Alvaro Herrera writes:
> Glitches fixed in this version; will apply shortly to 8.3 and HEAD.
Looks sane; one trivial grammar correction:
> + /* the minimum allowed time between two awakening of the launcher */
Should read "two awakenings".
regards, tom lane
--
Sent vi
Alvaro Herrera escribió:
> Tom Lane escribió:
>
> > Well, that code isn't even correct I think; you're not supposed to
> > modify a GUC variable directly. I think you should just silently
> > use a naptime of at least X without changing the nominal GUC variable.
> > And definitely without the WAR
Tom Lane escribió:
> Well, that code isn't even correct I think; you're not supposed to
> modify a GUC variable directly. I think you should just silently
> use a naptime of at least X without changing the nominal GUC variable.
> And definitely without the WARNING --- that's nothing but log spam.
Alvaro Herrera writes:
> I've been giving this some thought and tried several approaches. In the
> end the one that I like the most is raising autovacuum_naptime to a
> reasonable value for the exiting number of databases. The only problem
> I have with it is that it's trivial to change it in th
Tom Lane escribió:
> Bingo, that's surely exactly what was happening to the OP. He had 2000
> databases and naptime at (I assume) the default; so he was rerunning
> rebuild_database_list every 100ms.
>
> So that recovery code path needs some more thought. Maybe a lower bound
> on how often to d
2009/5/26 Tom Lane :
> Alvaro Herrera writes:
>> However I think there's a bigger problem here, which is that if the user
>> has set naptime too low, i.e. to a value lower than
>> number-of-databases * 100ms, we'll be running the (expensive)
>> rebuild_database_list function on each iteration ...
Alvaro Herrera writes:
> However I think there's a bigger problem here, which is that if the user
> has set naptime too low, i.e. to a value lower than
> number-of-databases * 100ms, we'll be running the (expensive)
> rebuild_database_list function on each iteration ... maybe we oughta put
> a low
Tom Lane escribió:
> Alvaro Herrera writes:
> > Tom Lane escribi�:
> >> Hmm, maybe we need to improve the code too. This example suggests that
> >> there needs to be some limit on the worker launch rate, even if there
> >> are so many databases that that means we don't meet naptime exactly.
>
>
Alvaro Herrera writes:
> Tom Lane escribió:
>> Hmm, maybe we need to improve the code too. This example suggests that
>> there needs to be some limit on the worker launch rate, even if there
>> are so many databases that that means we don't meet naptime exactly.
> We already have a 100ms lower b
Tom Lane escribió:
> Hmm, maybe we need to improve the code too. This example suggests that
> there needs to be some limit on the worker launch rate, even if there
> are so many databases that that means we don't meet naptime exactly.
We already have a 100ms lower bound on the sleep time (see
la
Alvaro Herrera writes:
> Sorry, it's the other way around actually -- correct for 8.3 onwards,
> wrong for 8.1 and 8.2. In the earlier versions, it would do one run in
> a chosen database, sleep during "naptime", then do another run.
> Tom Lane escribió:
>> I suppose the use of "minimum" means
W dniu 26 maja 2009 20:28 użytkownik Tom Lane napisał:
>>> I believe the interpretation of autovacuum_naptime is that it should
>>> examine each database that often, ie once a minute by default. So
>>> it's got more than 30 databases per second to look through.
>
>> Note that this is correct in 8
Tom Lane escribió:
> Alvaro Herrera writes:
> > Tom Lane escribi�:
> >> I believe the interpretation of autovacuum_naptime is that it should
> >> examine each database that often, ie once a minute by default. So
> >> it's got more than 30 databases per second to look through.
>
> > Note that thi
Alvaro Herrera writes:
> Tom Lane escribió:
>> I believe the interpretation of autovacuum_naptime is that it should
>> examine each database that often, ie once a minute by default. So
>> it's got more than 30 databases per second to look through.
> Note that this is correct in 8.1 and 8.2 but n
Tom Lane escribió:
> =?UTF-8?B?xYF1a2FzeiBKYWdpZcWCxYJv?= writes:
> > That autovacuum working hole time, shoudn't be run only when db needs ?
>
> With 2000 databases to cycle through, autovac is going to be spending
> quite a lot of time just finding out whether it needs to do anything.
> I belie
2009/5/25 Łukasz Jagiełło :
> W dniu 25 maja 2009 17:32 użytkownik Scott Marlowe
> napisał:
>>> Recent change postgresql server from Amazon EC2 small into large one.
>>> That gives me x86_64 arch, two core cpu and 7.5GB ram. Atm got almost
>>> ~2000 small databases at that server and autovacuum wo
2009/5/25 Tom Lane :
> With 2000 databases to cycle through, autovac is going to be spending
> quite a lot of time just finding out whether it needs to do anything.
> I believe the interpretation of autovacuum_naptime is that it should
> examine each database that often, ie once a minute by default
=?UTF-8?B?xYF1a2FzeiBKYWdpZcWCxYJv?= writes:
> That autovacuum working hole time, shoudn't be run only when db needs ?
With 2000 databases to cycle through, autovac is going to be spending
quite a lot of time just finding out whether it needs to do anything.
I believe the interpretation of autova
W dniu 25 maja 2009 17:50 użytkownik Grzegorz Jaśkiewicz
napisał:
>> So, in 2000 databases, there's only an average of 2 relations per db
>> and 102 dead rows? Cause that's all you got room for with those
>> settings.
>>
>> Whats the last 20 or so lines of vacuum verbose as run by a superuser say
W dniu 25 maja 2009 17:32 użytkownik Scott Marlowe
napisał:
>> Recent change postgresql server from Amazon EC2 small into large one.
>> That gives me x86_64 arch, two core cpu and 7.5GB ram. Atm got almost
>> ~2000 small databases at that server and autovacuum working hole time
>
>> postgresql.con
2009/5/25 Grzegorz Jaśkiewicz :
> 2009/5/25 Scott Marlowe :
>
>>
>> So, in 2000 databases, there's only an average of 2 relations per db
>> and 102 dead rows? Cause that's all you got room for with those
>> settings.
>>
>> Whats the last 20 or so lines of vacuum verbose as run by a superuser say?
2009/5/25 Scott Marlowe :
>
> So, in 2000 databases, there's only an average of 2 relations per db
> and 102 dead rows? Cause that's all you got room for with those
> settings.
>
> Whats the last 20 or so lines of vacuum verbose as run by a superuser say?
according to
http://www.postgresql.org/
2009/5/24 Łukasz Jagiełło :
> Hi,
>
> Recent change postgresql server from Amazon EC2 small into large one.
> That gives me x86_64 arch, two core cpu and 7.5GB ram. Atm got almost
> ~2000 small databases at that server and autovacuum working hole time
> postgresql.conf:
> max_fsm_pages = 204800
>
Hi,
Recent change postgresql server from Amazon EC2 small into large one.
That gives me x86_64 arch, two core cpu and 7.5GB ram. Atm got almost
~2000 small databases at that server and autovacuum working hole time
(witch isn't good for performance as I notice at cacti, one core is
busy in 60% hole
24 matches
Mail list logo
