On 2/4/16 3:37 PM, Andres Freund wrote:
> On 2016-02-03 09:28:24 -0500, Robert Haas wrote:
>> Would we still have some way of forcing the older checkpoint record to
>> be used if somebody wants to try to do that?
>
> I think currently the best way to force an arbitrary checkpoint to be
> used is creating a "custom" backup label. Not that nice. Not sure if we
> need something nice here, I don't really see a frequent need for this.
>
> We could add another option to pg_resetxlog alternatively :/
I guess you'd have to scan through WAL files by hand to find the next
oldest checkpoint?
I'm guessing that if this is happening in the field there's a decent
chance people aren't noticing it, so maybe the best thing for now is to
turn off the automatic behavior bust still have a relatively easy way to
re-enable it. In case this is more common than we think...
--
Jim Nasby, Data Architect, Blue Treble Consulting, Austin TX
Experts in Analytics, Data Architecture and PostgreSQL
Data in Trouble? Get it in Treble! http://BlueTreble.com