- Article Type: General
- Product: Aleph
- Product Version: 20
Even though the p_manage_102 log shows that the abc01 library was successfully locked:
Library UDM01 locked in param
In the following minute we see this:
aleph_admin@ALEPH20> 00:22:51 aleph_admin@ALEPH20> DROP TABLE UDM01.Z01
ERROR at line 1:
ORA-00054: resource busy and acquire with NOWAIT specified or timeout expired
and later in the log:
Sat Apr 14 01:22:25 EDT 2012 Creating Index z01_id5 ...
Sat Apr 14 01:22:29 EDT 2012 Error : Only 3 indexes out of 4 for table z01 were created
Sat Apr 14 01:22:29 EDT 2012 Error : index list for index creation was 'z01_id','z01_id2','z01_id3','z01_id5'
Sat Apr 14 01:22:29 EDT 2012 Error : Please check logs /exlibris/aleph/u20_1/abc01/scratch/create_ora_tables_z01_id*.log
aleph_admin@ALEPH20> 06:29:32 aleph_admin@ALEPH20> CREATE UNIQUE INDEX ABC01.Z01
_ID2 ON ABC01.Z01 (Z01_ACC_SEQUENCE ASC)
ERROR at line 1:
ORA-01452: cannot CREATE UNIQUE INDEX; duplicate keys found
The failure to drop/create the z01 meant that the existing z01's were not deleted. The job built and loaded new z01's, but there were thousands of duplicates and the z01_id2 -- which requires that the Z01_ACC_SEQUENCE's be Unique -- got the duplicate keys message.
The following line in the www_server_4991.log.1404.0505 log indicates that the www_server was still active at 00:22 after p_manage_102 was started:
2012-04-13 23:04:53 54  [vrb] server_main: OUT 0.1228 38213
2012-04-14 00:45:00 93  [vrb] IN 20120414 004500 ,<<<________
(If the www_server had been stopped, it would not have continued writing to this log. The log would have ended with an "exit" command at 2012-04-14 00:22 and we would have seen the first transaction from the www_server in an entirely *new* www_server log.)
KB 8192-7895 analyzes the "Resource busy" problem. It's very puzzling that the www_server (and pc_server) were not stopped.
A run of p_manage_102 the following night did not get the "resource busy" message and ran successfully.
- Article last edited: 10/8/2013