From: | Amin Abdulghani <amin(at)quantiva(dot)com> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Transaction Id wraparounds |
Date: | 2002-09-05 17:29:30 |
Message-ID: | web-772046@quantiva.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Thanks for the replies. As for the warning message,
probably
it may be better to put it at the DB connection time.
Another option might be to possibly do a forced automatic
VACUUM in such cases and thus avoid any potential
integrity problem. This could possibly be a configuration
option FORCE_VACUUM if age(FrozenXID)>threshold.
Thanks..
Amin
On Thu, 05 Sep 2002 12:08:04 -0400
Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>Amin Abdulghani <amin(at)quantiva(dot)com> writes:
>> From the administrator's guide I understand a
>> system wide VACUUM or VACUUM FREEZE would perform the
>>job
>> but for a large database (>10GB and growing) this takes
>>a
>> long time to run.
>
>So? You don't have to do it often, and it doesn't lock
>your
>tables against normal operations ... so I don't really
>see the
>objection. I would not bother with FREEZE unless you are
>trying
>to make a template database.
>
>> what should be the criteria to check when this issue
>>needs
>> to be addressed.
>
>When age(datfrozenxid) approaches 2 billion, it's time.
> I believe
>vacuum will start warning you well before that.
>
> regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Stephan Szabo | 2002-09-05 17:31:15 | Re: "...integer[] references..." = error |
Previous Message | frbn | 2002-09-05 17:22:23 | Re: PostgreSQL vs MySQL : strange results on insertion |