 2018-11-27 <
>> CAMa1XUgogtFxnYC5S+t6rO-zP2fNAjO23k8S8v__6Zgtj3Jivg(at)mail(dot)gmail(dot)com>
>> > the build is still not passing the regression tests. The good news is
>> > that the diff is the same on 9.5 and 11:
>>
>> Full log:
>>
>>
>> https://pgdgbuild.dus.dg-i.net/job/pg-fact-loader-binaries/5/architecture=amd64,distribution=sid/console
>>
>> > 15:13:28 Dropping cluster 9.5/regress ...
>> > 15:13:29 **** regression.diffs ****
>> > 15:13:29 ---
>> /tmp/autopkgtest.RWVqQv/tree/expected/07_launch_worker.out 2018-11-27
>> 15:26:32.000000000 +0000
>> > 15:13:29 +++ /tmp/autopkgtest.RWVqQv/tree/results/07_launch_worker.out
>> 2018-11-28 14:13:05.269366205 +0000
>> > 15:13:29 @@ -35,7 +35,7 @@
>> > 15:13:29 7 | 0001234567 | 35 | |
>> 0 |
>> > 15:13:29 8 | 0001234568 | 35 | |
>> 0 |
>> > 15:13:29 9 | 0001234569 | 35 | |
>> 0 |
>> > 15:13:29 - 10 | 0001234577 | 35 | |
>> 0 |
>> > 15:13:29 + 10 | 0001234560 | 35 | |
>> 0 |
>> > 15:13:29 (10 rows)
>> > 15:13:29
>> > 15:13:29 UPDATE fact_loader.fact_tables SET force_worker_priority =
>> FALSE WHERE fact_table_relid = 'test_fact.customers_fact'::REGCLASS;
>> > 15:13:29
>>
>> Christoph
>>
>