|
Hi all,
In our server Check pointer process is consuming 8 GB of memory, what could be the possible reason? Can any one please help.
|
|
On 5/30/13 8:09 AM, itishree sukla wrote:
> In our server Check pointer process is consuming 8 GB of memory, what > could be the possible reason? Can any one please help. That process will eventually access all of shared_buffers, which shows as a shared memory block for that process. That's what you're seeing there. It doesn't actually use any significant amount of memory on its own. -- Greg Smith 2ndQuadrant US Baltimore, MD PostgreSQL Training, Services, and 24x7 Support www.2ndQuadrant.com -- Sent via pgsql-performance mailing list ( ) To make changes to your subscription: |
|
In reply to
this post
by itishree sukla
Hi ,
Could you share the command, what you have used to confirm that, the checkpoint process is consuming 8GB. And also, please share the addition information like PostgreSQL version and the OS details.
I am suspecting that, your shared_buffers value is 8GB, and the "top" command is showing the used memory as 8GB.
Thanks.
Dinesh
-- Dinesh Kumar
Software Engineer
Ph : Follow us on Twitter @EnterpriseDB Visit EnterpriseDB for tutorials, webinars, whitepapers and more
On Thu, May 30, 2013 at 5:39 PM, itishree sukla
<
>
wrote:
|
|
In reply to
this post
by itishree sukla
> In our server Check pointer process is consuming 8 GB of memory, what could > be the possible reason? Can any one please help. Are you sure you're measuring the memory correctly? The RES field in top output, for example, includes shared memory, ie. the whole buffer cache. Shared memory isn't really "consumed" by the checkpointer process, but shared by all postgres processes. - Heikki -- Sent via pgsql-performance mailing list ( ) To make changes to your subscription: |
|
Thanks for the quick response. Below is the out put of Top Commnd.
Postgresql =9.2.3
3971 postgres 20 0 8048m 303m 301m S 0 0.9 0:04.34 /usr/lib/postgresql/9.2/bin/postgres -D /var/lib/postgresql/9.2/main -c config_file=/etc/postgre 3972 postgres 20 0 66828 1820 708 S 0 0.0 1:36.37 postgres: logger process 3974 postgres 20 0 8054m 7.6g 7.6g S 0 24.1 0:56.59 postgres: checkpointer process 3975 postgres 20 0 8051m 895m 891m S 0 2.8 0:04.98 postgres: writer process 3976 postgres 20 0 8051m 9m 9072 S 0 0.0 0:35.17 postgres: wal writer process 3977 postgres 20 0 70932 3352 716 S 0 0.0 0:05.19 postgres: stats collector process 1
On Thu, May 30, 2013 at 5:56 PM, Heikki Linnakangas
<
>
wrote:
Are you sure you're measuring the memory correctly? The RES field in top output, for example, includes shared memory, ie. the whole buffer cache. Shared memory isn't really "consumed" by the checkpointer process, but shared by all postgres processes. |
|
Can any one give more input, you can see my top out put, in %MEM its taking 24.1.
Regards,
On Thu, May 30, 2013 at 6:15 PM, itishree sukla
<
>
wrote:
|
|
itishree sukla <
> wrote:
> itishree sukla < > wrote: >> Heikki Linnakangas < > wrote: >>> itishree sukla wrote: >>> >>>> In our server Check pointer process is consuming 8 GB of >>>> memory, what could be the possible reason? Can any one please >>>> help. >>> >>> Are you sure you're measuring the memory correctly? The RES >>> field in top output, for example, includes shared memory, ie. >>> the whole buffer cache. Shared memory isn't really "consumed" >>> by the checkpointer process, but shared by all postgres processes. >> >> Thanks for the quick response. Below is the out put of Top >> Commnd. >> >> 3971 postgres 20 0 8048m 303m 301m S 0 0.9 0:04.34 3971 postgres 20 0 8048m 303m 301m S 0 0.9 0:04.34 /usr/lib/postgresql/9.2/bin/postgres -D /var/lib/postgresql/9.2/main -c config_file=/etc/postgre >> 3972 postgres 20 0 66828 1820 708 S 0 0.0 1:36.37 postgres: logger process >> 3974 postgres 20 0 8054m 7.6g 7.6g S 0 24.1 0:56.59 postgres: checkpointer process >> 3975 postgres 20 0 8051m 895m 891m S 0 2.8 0:04.98 postgres: writer process >> 3976 postgres 20 0 8051m 9m 9072 S 0 0.0 0:35.17 postgres: wal writer process >> 3977 postgres 20 0 70932 3352 716 S 0 0.0 0:05.19 postgres: stats collector process > Can any one give more input, you can see my top out put, in %MEM > its taking 24.1. It's not. It's referencing all of your shared_buffers. -- Kevin Grittner EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company -- Sent via pgsql-performance mailing list ( ) To make changes to your subscription: |
|
In reply to
this post
by itishree sukla
--On 30. Mai 2013 18:15:04 +0530 itishree sukla < > wrote: > Thanks for the quick response. Below is the out put of Top Commnd. > > 3971 postgres 20 0 8048m 303m 301m S 0 0.9 0:04.34 > /usr/lib/postgresql/9.2/bin/postgres -D /var/lib/postgresql/9.2/main -c > config_file=/etc/postgre > 3972 postgres 20 0 66828 1820 708 S 0 0.0 1:36.37 > postgres: logger > process > > > 3974 postgres 20 0 8054m 7.6g 7.6g S 0 24.1 0:56.59 > postgres: checkpointer > process > On Linux i often find the pmap utility a far better tool to get an idea what a process actually consumes of memory. The output can be large sometimes, but it's more "fine grained". -- Thanks Bernd -- Sent via pgsql-performance mailing list ( ) To make changes to your subscription: |
|
In reply to
this post
by itishree sukla
On Thu, Jun 6, 2013 at 1:44 AM, itishree sukla
<
>
wrote:
Depesz wrote a nice topic on his blog about this subject [1], read it and try the commands to see the real memory usage by checkpointer (when I say "real", I mean "private").
[1]
http://www.depesz.com/2012/06/09/how-much-ram-is-postgresql-using/
Regards,
--
Matheus de Oliveira Analista de Banco de Dados Dextra Sistemas - MPS.Br nível F! www.dextra.com.br/postgres |
| Free forum by Nabble | Edit this page |