Re: logical replication and PANIC during shutdown checkpoint in publisher

On Fri, May 5, 2017 at 5:33 PM, Pavan Deolasee <pavan(dot)deolasee(at)gmail(dot)com> wrote:
>
>
> On Fri, May 5, 2017 at 10:56 AM, Michael Paquier <michael(dot)paquier(at)gmail(dot)com>
> wrote:
>>
>> On Wed, May 3, 2017 at 12:25 AM, Peter Eisentraut
>>
>>
>> >>> Can we prevent HOT pruning during logical decoding?
>> >>
>> >> It does not sound much difficult to do, couldn't you just make it a
>> >> no-op with am_walsender?
>> >
>> > That's my hope.
>>
>> The only code path doing HOT-pruning and generating WAL is
>> heap_page_prune(). Do you think that we need to worry about FPWs as
>> well?
>
>
> IMO the check should go inside heap_page_prune_opt(). Do we need to worry
> about wal_log_hints or checksums producing WAL because of hint bit updates?
> While I haven't read the thread, I am assuming if HOT pruning can happen,
> surely hint bits can get set too.

Yeah, that's as well what I am worrying about. Experts of logical
decoding will correct me, but it seems to me that we have to cover all
the cases where heap scans can generate WAL.
--
Michael