Re: BUG #16016: deadlock with startup process, AccessExclusiveLockon pg_statistic's toast table

On 11/2/19 21:10, Tom Lane wrote:
> This does not look like a deadlock: the startup process is just biding
> its time until the standby delay timeout elapses, after which it's
> gonna kill the conflicting queries.
>
> It is, perhaps, arguable that it's a damn bad idea to allow
> max_standby_streaming_delay or max_standby_archive_delay to be
> set to "forever".
>
>

I agree, allowing max_standby_streaming_delay be set to -1 is perhaps
not a good idea. there is although case when such setting could be used
now, see "BUG #14044: Queries immediately conflict with recovery when 
recovery_min_apply_delay is used". [1]

[1] https://postgr.es/m/20160324204550.2920.87946%40wrigleys.postgresql.org

--
Alexey Ermakov