Re: Improving connection scalability: GetSnapshotData()

On 2020/09/08 13:11, Andres Freund wrote:
> Hi,
> 
> On 2020-09-08 13:03:01 +0900, Ian Barwick wrote:
(...)
>> I wonder if it's possible to increment "xactCompletionCount"
>> during replay along these lines:
>>
>>      *** a/src/backend/access/transam/xact.c
>>      --- b/src/backend/access/transam/xact.c
>>      *************** xact_redo_commit(xl_xact_parsed_commit *
>>      *** 5915,5920 ****
>>      --- 5915,5924 ----
>>               */
>>              if (XactCompletionApplyFeedback(parsed->xinfo))
>>                      XLogRequestWalReceiverReply();
>>      +
>>      +       LWLockAcquire(ProcArrayLock, LW_EXCLUSIVE);
>>      +       ShmemVariableCache->xactCompletionCount++;
>>      +       LWLockRelease(ProcArrayLock);
>>        }
>>
>> which seems to work (though quite possibly I've overlooked something I don't
>> know that I don't know about and it will all break horribly somewhere,
>> etc. etc.).
> 
> We'd also need the same in a few more places. Probably worth looking at
> the list where we increment it on the primary (particularly we need to
> also increment it for aborts, and 2pc commit/aborts).

Yup.

> At first I was very confused as to why none of the existing tests have
> found this significant issue. But after thinking about it for a minute
> that's because they all use psql, and largely separate psql invocations
> for each query :(. Which means that there's no cached snapshot around...
> 
> Do you want to try to write a patch?

Sure, I'll give it a go as I have some time right now.


Regards

Ian Barwick

-- 
Ian Barwick                   https://www.2ndQuadrant.com/
  PostgreSQL Development, 24x7 Support, Training & Services