Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Missing archive in hot backup

Missing archive in hot backup

From: <nixond_at_my-deja.com>
Date: Mon, 02 Aug 1999 16:23:47 GMT
Message-ID: <7o4gm5$qpd$1@nnrp1.deja.com>


We are running into a picular situaton on a Oracle 7.3.4 database running on HP/UX 10.20. If anyone has any insights on to what is going on, I would be interested in your comments.

We are running into an occasional problem with our online(hot) backup. This script has run with very little problems for years and then all of a sudden on one system it started having this problem.

We do the usual steps of putting the tablespaces into backup mode, etc. At end we cycle through the redo logs to create the archives need for recovery we capture the last sequence number and check in v$log that it is archived and the status is inactive. We then build a list of via the "ls" command of all the present archive logs and copy then to our nline backup directory. Has worked for years with out a hitch, and now the very last archive occasionally doesn't show up in the list of archives to be copied. I put in some debugging statements and the timestamps in the dumps, alert.log and backup.log all show are all in sync with when the archive log should have been created.

My questions is, according to the Oracle Archiver Best Practice (Note: 45042.1) the ufs archive file is created before the controlfile and redo log status is changed. If this is correct, the file should exist before we ever create the list of archive files. Is this ccurate? Are there some other circumstances that may cause this file to be written or finished being written to before the file can be seen at the OS level?

Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't. Received on Mon Aug 02 1999 - 11:23:47 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US