Jim C. Nasby wrote:
> Moving to -hackers

You forgot to actually do it apparently?

Sorry about posting the patch to -general, BTW. Anyway it was committed
to the 8.1 branch, so it is included in the new release (8.1.4?)

> Does this still obey stats_command_string?

Yes.

I considered having the ps display show the info, but it's not as useful
because you can only get the info if you have access to the process list
(i.e. not a remote client).

> I think it'd be very handy to either have autovac always report what
> it's doing (ignoring stats_command_string), or to provide it with it's
> own option for it.
> I doubt this should pose a performance issue, since unless you have a
> whole lot of small tables autovac shouldn't be issuing commands very
> frequently.

Hmm. While I understand your point, I wouldn't want to clutter
postgresql.conf with an option just for this though; seems to be very
unimportant.

--
Alvaro Herrera http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support

---------------------------(end of broadcast)---------------------------
TIP 4: Have you searched our list archives?

http://archives.postgresql.org

On May 22, 2006, at 2:37 PM, Alvaro Herrera wrote:

> Jim C. Nasby wrote:
>> Moving to -hackers
>
> You forgot to actually do it apparently?
>
> Sorry about posting the patch to -general, BTW. Anyway it was
> committed
> to the 8.1 branch, so it is included in the new release (8.1.4?)
>
>> Does this still obey stats_command_string?
>
> Yes.
>
> I considered having the ps display show the info, but it's not as
> useful
> because you can only get the info if you have access to the process
> list
> (i.e. not a remote client).

It would be useful for dba's watching the box directly, via ps or top
(which I find myself doing fairly often). This has always been a
great feature of postgres IMO. I'd put in a vote for having that for
that for the autovac backend as well FWIW.

In any event thanks a lot for the current fix, as is it's a big
improvement! 8^)

-Casey


---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match

On Mon, May 22, 2006 at 02:45:30PM -0700, Casey Duncan wrote:
>
> On May 22, 2006, at 2:37 PM, Alvaro Herrera wrote:
>
> >Jim C. Nasby wrote:
> >>Moving to -hackers
> >
> >You forgot to actually do it apparently?

Yup, I are SMRT.

> >Sorry about posting the patch to -general, BTW. Anyway it was
> >committed
> >to the 8.1 branch, so it is included in the new release (8.1.4?)
> >
> >>Does this still obey stats_command_string?
> >
> >Yes.
> >
> >I considered having the ps display show the info, but it's not as
> >useful
> >because you can only get the info if you have access to the process
> >list
> >(i.e. not a remote client).

Well, there's now been 2 calls for seperate options for the autovacuum
process; this one and the ability to give it a different log level.
Perhaps what would be best is having autovac read a second set of config
options that would over-ride settings in the main postgresql.conf. This
could be a seperate GUC, or perhaps a seperate config file (which would
allow for removing all the autovac controls from postgresql.conf).

> In any event thanks a lot for the current fix, as is it's a big
> improvement! 8^)

Ditto.
--
Jim C. Nasby, Sr. Engineering Consultant jnasby@pervasive.com
Pervasive Software http://pervasive.com work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461

---------------------------(end of broadcast)---------------------------
TIP 1: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo@postgresql.org so that your
message can get through to the mailing list cleanly