Re: db hang on shutdown

If this problem can be replicated easily (i.e it occurs quite often) then do the following: 1. Before doing a shutdown immediate connect to the db as "system" using a db monitoring tool

    [Platypus is FREE to download and register from http://www.websoft.com.au/downloads.html ]     Firstly, in the LeftHandSide [LHS] frame click on the "Database Settings".     You must change the default mode from "Read Only" = "Y" to "N"     The screen in Platypus to do this looks like http://www.websoft.com.au/images/puser_dbsetting1_shot.jpg 2. Current User Logins

    Click on the "Group" on the LHS tree and "Users Currently Logged In" Tab on the RHS.     The screen should look like this http://www.websoft.com.au/images/puser_userslogged1_shot.jpg     Note if there are any users logged on other than your platypus session.     If there ARE other users logged on then click on the row showing they logon to see:

1. What Objects they have opened and
2. What SQL they last executed in the Oracle SGA. You must now decide what steps you wish to take to log them out.
3. Find the user/process and close it normally or
4. Highlight the logon row and click on the "-" icon in Platypus to kill and terminate they session. Now refresh the screen by clicking "recource name" row on the Top RHS of the platypus screen to confirm that they have removed. 3. DBMS_JOBS running I know you said that you do not have any dbms_jobs defined in your db. However, please confirm this by: Clicking on the "DBMS_JOBS" Tab on the RHS. The screen should look like this http://www.websoft.com.au/images/puser_dbmsjob1_shot.jpg If there ARE jobs defined in the db then Platypus will tell you if they are currently running [a giff annimation of a runner is displayed in the RHS column called "Running" as in the screenshot example above Job#25 is running] At this stage you can now either:
5. "REMOVE" the job permanently from the job queue by highlighting the job on the RHS and clicking on the "-" icon. or
6. "DISABLE" (or Oracle uses the word BREAK" a job by highlighting the job and clicking on the job "STATUS" column to toggle it from a Green flag to a Red flag.

Please let me know if you still can't issue a shutdown immediate even after you have followed the above and confirmed that: a) No users logged on
b) No jobs running
Regards,
Earl.

Dusan Bolek wrote (on 15 Jan 2002 05:03:00):

> Ed_Stevens_at_nospam.noway.nohow (Ed Stevens) wrote in message news:<3c433fbd.7157151_at_ausnews.austin.ibm.com>...
>
> > Any takers to either question?
> > 1 - Why does this one db seem to hang on a shutdown immediate
> > 2 - Why does this one db show well over 100 sessions (some up to several days
> > old) when there are only six client machines in existence?
>
> I had exactly the same problem few years ago. Some session just hanged
> and had no intentions to die honorably. I had advantage that platform
> was HP-UX, so I used "kill -9" and everything was OK. These zombie
> sessions were not killable by internal Oracle tools.
> You can use timeout in listener.ora, or check if your applications are
> logging out correctly. However neither first or second helped me.
>
> --
> _________________________________________
>
> Dusan Bolek, Ing.
> Oracle team leader
>
> Note: pagesflames_at_usa.net has been cancelled due to changes (maybe we
> can call it an overture to bankruptcy) on that server. I'm still using
> this email to prevent SPAM. Maybe one day I will change it and have a
> proper mail even for news, but right now I can be reached by this
> email.
Received on Tue Jan 15 2002 - 23:12:06 CST