Re: logical decoding : exceeded maxAllocatedDescs for .spill files

On Tue, 24 Dec 2019 at 10:41, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
>
> On Fri, Dec 20, 2019 at 9:31 AM Amit Khandekar <amitdkhan(dot)pg(at)gmail(dot)com> wrote:
> > Attached are the patches from master back up to 94 branch.
> >
> > PG 9.4 and 9.5 have a common patch to be applied :
> > pg94_95_use_vfd_for_logrep.patch
> > From PG 9.6 onwards, each version has a separate patch.
> >
> > For PG 9.6, there is no logical decoding perl test file. So I have
> > made a new file 006_logical_decoding_spill.pl that has only the
> > specific testcase. Also, for building the test_decoding.so, I had to
> > add the EXTRA_INSTALL=contrib/test_decoding line in the
> > src/test/recovery/Makefile, because this is the first time we are
> > using the plugin in the 9.6 tap test.
> >
>
> I am not sure if we need to go that far for 9.6 branch. If the other
> tests for logical decoding are not present, then I don't see why we
> need to create a new test file for this test only. Also, I think this
> will make the patch the same for 9.4,9.5 and 9.6.

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.

>
> > From PG 10 onwards, pgstat_report_*() calls around read() are removed
> > in the patch, because FileRead() itself reports the wait events.
> >
>
> Why there are different patches for 10 and 11?
For PG10, OpenTransientFile() and PathNameOpenFile() each have an
extra parameter for specifying file creation modes such as S_IRUSR or
S_IWUSR. For 11, these functions don't accept the flags, rather the
file is always opened with PG_FILE_MODE_OWNER. Because of these
differences in the calls, the PG 10 patch does not apply on 11.

>
> We should try to minimize the difference between patches in different branches.
Ok.

--
Thanks,
-Amit Khandekar
EnterpriseDB Corporation
The Postgres Database Company