Re: [HACKERS] Why does logical replication launcher setapplication_name?

On 6/2/17 23:23, Peter Eisentraut wrote:
> On 6/2/17 15:08, Peter Eisentraut wrote:
>> On 5/30/17 23:10, Peter Eisentraut wrote:
>>> Here is a proposed solution that splits bgw_name into bgw_type and
>>> bgw_name_extra.  bgw_type shows up in pg_stat_activity.backend_type.
>>> Uses of application_name are removed, because they are no longer
>>> necessary to identity the process type.
>>>
>>> This code appears to be buggy because I sometimes get NULL results of
>>> the backend_type lookup, implying that it couldn't find the background
>>> worker slot.  This needs another look.
>>
>> I would like some more input on this proposal, especially from those
>> have have engineered the extended pg_stat_activity content.
>>
>> If we don't come to a quick conclusion on this, I'd be content to leave
>> PG10 as is and put this patch into the next commit fest.
> 
> If there are no new insights into this by Monday, I will commit patches
> that remove the setting of application_name, which was originally
> complained about, and postpone the rest of this patch.

Done, and added the rest of the patch to the next commit fest:
https://commitfest.postgresql.org/14/1165/

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