Re: Track query status - Mailing list pgsql-sql

From Rajesh Kumar Mallah
Subject Re: Track query status
Date
Msg-id a97c77030703291248h1c1e0ab3h37e96b9775bc16d7@mail.gmail.com
Whole thread Raw
In response to Re: Track query status  (Andrew Sullivan <ajs@crankycanuck.ca>)
List pgsql-sql
more methods:

1. select * from pg_stat_activity;
see what all running and pid of your current query also.

2. SELECT h.pid AS blocker, w.pid AS blockee  FROM ONLY pg_locks h, ONLY pg_locks w WHERE h."granted" AND NOT
w."granted"AND (h.relation = w.relation
 
AND h."database" = w."database" OR h."transaction" = w."transaction");

make sure that the pid in 1 is not listed as a blockee in result of
above query m, ie The the update is waiting for anything.

3. go the shell , su - postgres ,

strace -p <pid> of the backend got in 1.
observe the system calls , see if anything awkward
there shoud be a lot of writes

4. iostat ofcourse, unless the above sql is the only sql(update)  running running in server .

my 4cents
regds
mallah.

On 3/29/07, Andrew Sullivan <ajs@crankycanuck.ca> wrote:
> On Wed, Mar 28, 2007 at 04:59:24PM -0400, Sumeet wrote:
> > Hi all,
> >
> > I have ran a update query on a dataset which has about 48 million records
> > and the query is already running for the third day....im so tempted to the
> > kill this query now....is there a way to know if the query is running??
> > here is the query i've ran
>
> 48 million records is a lot.  You oughta see activity with iostat or
> something.
>
> A
>
> --
> Andrew Sullivan  | ajs@crankycanuck.ca
> When my information changes, I alter my conclusions.  What do you do sir?
>                 --attr. John Maynard Keynes
>
> ---------------------------(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
>


pgsql-sql by date:

Previous
From: "Rajesh Kumar Mallah"
Date:
Subject: Re: Empty Table
Next
From: "Jon Horsman"
Date:
Subject: Re: Foreign Unique Constraint