Veritas-bu

[Veritas-bu] Restore of Oracle Dump Files Failing

2002-06-13 14:20:05
Subject: [Veritas-bu] Restore of Oracle Dump Files Failing
From: Paul_Yahnig AT prnewswire DOT com (Paul_Yahnig AT prnewswire DOT com)
Date: Thu, 13 Jun 2002 14:20:05 -0400
--0__=42ad2cmgJI8OgEG26A3CaQbKuHQkTmTr4E2PePHgxpm1flUFYYl0JITt
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline



Hello All.  I would like to say thank you at the start of this email for all of
the help.

I am running NBU 3.4 DC under Solaris 7, with a SpectraLogics 64000 that has 8
AIT2 TapeDrives.

We are in the process of re-building a Oracle Database Server that we backup via
Oracle Dump files and when I do the restore it is only getting a partial
completion.  It gives a media error.
To be safe I tried a different system that does the same type of Oracle Backup 
and
I have the Exact same problem.

I have been going around the web and newsgroups finding different information 
from
AIT strings in st.conf to fix vs. variable block writes.

I have put at the bottom of this email the errors that I have received in my
/var/adm/messages file and also the info that I have found.  I am hoping that
someone out there has had this problem and can help me shed some light on what
path I should take.


/var/adm/messages errors:
Jun 13 11:04:57 hsvnetbkup unix: /pci@1f,2000/scsi@1 (glm2):
Jun 13 11:04:57 hsvnetbkup      Cmd (0x24bcac0) dump for Target 6 Lun 0:
Jun 13 11:04:57 hsvnetbkup unix: /pci@1f,2000/scsi@1 (glm2):
Jun 13 11:04:57 hsvnetbkup              cdb=[ 0x8 0x0 0x0 0x80 0x0 0x0 ]
Jun 13 11:04:57 hsvnetbkup unix: /pci@1f,2000/scsi@1 (glm2):
Jun 13 11:04:57 hsvnetbkup      pkt_flags=0x0 pkt_statistics=0x61 pkt_state=0x7
Jun 13 11:04:57 hsvnetbkup unix: /pci@1f,2000/scsi@1 (glm2):
Jun 13 11:04:57 hsvnetbkup      pkt_scbp=0x0 cmd_flags=0x8e1
Jun 13 11:04:57 hsvnetbkup unix: WARNING: /pci@1f,2000/scsi@1 (glm2):
Jun 13 11:04:57 hsvnetbkup      Disconnected command timeout for Target 6.0
Jun 13 11:04:57 hsvnetbkup unix: WARNING: ID[SUNWpd.glm.cmd_timeout.6016]
Jun 13 11:04:57 hsvnetbkup unix: WARNING: /pci@1f,2000/scsi@1/st@6,0 (st20):
Jun 13 11:04:57 hsvnetbkup      SCSI transport failed: reason 'timeout': giving 
up

Jun 13 11:25:07 hsvnetbkup unix: /pci@6,2000/scsi@1,1 (glm7):
Jun 13 11:25:07 hsvnetbkup      Cmd (0x146d508) dump for Target 6 Lun 0:
Jun 13 11:25:07 hsvnetbkup unix: /pci@6,2000/scsi@1,1 (glm7):
Jun 13 11:25:07 hsvnetbkup              cdb=[ 0x8 0x0 0x0 0x80 0x0 0x0 ]
Jun 13 11:25:07 hsvnetbkup unix: /pci@6,2000/scsi@1,1 (glm7):
Jun 13 11:25:07 hsvnetbkup      pkt_flags=0x0 pkt_statistics=0x61 pkt_state=0x7
Jun 13 11:25:07 hsvnetbkup unix: /pci@6,2000/scsi@1,1 (glm7):
Jun 13 11:25:07 hsvnetbkup      pkt_scbp=0x0 cmd_flags=0x8e1
Jun 13 11:25:07 hsvnetbkup unix: WARNING: /pci@6,2000/scsi@1,1 (glm7):
Jun 13 11:25:07 hsvnetbkup      Disconnected command timeout for Target 6.0
Jun 13 11:25:07 hsvnetbkup unix: WARNING: ID[SUNWpd.glm.cmd_timeout.6016]
Jun 13 11:25:07 hsvnetbkup unix: WARNING: /pci@6,2000/scsi@1,1/st@6,0 (st55):
Jun 13 11:25:07 hsvnetbkup      SCSI transport failed: reason 'timeout': giving 
up


Current st.conf:

"SONY    SDK-500C",             "SONY AIT-2 8mm", "SONY_AIT";
SONY_AIT        =       1,0x34,0,0x3d639,1,0x30,0;

WebPage Info for AIT on E450 st.conf file:

        "SONY SDX-500C", "Sony AIT2 8mm", "SONY_TAPE";


        SONY_TAPE = 1,0x34,0,0x1639,4,0x00,0x00,0x00,0x00,3;



Info from Newsgroup:
Your tapes had been written to with a fixed block, NetBackup expects =
Variable blocks, Solaris 2.6 by default uses fixed blocks on tapes, thus =
by editing the st.conf you have now forced it into using variable, so it =
doesn't recognize the tapes. =20

If you change your st.conf back and boot -r then the error should change =
back.

At any rate you need to get the touch file that makes NetBackup use =
Fixed Blocks, and I know it exists, you just need to ask the support =
guys to escalate the call, the Media Manager guru's know about it, that =
should fix your problem for the time being, however you need to set that =
for restores, freeze all your tapes till they expire, remove it before =
backups and bplabel your tapes as they come off retention. =20

All while using the new updated st.conf.

Bob


URL:  http://mailman.eng.auburn.edu/pipermail/veritas-bu/2001-July/004922.html


Thank you again for any help,
Paul Yahnig


Paul Yahnig
Sr. Systems Administrator
PR Newswire
Tel: (201) 369-5022
Fax: (201) 369-0544
(Embedded image moved to file: pic09836.gif)
C. Plinius Caecilius Secundus - "Difficile est tenere quae acceperis nisi
exerceas"














_______________________________________________________________________________________________


Disclaimer:

Any views or opinions are solely those of the
author and do not necessarily represent those
of PR Newswire. The contents are intended
only for the addressee and may contain confidential
and/or privileged material. If you are not the
intended recipient, please do not read, copy,
use or disclose this communication and notify
the sender.

--0__=42ad2cmgJI8OgEG26A3CaQbKuHQkTmTr4E2PePHgxpm1flUFYYl0JITt--


<Prev in Thread] Current Thread [Next in Thread>
  • [Veritas-bu] Restore of Oracle Dump Files Failing, Paul_Yahnig AT prnewswire DOT com <=