Oracle FAQ Your Portal to the Oracle Knowledge Grid

Home -> Community -> Mailing Lists -> Oracle-L -> RE: Hosing of Data Dict by "catalog.sql"

RE: Hosing of Data Dict by "catalog.sql"

From: Koivu, Lisa <>
Date: Tue, 30 Oct 2001 06:59:31 -0800
Message-ID: <>

Hi James,

Been there, done that...

As long as your database names are not hosed (check v$database) and you haven't lost any data you may be OK.  Try running catalog.sql in it's entirety.  If it doesn't fix your problem, restore from backup.

Good luck.

Lisa Koivu
Oracle Database Monkey
Fairfield Resorts, Inc.

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

From:   James Damiano []
Sent:   Tuesday, October 30, 2001 9:15 AM
To:     Multiple recipients of list ORACLE-L
Subject:        Hosing of Data Dict by "catalog.sql"

Greetings fellow DBAs,

I have one small Oracle8i (8.1.6) database (noarchivelog mode) installation on a Windows/NT 4 box to which I believe I have done some serious damage. Long story short - I inadvertently kicked off the build script for the database which lives there.

    The first thing the script did was an "oradim", trying to recreate the service which was already running.  I don't think that caused any problems but I haven't been able to locate any log of the event to indicate one way or the other.

    The second thing it did was try to "startup nomount" and "create database" which of course outright failed - no problem.

    But, the third script which fired caused the problem.  After a series of "create tablespace" and etc. failures, the "catalog.sql" script started up. It was in the process of reading in the first "create package" statement (consisting of about 2500 text lines) in preparation for executing when I cut it off.  I thought I had gotten it before it even had read through to the end of that "create package".  But apparantly not so.  It must have executed some of the statements within that definition (e.g. "create package body", etc.), because though everything looked fine thereafter, I tried to do a full database export and received the following result:

. . exporting table                       WORKSTOP
                                                            0 rows exported

. about to export ADAM's tables via Conventional Path ...
. exporting referential integrity constraints
. exporting synonyms
. exporting views
. exporting stored procedures
. exporting operators
. exporting indextypes
. exporting bitmap, functional and extensible indexes
. exporting posttables actions

EXP-00008: ORACLE error 4045 encountered ORA-04045: errors during recompilation/revalidation of SYS.DBMS_AQ_IMPORT_INTERNAL
ORA-06552: PL/SQL: Compilation unit analysis terminated
ORA-06553: PLS-905: object SYS.AQ$_SUBSCRIBERS is invalid
ORA-06508: PL/SQL: could not find program unit being called
ORA-06512: at line 1
ORA-06512: at "SYS.DBMS_SYS_SQL", line 1076
ORA-06512: at "SYS.DBMS_SQL", line 316
ORA-06512: at "SYS.DBMS_EXPORT_EXTENSION", line 84
ORA-06512: at "SYS.DBMS_EXPORT_EXTENSION", line 135
ORA-06512: at line 1
EXP-00000: Export terminated unsuccessfully

In my mind, I have the following options:

  (1) Try to just execute the full "catalog.sql" script again and             see if that clears up the problem.

  (2) Export just the users, their schemas and data, recreate             the database and import the users again.

  (3) Restore the datafiles from a week-old cold backup,
            drop the users, then import the same kind of export
            I mentioned in #2.

Any thoughts or opinions would be greatly appreciated.  Thanks in advance.

Jim Damiano

Please see the official ORACLE-L FAQ:
Author: James Damiano

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
To REMOVE yourself from this mailing list, send an E-Mail message
to: (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).
Received on Tue Oct 30 2001 - 08:59:31 CST

Original text of this message