I upgraded my 9i to 10g and then moved it to another machine that was
already running 10g with a starter database, which I replaced with the
upgraded one. I followed all the steps, but got errors on the ALTER
DATABASE OPEN RESETLOGS. The database has been opening, but I can't get
it to stay up -- keeps crapping out about every 20-30 minutes with the
following error (SMON creates it): ORA-0600 [4194].

I look in the alert trace file and find that the smon is erroring out.
Error was NO VALID LOG MEMBER SEQ #35 FOR THREAD 1!

I shutdown the database (but that errored out) and finally when it had
killed itself, I opened it up STARTUP MOUNT. Then I did a
DATABASE RECOVER USING BACKUP CONTROLFILE UNTIL CANCEL
And gave it the sequence it was erroring on.

That was accepted with MEDIA RECOVERY COMPLETE. So I ran the ALTER
DATABASE OPEN RESETLOGS again. I got DATABASE ALTERED which I thought
meant everything was now fixed and good. No -- when I went to shutdown
the database I got the same error again.

I have run the ALTER DATABASE CLEAR UNARCHIVED LOG GROUP # for all
my logs. I can then shutdown the database 1 time, before I get the
errors again. I don't get the errors on STARTUP. Alert log just shows
problems when trying to write to the redo logs. But I recreated all of
them and dropped all the old ones.

Can anyone help?

Do I need to go back to the start with moving the db from the other
machine????

Thanks in advance,
Barb

Mrs. Barbara E. Parker
Helpdesk Manager
Assistant IAM
Comm: 850-452-0350
DSN: 922-0350
barbara.e.[Email address protected]





Oracle LazyDBA home page