Hi,
Doing the UPDATE on the child table (provided that the table does exist) as
you recommended solved all my memory consumption issue.
Thanks a lot,
Alessandro.
On Tuesday 23 December 2014 15:27:41 Tom Lane wrote:
> Alessandro Ipe writes:
> > I guess the memory consumption is
les&t=5e0e9e1bb06dce1d12c95662a9ee1c03
The queries causing the issue are given in files
- tmp.OqOavPYbHa (with the new upsert_func function)
- tmp.f60wlgEDWB (with WITH .. AS statement)
I hope it will help. Thanks.
Regards,
A.
On Thursday 18 December 2014 12:05:45 Tom Lane wrote:
> Alessandro Ipe writes:
> &g
Hi,
I can send a full dump of my database (< 2MB) if it is OK for you.
Thanks,
A.
On Thursday 18 December 2014 12:05:45 Tom Lane wrote:
> Alessandro Ipe writes:
> > Hi,
> > I tried also with an upsert function
> > CREATE FUNCTION upsert_func(sql_insert text, sql_up
;
RETURN;
END;
$$;
with the same result on the memory used...
The tables hold 355000 rows in total.
Regards,
A.
On Thursday 18 December 2014 12:16:49 Alessandro Ipe wrote:
> Hi,
>
>
> A grep in a nightly dump of this database did not return any AFTER trigger.
> Th
On Thursday 18 December 2014 08:51:47 Torsten Förtsch wrote:
> On 17/12/14 16:14, Alessandro Ipe wrote:
> > 2014-12-15 17:54:07 GMT LOG: server process (PID 21897) was
terminated
> > by signal 9: Killed
>
> since it was killed by SIGKILL, maybe it's the kernel'
A.
On Wednesday 17 December 2014 12:49:03 Tom Lane wrote:
> Alessandro Ipe writes:
> > My dtrigger definition is
> > CREATE TRIGGER msg_trigger BEFORE INSERT ON msg FOR EACH ROW EXECUTE
> > PROCEDURE msg_function(); so it seems that it is a BEFORE trigger.
>
> Hm, n
Hi,
My dtrigger definition is
CREATE TRIGGER msg_trigger BEFORE INSERT ON msg FOR EACH ROW EXECUTE PROCEDURE
msg_function();
so it seems that it is a BEFORE trigger.
To be totally honest, I have "really" limited knownledge in SQL and postgresql
and all these were gathered from recipes found on
Hi Torsten,
Thanks for your answer.
I have modified
(SELECT * FROM upsert)
to
(SELECT * FROM upsert WHERE slot=to_timestamp('201212032145', 'MMDDHH24MI')
and MSG=2)
according to your suggestion to reduce the result-set to a single row. However,
the INSERT process is still consuming the sam
Hi,
Software and hardware running postgresql are:
- postgresql92-9.2.3-1.1.1.x86_64
- openSuSE 12.3 x64_86
- 16 GB of RAM
- 2 GB of swap
- 8-core Intel(R) Xeon(R) CPU E5-2407 0 @ 2.20GHz
- ext4 filesystem hold on a hardware Dell PERC H710 RAID10 with 4x4TB SATA HDs.
- 2 GB of RAM are reserved for
