Backup Central

Hi,
we have fixed the block size to 256K on the script /sbin/init.d/networker =

on the servers using this:
NSR=5FDEV=5FBLOCK=5FSIZE=5FSDLT=3D256;export NSR=5FDEV=5FBLOCK=5FSIZE=5FSDLT
on the device configuration under networker
on the oracle backup script and oracle restore script.
And we are sure the tape label is 256k (confirmed by a scanner command)

thanks,

tmj




"Girot, Jean-Pierre" <JGirot < at > FR.MTI.COM>=20
Envoy=E9 par : Legato NetWorker discussion <NETWORKER < at > LISTMAIL.TEMPLE.EDU>
14/12/2004 17:36
Veuillez r=E9pondre =E0
Legato NetWorker discussion <NETWORKER < at > LISTMAIL.TEMPLE.EDU>; Veuillez=20
r=E9pondre =E0
"Girot, Jean-Pierre" <JGirot < at > FR.MTI.COM>


Pour
NETWORKER < at > LISTMAIL.TEMPLE.EDU
cc

Objet
Re: [Networker] hp-ux 11.23 update 2 and networker 7.1.1 and nmo4.1






Hello,

Perhaps the other SAN Storage Node doesn't use the same block size

Kind Regards
Jean-Pierre=20

-----Message d'origine-----
De : therese-marie Jonnaert [mailto:tmjonnaert < at > COFIDIS.FR]=20
Envoy=E9 : mardi 14 d=E9cembre 2004 17:33
=C0 : NETWORKER < at > LISTMAIL.TEMPLE.EDU
Objet : [Networker] hp-ux 11.23 update 2 and networker 7.1.1 and nmo4.1

hi,
few days ago, I have posted a message speaking about oracle data base
restore using Rman and using a San.
To day, after many tests, we have detected that the problem is only using=20
a
new hp server unix version 11.23 update 2 (september=20
2004,11.23.04.09).When
we make a backup on the san with this dedicated storage node --> no error
But when we try to restore this database on the san,another dedicated
storage node (hp-us, 11.23), rman hang with this message:
ORA-19501: read error on file "nhg73lmf=5F1=5F1", blockno 1456897
(blocksize=3D1024)
ORA-27190: skgfrd: sbtread2 returned error
If we make the backup and the restore on the san with an hp-ux 11.23=20
server
all is ok.
We use oracle 9.2.0.5.0, networker 7.1.1
The hp server are 64bits itanium
No help from legato, no help from hp because we haven't error on the=20
system
logs
Have you an idea?
thanks

Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu
=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D=
*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D

Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu
=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D=
*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D


Note: To sign off this list, send a "signoff networker" command via email
should be sent to stan < at > temple.edu
=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D

This is not necessarily a legato error. I've had the same problem before=20
when I tried to read a backup created by rman using the blksize parameter=20
like the following=20

allocate channel t1 type 'SBT=5FTAPE'=20
parms=20
'ENV=3D(NSR=5FSERVER=3Dcmisnsr2,NSR=5FCLIENT=3Daddcsmdev,NSR=5FDATA=5FVOLUM=
E=5FPOOL=3Dadddev),BLKSIZE=3D524288';=20

Then, when you restore, you must ensure you use the same parms. If not,=20
it defaults I think to 65K and ends up not finding the files or filesets=20
that were backed up by oracle/rman. =20

Take a look at your rman backup scripts and see what they use.=20





therese-marie Jonnaert <tmjonnaert < at > COFIDIS.FR>
Sent by: Legato NetWorker discussion <NETWORKER < at > LISTMAIL.TEMPLE.EDU>
12/15/2004 02:56 AM
Please respond to Legato NetWorker discussion; Please respond to=20
tmjonnaert

=20
To: NETWORKER < at > LISTMAIL.TEMPLE.EDU
cc:=20
Subject: [Networker] R=E9f. : Re: [Networker] hp-ux 11.23 up=
date 2 and networker=20
7.1.1 and nmo4.1
hi,
few days ago, I have posted a message speaking about oracle data base
restore using Rman and using a San.
To day, after many tests, we have detected that the problem is only using=20
a
new hp server unix version 11.23 update 2 (september=20
2004,11.23.04.09).When
we make a backup on the san with this dedicated storage node --> no error
But when we try to restore this database on the san,another dedicated
storage node (hp-us, 11.23), rman hang with this message:
ORA-19501: read error on file "nhg73lmf=5F1=5F1", blockno 1456897
(blocksize=3D1024)
ORA-27190: skgfrd: sbtread2 returned error
If we make the backup and the restore on the san with an hp-ux 11.23=20
server
all is ok.
We use oracle 9.2.0.5.0, networker 7.1.1
The hp server are 64bits itanium
No help from legato, no help from hp because we haven't error on the=20
system
logs
Have you an idea?
thanks




--=_alternative 004C5F2A86256F6B_=