pg.ddx.io pgsql-admin@postgresql.org mailing list archive
help / color / mirror / Atom feedProcess
2+ messages / 2 participants
[nested] [flat]
* Process
@ 2007-04-10 13:21 Alexander B. <burbello3000@yahoo.com.br>
0 siblings, 1 reply; 2+ messages in thread
From: Alexander B. @ 2007-04-10 13:21 UTC (permalink / raw)
To: pgsql-admin
Hi,
I have a Postgres server that has more than 500MB ( shared_buffers =
64000) and looking at TOP command, I couldn't understand why appears
some process with 536m and other with 23m.
What's the difference between?
Wouldn't be all process with 536m, including client connections, but not
allocating all this memory, just server process?
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
4039 postgres 16 0 537m 13m 536m S 0.0 0.7 0:01.53 postmaster
4040 postgres 16 0 25572 3356 23m S 0.0 0.2 0:00.14 postmaster
4042 postgres 15 0 537m 400m 536m S 0.0 20.0 0:01.63 postmaster
4043 postgres 16 0 25568 3424 23m S 0.0 0.2 0:00.01 postmaster
4044 postgres 16 0 26572 4396 23m S 0.0 0.2 0:01.85 postmaster
4045 postgres 15 0 25848 3584 23m S 0.0 0.2 0:00.96 postmaster
Thanks in advance for you help.
^ permalink raw reply [nested|flat] 2+ messages in thread
* Re: Process
@ 2007-04-10 14:36 Tom Lane <tgl@sss.pgh.pa.us>
parent: Alexander B. <burbello3000@yahoo.com.br>
0 siblings, 0 replies; 2+ messages in thread
From: Tom Lane @ 2007-04-10 14:36 UTC (permalink / raw)
To: Alexander B. <burbello3000@yahoo.com.br>; +Cc: pgsql-admin
"Alexander B." <burbello3000@yahoo.com.br> writes:
> I have a Postgres server that has more than 500MB ( shared_buffers =
> 64000) and looking at TOP command, I couldn't understand why appears
> some process with 536m and other with 23m.
> What's the difference between?
Some versions of top only count the pages of shared memory that a
given process has actually touched so far. So for example in the
shared buffer arena, the reported usage will rise as the process runs
longer and chances to use buffers other than the ones it's used already.
It's pretty misleading, because it makes normal behavior look like a
memory leak :-(
regards, tom lane
^ permalink raw reply [nested|flat] 2+ messages in thread
end of thread, other threads:[~2007-04-10 14:36 UTC | newest]
Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2007-04-10 13:21 Process Alexander B. <burbello3000@yahoo.com.br>
2007-04-10 14:36 ` Tom Lane <tgl@sss.pgh.pa.us>
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox