Re: BUG #16129: Segfault in tts_virtual_materialize in logical replication worker

On Thu, Nov 21, 2019 at 02:46:45PM +0100, Ondřej Jirman wrote:
>On Thu, Nov 21, 2019 at 02:37:14PM +0100, Tomas Vondra wrote:
>> On Thu, Nov 21, 2019 at 02:12:02PM +0100, Ondřej Jirman wrote:
>> > On Thu, Nov 21, 2019 at 01:59:30PM +0100, Tomas Vondra wrote:
>> > > On Thu, Nov 21, 2019 at 12:53:26PM +0100, Ondřej Jirman wrote:
>> > > >
>> > > > ...
>> > > >
>> > > OK, so this is the problematic table. Any idea what operation triggers
>> > > the issue? Having a small reproducer (say, a single-row insert or so)
>> > > would be very helpful.
>> >
>> > Can something be extracted from the logical decoding data I sent in the other
>> > e-mail? That seems to contain the failing operation, but I don't know how to
>> > decode it off the top of my head from that binary representation.
>> >
>>
>> Well, why not use pg_logical_slot_peek_changes()? That should be human
>> readable, I think.
>
>I would, but how?
>
>SELECT * FROM pg_logical_slot_peek_changes('l5_hometv', NULL, NULL,
> 'proto_version', '1', 'publication_names', 'pub');
>ERROR: logical decoding output plugin "pgoutput" produces binary output, but
>function "pg_logical_slot_peek_changes(name,pg_lsn,integer,text[])" expects
>textual data
>

Ah, right. Sorry, I forgot this is pgoutput.

regards

--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services