BRARCHIVE - strange behaviour

[symon_braunbaer]

Dear Experts,

I have been solving so many times a full oraarch situation. But what happened today is a little strange.

During the DB backup, there were 70 archive log files (which is not that much). The backup finished at

2:44. Then oraarch got full and the DB got stuck.

Around 8:40 I have started BRARCHIVE manually, which solved the problem.

The strange here is, that BRARCHIVE has been running at 3:44, 4:44, etc. and failing to archive ANY

logs with the following error message:

BR0280I BRARCHIVE time stamp: 2015-08-18 04.44.01

BR0301W SQL error -257 at location BrDbConnect-2, SQL statement:

'CONNECT /'

ORA-00257: archiver error. Connect internal only, until freed.

BR0310W Connect to database instance <SID> failed

There are no records in DB14 of these runs, but I think that when the DB is stuck, the corresponding

tables couldn't be updated.

My question: Why the automatic BRARCHIVE runs were failing and when I manually started exactly

the same command, it worked ???

Thanks a lot!

[symon_braunbaer]

Hi everybody,

I am SORRY !! Now I analyzed the situation again, and what I wrote in my initial post was completely wrong. So, here is the real deal:

BRBACKUP has been running until 02:44:43. BRARCHIVE was called at 02:44:06 and it said:

BR0280I BRARCHIVE time stamp: 2015-08-18 02.44.06

BR0037W Offline redolog files starting with file /oracle/<SID>/oraarch/<SID>arch1_276408_871557105.dbf are locked by BRBACKUP for an online consistent backup

When it was called for the next time, which was one hour later, it got STUCK without any entry in the log file, until 8:32h when I intervened:

BR0006I Start of offline redolog processing: aerewbsy.svd 2015-08-18 03.44.00

BR0484I BRARCHIVE log file: /oracle/<SID>/saparch/aerewbsy.svd

BR0280I BRARCHIVE time stamp: 2015-08-18 08.32.36

BR0261E BRARCHIVE cancelled by signal 15

Of course, all other calls were failing, because of the .lock file.

And indeed, I found a running process of BRARCHIVE from 03:44, which I killed, then I deleted the .lock file and I manually launched

BRARCHIVE once again, which then resolved the situation.

So, my question is, how to find out WHY the BRARCHIVE process got stuck, when the log file has no entry for those almost 5 hours...

[Former Member]

I think you are contradicting yourself

You only gave the brarchive logs

BR0280I BRARCHIVE time stamp: 2015-08-18 04.44.01

BR0301W SQL error -257 at location BrDbConnect-2, SQL statement:

'CONNECT /'

ORA-00257: archiver error. Connect internal only, until freed.

BR0310W Connect to database instance <SID> failed

According to which the scheduled job did run at 4:44:01 and cancelled due to oraarch being 100% used and now you figured out that the jobs did not run after 3:44 .

Thanks

Amit

[symon_braunbaer]

No, Amit. Just ignore whatever I wrote in my first post please.

The jobs have been triggered, but didn't do anything, as the BRARCHIVE launched at 3:44h was hanging until I killed it at 8:32h. I have absolutely no logs and no idea how to find out why was it hanging:

BR0006I Start of offline redolog processing: aerewbsy.svd 2015-08-18 03.44.00

BR0484I BRARCHIVE log file: /oracle/T07/saparch/aerewbsy.svd

BR0280I BRARCHIVE time stamp: 2015-08-18 08.32.36

BR0261E BRARCHIVE cancelled by signal 15

You can see the almost 5 hour gap in the log, which can't help to find out, why did the process hang.

[Former Member]

Hi Symon,

This can happen, if your database is already stuck or oraarch was full before the scheduled brarchive job could run via DB13 , it can fail with the error messages,

ORA-00257: archiver error. Connect internal only, until freed.  This simply means "The archiver process received an error while trying to archive a redo log "

Thanks

Amit

[symon_braunbaer]

Hi Amit,

yes, of course! But when I launched it at the command line it just worked...

Indeed I have had situations, where I had to move a few files, in order to get BRARCHIVE working, but in this case it just started archiving. It's quite strange...

[Former Member]

Hi Symon,

Yes agree with you. I had similar situation where I was able to run the brarchive from OS level although oraarch was 100% but the one scheduled through DB13 failed kept on failing.

391 - Archiver stuck

Thanks

Amit