David Rowley wrote:
> Okay, I've written and attached a fix for this. I'm not 100% certain
> that this is the cause of the problem on pademelon, but the code does
> look wrong, so needs to be fixed. Hopefully, it'll make pademelon
> happy, if not I'll think a bit harder about what might be causing that
> instability.
Pushed it just now. Let's see what happens with pademelon now.
> The misleading comment claimed we unset the extern params so we didn't
> perform pruning again using these. I'd failed to update this comment
> after I realised that we still need to attempt to prune again with the
> external params since quals against the partition key may actually
> contain a mix of exec and external params, which would mean that it's
> only possible to prune partitions using both types of params. No
> actual code needs to be updated since the 2nd pass of pruning uses
> "allparams" anyway. We could actually get away without the bms_free()
> and set to NULL in the lines below the comment, but I wanted some way
> to ensure that we never write any code which calls the function twice
> on the same PartitionPruneState, but maybe I'm just overly paranoid
> there.
Pushed this earlier today, thanks.
--
Álvaro Herrera https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
