Re: Backend handling replication slot stuck using 100% cpu, unkillable

On Tue, Jul 04, 2023 at 01:30:21PM +0200, Tomas Vondra wrote:
> On 7/3/23 14:58, hubert depesz lubaczewski wrote:
> So is it an infinite loop in ReorderBufferExecuteInvalidations, or is it
> just the case that there are many invalidations? I can't really deduce
> that from the backtraces.
> 
> How many invalidations does the transaction have? Should be enough to
> 
>   print txn->ninvalidations
> 
> Also, is there anything interesting about the transaction? You know the
> XID (2741814901) so maybe use pg_waldump to see what it did.

I don't have the process anymore. Pg was restarted, and the app was
forced to not take data from this place.

I can run pg_waldump, no problem, but I'm not an expert on this. Which
wal file(s) should I try to dump? and what to look for (or should I just
provide full output of dump?

Best regards,

depesz