Backup Central

Content-Transfer-Encoding: quoted-printable

Hi,

I am using Netbackup 4.5 in a Windows2000 environment. There is the
following SSO configuration:

An ADIC Scalar1000 with 24 AIT-3 drives attached to 6 SNCs (storage
network controllers - fibrechannel-to-scsi bridges).

The SNCs are attached to Brocade Silkworm 3800 switches.

The 3 Netbackup media servers are also attached to the Silkworm 3800
switches (using Qlogic HBAs).

=20

The problem I am having is that during a backup, the media servers will
lose connection to some of the drives (you actually get the
plug-and-play message on the server that a device was disconnected).
Sometimes it is just one of the drives on an SNC, and sometimes it is
the whole SNC. We can cycle everything (including the SNCs) and
everything will work ok for a day or two and then the same thing happens
again. It is not limited to a specific media server, drive or SNC.

=20

Has anyone seen this problem before? Any help is greatly appreciated.

Content-Transfer-Encoding: quoted-printable

I had the same thing on some Brocade switches. If I am not mistaken the
field engineer had to set the switches to not use the auto-detect. It
was re-targeting the luns which in turn would cause NBU to down the
drive.

=20

________________________________

From: veritas-bu-admin < at > mailman.eng.auburn.edu
[mailto:veritas-bu-admin < at > mailman.eng.auburn.edu] On Behalf Of Sanders,
Donna
Sent: Tuesday, January 25, 2005 3:18 PM
To: veritas-bu < at > mailman.eng.auburn.edu
Subject: [Veritas-bu] netbackup media servers losing connections to
drives through SNCs

=20

Hi,

I am using Netbackup 4.5 in a Windows2000 environment. There is the
following SSO configuration:

An ADIC Scalar1000 with 24 AIT-3 drives attached to 6 SNCs (storage
network controllers - fibrechannel-to-scsi bridges).

The SNCs are attached to Brocade Silkworm 3800 switches.

The 3 Netbackup media servers are also attached to the Silkworm 3800
switches (using Qlogic HBAs).

=20

The problem I am having is that during a backup, the media servers will
lose connection to some of the drives (you actually get the
plug-and-play message on the server that a device was disconnected).
Sometimes it is just one of the drives on an SNC, and sometimes it is
the whole SNC. We can cycle everything (including the SNCs) and
everything will work ok for a day or two and then the same thing happens
again. It is not limited to a specific media server, drive or SNC.

=20

Has anyone seen this problem before? Any help is greatly appreciated.

Set your HBA's to "Persistent Binding". This will prevent re-targeting of the luns.

Bobby Williams
Sr. Systems Engineer
4250-A Benton Drive
Chattanooga, Tennessee 37406
Pager: 423-819-3336
Phone: 423-624-2600 x122


________________________________

From: veritas-bu-admin < at > mailman.eng.auburn.edu on behalf of Zermeno, Brandon (Contractor)
Sent: Tue 1/25/2005 3:59 PM
To: Sanders, Donna; veritas-bu < at > mailman.eng.auburn.edu
Subject: RE: [Veritas-bu] netbackup media servers losing connections to drives through SNCs



I had the same thing on some Brocade switches. If I am not mistaken the field engineer had to set the switches to not use the auto-detect. It was re-targeting the luns which in turn would cause NBU to down the drive.



________________________________

From: veritas-bu-admin < at > mailman.eng.auburn.edu [mailto:veritas-bu-admin < at > mailman.eng.auburn.edu] On Behalf Of Sanders, Donna
Sent: Tuesday, January 25, 2005 3:18 PM
To: veritas-bu < at > mailman.eng.auburn.edu
Subject: [Veritas-bu] netbackup media servers losing connections to drives through SNCs



Hi,

I am using Netbackup 4.5 in a Windows2000 environment. There is the following SSO configuration:

An ADIC Scalar1000 with 24 AIT-3 drives attached to 6 SNCs (storage network controllers - fibrechannel-to-scsi bridges).

The SNCs are attached to Brocade Silkworm 3800 switches.

The 3 Netbackup media servers are also attached to the Silkworm 3800 switches (using Qlogic HBAs).



The problem I am having is that during a backup, the media servers will lose connection to some of the drives (you actually get the plug-and-play message on the server that a device was disconnected). Sometimes it is just one of the drives on an SNC, and sometimes it is the whole SNC. We can cycle everything (including the SNCs) and everything will work ok for a day or two and then the same thing happens again. It is not limited to a specific media server, drive or SNC.



Has anyone seen this problem before? Any help is greatly appreciated.