Re: expdp error - Not Even Sure Where to Look - Found It
Date: Wed, 3 Sep 2014 12:29:42 -0500
Message-ID: <CAFH+ifdp4RSgr74xTsZmSnBOM+rDO53bSzTgYOCzhQkAXb7Usw_at_mail.gmail.com>
Thanks for all the replies. This is a third party application. The database was set up IAW their Database Server and Schema Setup Guide - Oracle 2013.
The document explicitly shows the following on page 3 in the section titled "Initialization Parameters Recommendation":
nls_date_format 'YYYY-MM-DD HH24:MI:SS'
nls_timestamp_format 'YYYY-MM-DD HH24:MI:SS:FF'
HOWEVER - being desperate I figured how many ways can one mess with the nls date format? How about a logon trigger?
Yep. In the scripts they executed to create the schemas, etc. there is this:
TRIGGER_TYPE TRIGGERING_EVENT
- ------------------------------
TRIGGER_BODY
AFTER EVENT LOGON BEGIN EXECUTE IMMEDIATE 'alter session set NLS_TIMESTAMP_FORMAT = ''YY-MM-DD HH24:MI:SS.FF''';
END ALTERSESS_TRIG; Notice the difference? Anyway, if I remove the initialization parameters, expdp works. Similarly, if I delete the triggers and leave the initialization parameters, expdp works. Guess they can't have it both ways. I'm waiting on the vendor to determine which is correct.
On Wed, Sep 3, 2014 at 11:54 AM, Freek D'Hooge <freek.dhooge_at_gmail.com> wrote:
> David,
>
> Checkout MOS note: How To Find The Tables From Data Dictonary Which Are
> Generating ORA-1801 While Running The Export? (Doc ID 1364321.1)
> It is not the exact same error, but it seems plausible to me.
>
>
> regards,
>
> Freek
>
>
>
> On wo, 2014-09-03 at 10:55 -0500, David Barbour wrote:
>
> RHEL 6.4 Oracle 11.2.0.3.o
>
>
> Had a request to export a schema. No problem. Set up a directory,
> granted rights, made sure directory ownership and permissions on the server
> were correct and issued the command:
>
> oracle_at_rchr1p04 P244 /tmp_mnt/P24_Dump $ expdp system/<password>_at_P244
> schemas=MSFPRD,TLMPRD directory=P24_Dump
> dumpfile=MSFPRD_TLMPRD_09022014.dmp logfile=MSFPRD_TLMPRD_09022014.log
>
>
> Got this back:
>
> Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 -
> 64bit Production
> With the Real Application Clusters and Automatic Storage Management
> options
> ORA-31626: job does not exist
> ORA-31637: cannot create job SYS_EXPORT_SCHEMA_01 for user SYSTEM
> ORA-06512: at "SYS.DBMS_SYS_ERROR", line 95
> ORA-06512: at "SYS.KUPV$FT", line 1569
> ORA-39062: error creating master process DM00
> ORA-39107: Master process DM00 violated startup protocol. Master error:
> ORA-01821: date format not recognized
>
>
> Huh?
>
>
>
> Alert log shows:
>
> DM00 started with pid=53, OS id=21595, job SYSTEM.SYS_EXPORT_SCHEMA_01
> opidrv aborting process DM00 ospid (21595) as a result of ORA-447
> Wed Sep 03 10:17:00 2014
> ALTER SYSTEM SET service_names='SYS$SYS.KUPC$
> S_4_20140903101659.P24.LENNOXINTL.COM','P244.lennoxintl.com' SCOPE=MEMORY
> SID='P244';
> ALTER SYSTEM SET service_names='P244.lennoxintl.com' SCOPE=MEMORY
> SID='P244';
>
>
> DM00 Trace file shows:
>
> *** 2014-09-03 10:16:59.659
> *** SESSION ID:(950.153) 2014-09-03 10:16:59.659
> *** CLIENT ID:() 2014-09-03 10:16:59.659
> *** SERVICE NAME:(SYS$USERS) 2014-09-03 10:16:59.659
> *** MODULE NAME:() 2014-09-03 10:16:59.659
> *** ACTION NAME:() 2014-09-03 10:16:59.659
>
> KUPP: Error 1821 detected in master process DM00
> OPIRIP: Uncaught error 447. Error stack:
> ORA-00447: fatal error in background process
> ORA-31666: Master process DM00 had an unhandled exception.
> ORA-01821: date format not recognized
>
>
> Any ideas?
>
> This actually happened yesterday as well and I've tried some different
> things with NLS and Googled the heck out of this and looked through MOS
> with no luck. Put in an SR with this information and they want the full
> command (was provided), the error (was provided) and the alert log contents
> (were provided). Not sure if they read the submission. Not sure why I
> even try that route any more.
>
>
-- http://www.freelists.org/webpage/oracle-lReceived on Wed Sep 03 2014 - 19:29:42 CEST