| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | strk <strk(at)keybit(dot)net> |
| Cc: | pgsql-hackers(at)postgresql(dot)org, postgis-devel(at)postgis(dot)refractions(dot)net |
| Subject: | Re: index leaks ? |
| Date: | 2004-03-12 15:04:20 |
| Message-ID: | 9180.1079103860@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
strk <strk(at)keybit(dot)net> writes:
> The size growth is reported by 'top' in the fields
> SIZE, RSS and SHARE.
> Can it be a memory leak in postgres code ?
No, you are misinterpreting the 'top' output.
You didn't say what platform you are on, but on some systems 'top'
increases the reported size of a backend process each time that process
touches a page of the shared memory area that it had not touched before.
In this case you are simply watching the process use shared buffers it
hadn't previously touched.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andrew Dunstan | 2004-03-12 15:14:06 | Re: The Name Game: postgresql.net vs. pgfoundry.org |
| Previous Message | Jeroen T. Vermeulen | 2004-03-12 14:55:54 | Re: The Name Game: postgresql.net vs. pgfoundry.org |