From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Alvaro Herrera <alvherre(at)commandprompt(dot)com> |
Cc: | Pavan Deolasee <pavan(dot)deolasee(at)gmail(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Snapshot warning |
Date: | 2008-11-25 19:37:01 |
Message-ID: | 4731.1227641821@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Alvaro Herrera <alvherre(at)commandprompt(dot)com> writes:
> Tom Lane escribi:
>> Hmm, that's a bit problematic because resowner.c doesn't have any global
>> notion of what resource owners exist. I think you still need to have
>> snapmgr.c maintain a list of all known snapshots. resowner.c can only
>> help you with tracking reference counts for particular snapshots.
> A counter seems to suffice. Patch attached.
Looks sane to me. The list solution might be needed later, if we wanted
to get smarter about advancing our xmin after deleting only some of our
snapshots ... but this'll do for now.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2008-11-25 19:42:15 | Re: Exporting PGINTERVALSTYLE prevents access to older server versions |
Previous Message | Alvaro Herrera | 2008-11-25 19:30:18 | Re: Snapshot warning |