| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Jakub Ouhrabka <kuba(at)comgate(dot)cz> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: ERROR: Out of memory - when connecting to database |
| Date: | 2010-11-08 20:51:44 |
| Message-ID: | 29947.1289249504@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Jakub Ouhrabka <kuba(at)comgate(dot)cz> writes:
>>> They clearly were: notice the reference to "Autovacuum context" in the
>>> memory map. I think you are right to suspect that auto-analyze was
>>> getting blown out by the wide bytea columns. Did you have any
>>> expression indexes involving those columns?
> Yes, there are two unique btree indexes:
> (col1, col2, col3, md5(array_to_string(col_bytea_arr, ''::text)) where
> col4 is not null
> (col1, col2, col3, md5(array_to_string(col_bytea_arr, ''::text)) where
> col4 is null
> What is the cause of the out of memory then?
Hmm. I suspect that evaluation of that md5() call is resulting in
memory leakage, but not sure why or where exactly.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2010-11-08 21:03:25 | Re: Removing pgsql_tmp files |
| Previous Message | Tom Lane | 2010-11-08 20:45:12 | Re: finding the other statement causing a sharelock |