Re: Multixacts wraparound monitoring

Yes, VACUUM helps to solve the problem and the WARNING gone away.
But, the problem is that the VACUUM for the entire database (2.4T) takes over 7 hours, and it has to run every 15-20 hours (about 300 millions transactions), otherwise:
ERROR: multixact "members" limit exceeded - and server stops working.
The question is how to start the VACUUM at least once in three days.

Kind regards,
 
Vladimir Pavlov

-----Original Message-----
From: Adrian Klaver [mailto:adrian(dot)klaver(at)aklaver(dot)com]
Sent: Wednesday, March 30, 2016 4:52 PM
To: Pavlov Vladimir; 'Alvaro Herrera'
Cc: 'pgsql-general(at)postgresql(dot)org'
Subject: Re: [GENERAL] Multixacts wraparound monitoring

On 03/30/2016 06:24 AM, Pavlov, Vladimir wrote:
> Hello,
> There is no news?
> Now I have to do VACUUM every night, so that the server worked.

So has the WARNING gone away?:

WARNING: oldest multixact is far in the past
HINT: Close open transactions with multixacts soon to avoid wraparound problems.

Or to put it another way, define worked.

> Maybe run VACUUM FREEZE?
>
> Kind regards,
>
> Vladimir Pavlov
>
>
> -----Original Message-----
> From: Pavlov Vladimir
> Sent: Friday, March 25, 2016 9:55 AM
> To: 'Alvaro Herrera'
> Cc: 'Adrian Klaver'; pgsql-general(at)postgresql(dot)org
> Subject: RE: [GENERAL] Multixacts wraparound monitoring
>
> Hi, thank you very much for your help.
> Pg_control out in the attachment.
>
> Kind regards,
>
> Vladimir Pavlov
>
>
> -----Original Message-----
> From: Alvaro Herrera [mailto:alvherre(at)2ndquadrant(dot)com]
> Sent: Friday, March 25, 2016 12:25 AM
> To: Pavlov Vladimir
> Cc: 'Adrian Klaver'; pgsql-general(at)postgresql(dot)org
> Subject: Re: [GENERAL] Multixacts wraparound monitoring
>
> Pavlov, Vladimir wrote:
>> There is nothing:
>> select * from pg_prepared_xacts;
>> transaction | gid | prepared | owner | database
>> -------------+-----+----------+-------+----------
>> (0 rows)
>> It is also noticed that a lot of files in a directory main/pg_multixact/members/, now - 69640.
>
> Can you attach pg_controldata output?
>

--
Adrian Klaver
adrian(dot)klaver(at)aklaver(dot)com