RE: wierd locking issue

From: Powell, Mark D <mark.powell_at_eds.com>
Date: Mon, 25 Feb 2008 16:13:06 -0500
Message-ID: <D1DC33E67722D54A93F05F702C99E2A901FE19AE@usahm208.amer.corp.eds.com>

 

I thought someone posted an Oracle support document that the ddl_wait_for_locks parameter does not work in 10g as it was intended to be an 11g parameter, where I guess it actually works.

  • Mark D Powell -- Phone (313) 592-5148

-----Original Message-----

From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of ryan_gaffuri_at_comcast.net
Sent: Monday, February 25, 2008 12:29 PM To: oracle-l_at_freelists.org
Subject: wierd locking issue

10.2.0.3
Linux Redhat

I have code that adds and drops partitions. I am constantly getting this error. I am doing a few things to track this down and alleviate it.

  1. at the beginning of the execution of my package I issue: alter session set ddl_wait_for_locks=true;

now I thought that with this set to true, I would just wait when adding or dropping a partition instead of erroring out?

2. Inside my exception block, I run the following query and dbms_output the results to the screen:

  cursor cur_get_lock is
    select a.*,b.object_name
    from v$locked_object a, user_objects b     where a.object_id = b.object_id;

I get no rows returned and this is fires in the exception block which is triggered when the error is fired? So I would guess the lock is occurring in a fraction of a second and then ending.

any ideas?
--

http://www.freelists.org/webpage/oracle-l

--

http://www.freelists.org/webpage/oracle-l Received on Mon Feb 25 2008 - 15:13:06 CST

Original text of this message