|From:||Michael Paquier <michael(at)paquier(dot)xyz>|
|To:||Dagfinn Ilmari Mannsåker <ilmari(at)ilmari(dot)org>|
|Subject:||Re: Logging query parmeters in auto_explain|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
On Thu, Jun 09, 2022 at 11:55:11PM +0100, Dagfinn Ilmari Mannsåker wrote:
> Done (and more tests added), v3 attached.
One thing I am wondering is if we'd better mention errdetail_params()
at the top of the initial check done in ExplainQueryParameters().
That's a minor issue, though.
Perhaps a short description at the top of this routine to explain it
is worth it? The test still does a set of like() and unlike() after
running each query when the parameter updates are done. One thing I
would have played with is to group the set of logs expected or not
expected as parameters of the centralized routine, but that would
reduce the customization of the test names, so at the end the approach
you have taken for query_log() looks good to me.
There is no need for that. The END block of Cluster.pm does that
|Next Message||Bharath Rupireddy||2022-06-14 07:50:39||Re: A proposal to force-drop replication slots to make disabling async/sync standbys or logical replication faster in production environments|
|Previous Message||Peter Smith||2022-06-14 07:35:31||Re: tablesync copy ignores publication actions|