Re: logical decoding : exceeded maxAllocatedDescs for .spill files

On Thu, Jan 2, 2020 at 5:44 PM Amit Kapila <[email protected]> wrote:

> On Tue, Dec 24, 2019 at 2:31 PM Amit Khandekar <[email protected]>
> wrote:
> >
> >
> > Ok. I tested pg94_95_use_vfd_for_logrep.patch for 9.6 branch, and it
> > works there. So please use this patch for all the three branches.
> >
>
> Pushed!
>


I see one failure in REL_10_STABLE [1] which seems to be due to this commit:

Test Summary Report
-------------------
t/006_logical_decoding.pl          (Wstat: 7424 Tests: 10 Failed: 0)
  Non-zero exit status: 29
  Parse errors: Bad plan.  You planned 11 tests but ran 10.
Files=14, Tests=122, 1968 wallclock secs ( 0.10 usr  0.03 sys + 19.00 cusr
21.98 csys = 41.11 CPU)
Result: FAIL
Makefile:19: recipe for target 'check' failed
make: *** [check] Error 1


See below snippet from 006_logical_decoding_master.log
..
..
2020-01-03 01:30:48.254 UTC [12189836:9] t/006_logical_decoding.pl
STATEMENT:  SELECT data from pg_logical_slot_get_changes('test_slot', NULL,
NULL)
   WHERE data LIKE '%INSERT%' ORDER BY lsn LIMIT 1;
2020-01-03 01:30:51.990 UTC [6882174:3] LOG:  server process (PID 12189836)
was terminated by signal 11
2020-01-03 01:30:51.990 UTC [6882174:4] DETAIL:  Failed process was
running: SELECT data from pg_logical_slot_get_changes('test_slot', NULL,
NULL)
   WHERE data LIKE '%INSERT%' ORDER BY lsn LIMIT 1;
2020-01-03 01:30:51.990 UTC [6882174:5] LOG:  terminating any other active
server processes

The strange thing is that the same test passes on master on the same
machine [2] and for 10 as well, it passes on other machines, so not sure
what could cause this.  Any clue?

[1] -
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=tern&dt=2020-01-02%2023%3A36%3A31
[2] -
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=tern&dt=2020-01-02%2010%3A37%3A33

-- 
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com