Re: EM12c reporting two sessions with same SID

From: Alex Fatkulin <afatkulin_at_gmail.com>
Date: Wed, 22 Aug 2012 13:58:12 -0400
Message-ID: <CAMVw97+0CkQYKxEtDpwBiGquP8q+ekuHQ0_sVgBZmuyf0rZNbQ_at_mail.gmail.com>

It is in fact impossible to get sessions with the same SID at the same time on the same instance (at the same time is the important part!). However, it's pretty easy to get into the situation that you describe because you can have any number of sessions connected/disconnected during the snap period and all these sessions can have the same SID because they all existed at a different points in time.

This is why v$session has a serial# column which can be used to distinguish sessions with the same SID, however, (SID, SERIAL#) is only unique for the duration of instance's lifetime. If you need a truly unique identifier then AUDSID is what you will be looking for (minus a couple of special cases where AUDSID equals 0 like for SYS sessions, etc.).

On Wed, Aug 22, 2012 at 9:44 AM, Rich Jesse <rjoralist2_at_society.servebeer.com> wrote:
> Hey all,
>
> While I'm trying to debug some messes, I see that EM12c's Top Activity
> historical data shows at least one 30-minute time period from this morning
> where two of the Top Sessions have the same SID. Since I know the app
> generating these two sessions must do so serially -- one will finish and
> disconnect before the other starts -- could this really be the case? Can
> two different direct-connect (no pooling) sessions created within a "short"
> period of time have the same SID?
>
> Evidence is on 11.2.0.3, but I know I've also seen this on 10.1.0.5 (which
> is now thankfully gone!).
>
> TIA!
> Rich
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>

-- 
Alex Fatkulin,
http://afatkulin.blogspot.com

Enkitec,
http://www.enkitec.com
--
http://www.freelists.org/webpage/oracle-l

Received on Wed Aug 22 2012 - 12:58:12 CDT