Postgres Pro
Downloads
Home   >   mailing lists

Thread: COMMIT in ps output

COMMIT in ps output

From
Dr NoName
Date:
Hi all,

While investigating the causes of the deadlock I
described previously, we noticed that ps output would
often show some postgresql processes doing COMMIT. The
developer in charge of the application(*) assures me
that they are not using transactions (or at least not
in any of the code he checked). My question is, would
postgresql ps string show COMMIT at the end of
INSERT/UPDATE even if explicit transactions are not
used?

(*)I am posting this on behalf of another department,
so I am not entirely familiar with all the crap they
have writing to the database (and, evidently, neither
are they ;-)). I was called in as a resident
postgresql guru. But these problems are so weird that
I need to turn to the experts.

At one point, they had a problem with pqxx library. It
would create pqxxlog_condor table and write some crap
there. This caused a huge performance problem, so
eventually they disabled whatever option was causing
it. However, yesterday I noticed pqxxlog_condor table
in the database. It was empty, so I dropped it. I
checked this morning and it's back again, also empty.
Could COMMITs be coming from the library?


thanks in advance,

Eugene



____________________________________________________
Start your day with Yahoo! - make it your home page
http://www.yahoo.com/r/hs

Re: COMMIT in ps output

From
Tom Lane
Date:
Dr NoName <spamacct11@yahoo.com> writes:
> ... My question is, would
> postgresql ps string show COMMIT at the end of
> INSERT/UPDATE even if explicit transactions are not
> used?

No.

            regards, tom lane

Re: COMMIT in ps output

From
Bruce Momjian
Date:
Tom Lane wrote:
> Dr NoName <spamacct11@yahoo.com> writes:
> > ... My question is, would
> > postgresql ps string show COMMIT at the end of
> > INSERT/UPDATE even if explicit transactions are not
> > used?
>
> No.

My guess is that his interface library is doing it.

--
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073

Re: COMMIT in ps output

From
Dr NoName
Date:
yep. looks like it was the library.

thanks,

Eugene


--- Bruce Momjian <pgman@candle.pha.pa.us> wrote:

> Tom Lane wrote:
> > Dr NoName <spamacct11@yahoo.com> writes:
> > > ... My question is, would
> > > postgresql ps string show COMMIT at the end of
> > > INSERT/UPDATE even if explicit transactions are
> not
> > > used?
> >
> > No.
>
> My guess is that his interface library is doing it.
>
> --
>   Bruce Momjian                        |
> http://candle.pha.pa.us
>   pgman@candle.pha.pa.us               |  (610)
> 359-1001
>   +  If your life is a hard drive,     |  13 Roberts
> Road
>   +  Christ can be your backup.        |  Newtown
> Square, Pennsylvania 19073
>
> ---------------------------(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
>


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com