Good news for bad databases...
I just closed a long running escalation whereby old oracle connections would appear to hang, and new ones could not be established. A truss of a new sqlplus session showed that the oracle process was sleeping on a semop system call.sqlplus/1: read(10, 0x10024FF66, 2064) (sleeping...)^M oracle/1: semop(1245185, 0xFFFFFFFF7FFFAF9C, 1) (sleeping...)^M sqlplus/1: 556.9328 Received signal #2, SIGINT, in read() ^^^Here we see the process is eventually killed with CTRL-CAnyhow the fix was to apply Oracle update 9.2.0.5 (9205) which according to my Oracle contacts contains around 1,100 fixes. 9.2.0.4 contained in the region of 800 compared to 9.2.0.3 Cor blimey, what a whopper . ( Jul 19 2004, 04:12:07 PM BST ) Permalink
Comments:
Post a Comment:
Comments are closed for this entry.

