Re: Procedural Consistency

2002-09-27 Thread Jared . Still
Dan, The default is for oracle to determine this by timestamp. You may want to experiment with the REMOTE_DEPENDENCIES_MODE parameter. This may not apply to your situation as I don't know if it applies to dependencies on a local database as well. e.g. ALTER SESSION SET

RE: Procedural Consistency

2002-09-27 Thread Fink, Dan
Thanks, I'll take a look.. -Original Message- Sent: Friday, September 27, 2002 12:03 PM To: Multiple recipients of list ORACLE-L Dan, The default is for oracle to determine this by timestamp. You may want to experiment with the REMOTE_DEPENDENCIES_MODE parameter. This may not apply

RE: Procedural Consistency

2002-09-27 Thread Fink, Dan
The app was getting an ORA-600, which metalink identified as a 'bug' in 9i. The app uses persistent connections and the 'login' package had been recompiled without removing the connections. All of a sudden the ORA-600s started being thrown. It made me curious (call me George) as to how Oracle

Re: Procedural Consistency

2002-09-26 Thread Bill Buchan
What is the error and how is the error occuring? My experience has been: 1. Long running SQL calling PL/SQL function:will die due to invalidated state of the function. 2. Long running PL/SQL calling PL/SQL procedure (statically): cannot re-compile the procedure during the run as it is

Re: Procedural Consistency

2002-09-24 Thread Stephane Faroult
Fink, Dan wrote: Okay, I know I'm being a little lazy on this one, but I'm very interested to hear the ideas/conjecture/proof. So away we go We recently encountered a bug in Oracle where a long running process attempted to execute a procedure that was within a package that had been