Re: [ADSM-L] Problems with Oracle TDP

Well, I knew about the object not found part, what was bugging me was
why it was getting to that.  However, I opened a call with IBM to find
out, and found this now closed note:
"http://www-01.ibm.com/support/docview.wss?rs=0&q1=IC68708&uid=swg1IC687
08&loc=en_US&cs=utf-8&cc=us&lang=en"
It appears that prior to the 6.2.2 BA Client (and API by extension) if
the transition from disk pool to tape pool exceeded the maximum allowed
mount points the transaction would be aborted with no entry in the log
as to why, and no retry.  So, I'm hoping this was the problem, as we
have increased the pool and bumped up the maxnummp setting by 1 to give
it that extra one in case the disk pool gets full.  We'll see over the
weekend.


See Ya'
Howard Coles Jr., RHCE, CNE, CDE
John 3:16!

-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
Stackwick, Stephen
Sent: Thursday, August 12, 2010 9:27 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: [ADSM-L] Problems with Oracle TDP

Howard,

Dunno about the others, but the "object not found" errors are simply the
result of the control-file autobackup. RMAN apparently checks for a
control file backup of the same name (thus the "object not found")
before doing the control file backup. You can ignore those errors.

Steve

Stephen J. Stackwick  |  Senior Consultant  |  410.539.1135  |
sstackwick AT icfi DOT com  |  www.jasi.com
ICF Jacob & Sundstrom  |  401 E. Pratt St, Suite 2214, Baltimore, MD
21202  |  301.518.6352 (m)

________________________________________
From: ADSM: Dist Stor Manager [ADSM-L AT VM.MARIST DOT EDU] on behalf of Howard
Coles [Howard.Coles AT ARDENTHEALTH DOT COM]
Sent: Wednesday, August 11, 2010 15:21
To: ADSM-L AT VM.MARIST DOT EDU
Subject: [ADSM-L] Problems with Oracle TDP

Gents, I'm having some weirdness with the TDP that's driving me crazy
because my DBQ is wearing me out about it.

The problem is that we're seeing errors that a transaction failed,
mainly with the "Requested data was not found" errors.  My contention is
that these jobs are getting retried, as was the case in times past.
However, I don't see that in the logs in later versions of the TDP.



Here are some snips from the logs:
>From dsmerror.log:

08/07/10   23:00:31 ANS0278S The transaction will be aborted.

08/07/10   23:00:31 ANS1309I The requested data is offline.

08/07/10   23:00:33 ANS1303E Client ended transaction



-- These are warnings, unlike the ones on 8/7 and 8/10

08/08/10   23:58:10 ANS4992W TDP Oracle AIX ANU0599 TDP for Oracle:
(21884988):

=>(tulhci01-oracle) ANU2604W The object /tulhci01_hcidb/
/c-506694242-20100808-06 was not found on the TSM Server



08/09/10   00:10:44 ANS4992W TDP Oracle AIX ANU0599 TDP for Oracle:
(21266518):

=>(tulhci01-oracle) ANU2604W The object /tulhci01_hcidb/
/c-251359303-20100809-00 was not found on the TSM Server



08/10/10   23:00:42 ANS0278S The transaction will be aborted.

08/10/10   23:00:42 ANS1309I The requested data is offline.

08/10/10   23:02:07 ANS1303E Client ended transaction



>From tdpoerror.log:

08/07/10   23:00:31 TID<26927212> ==> ANS0278S (RC157)  The transaction
will be aborted.

08/07/10   23:00:31 TID<26927212> ==> ANS1309I (RC9)    The requested
data is offline.

08/07/10   23:00:31 TID<26927212> ==> ANU2539E sbtwrite2(): Error - buf
pointer is NULL.

08/07/10   23:00:31 TID<26927212> ==> ANU2539E sbtwrite2(): Error - buf
pointer is NULL.

08/07/10   23:00:31 TID<26927212> ==> ANU2539E sbtwrite2(): Error - buf
pointer is NULL.





08/10/10   23:00:42 TID<28192938> ==> ANS0278S (RC157)  The transaction
will be aborted.

08/10/10   23:00:42 TID<28192938> ==> ANS1309I (RC9)    The requested
data is offline.

08/10/10   23:00:42 TID<28192938> ==> ANU2539E sbtwrite2(): Error - buf
pointer is NULL.

08/10/10   23:00:42 TID<28192938> ==> ANU2539E sbtwrite2(): Error - buf
pointer is NULL.

08/10/10   23:00:42 TID<28192938> ==> ANU2539E sbtwrite2(): Error - buf
pointer is NULL.



Information from the Client/Server:

Client OS:  AIX 5.3.11

TSM BA Client: 6.2.1

TSM Server: 5.5.3

TSM TDP: 5.5.2



I found the doc about the "Requested Data not found" and I understand
what that means.  However, the TDP is reporting the backup completed,
(as shown by looking at the details of the filespaces), and this is
coming from a full backup.

My stance is that the backups are getting done, and that this
transaction is being retried, but the DBA doesn't see it that way, and
since I can't show it definitively in the log I need some help on this.





See Ya'

Howard Coles Jr.

RHCE, CDE, CNE

Sr. Systems Engineer

(615) 296-3416

John 3:16!