CASTalk.com

Eli Asulin · Guest

Hi all,

been around it for 2 weeks...hopeless.

I have 2 Win2000 SP4 configured as A/P Cluster attached to IBM SAN via FC
HBA.
I want to connect the servers to another storage (NetApp Filer) via iSCSI (1
Gbit NIC).

After installing the NIC, MS iSCSI initiator (v1.06) and Netapp SnapDrive
(v3.1) I moved all Cluster Resources to one node and make the iSCSI LUN
connection on the other.

An iSCSI session successfully created but when trying to do "Connect to
Virtual Disk" (SnapDrive snap-in), the operation is timed-out by error
message, although on the Netapp filer I can see a LUN Mapping was created.

At this point, the server will not reboot (hung) - only hard reset.
After the server start, I have to delete the LUN Mapping from the Netapp
filer inorder to successfully restart the server.

I've tried almost every thing I could think of, starting with newer versions
of iSCSI Initiator and SnapDrive, stop services on the server including
Cluster Service....

The only thing I didn't do is to uninstall IBM Drivers and Software for the
storage connection / management (SMClient, RDAC virtual Driver and SMAgent).
I did try to remove the FC HBA of the server and restart it...same thing.

Dose anyone know of conflict or incompatibility of any kind for the current
configuration?

Any help will appreciated

Eli.

Jan Hugo Prins · Guest

On Sun, 23 Oct 2005 11:44:05 +0200, Eli Asulin wrote:

Eric Bursley [MVP] · Guest

Part of the problem with iSCSI cluster is starting of the shared storage.
Typically shared storage is brought up before networking is, but in the case
of iSCSI, networking needs to be up first. Some storage vendors have
accomplished this.

Eric Bursley
eric at bursley dot net
Microsoft MVP
RHCE, MCSE, BCFP, EEIE-CS, ESCE-CS
GPG Signature:
Key fingerprint = CEAE CF3A 3876 7ECE 9DA7 946F DA9F DDCA C392 6DCB

"Jan Hugo Prins" <jhp@jhprins.org> wrote in message
news:pan.2005.10.24.15.17.11.312905@jhprins.org...

Jeff Goldner [MS] · Guest

you might want to try removing RDAC. It takes over more than it should, so
it could very likely be interfering with the iSCSI initiator (even if you
remove the FC HBA).

"Eli Asulin" <eli@gadotbio.com> wrote in message
news:u5eAuZ71FHA.664@tk2msftngp13.phx.gbl...

Kigall · Guest

I have built an SQL Cluster using 2x HP DL380's with Windows 2003 Enterprise
on and using iSCI initiators of 1.6 with the Qurorom, MSDTC and resources all
located on a NetApps FAS960 using LUN's so I know it is possible.

SnapDrive does not allow the use of FCP and iSCI on the same host system so
I'm not sure if that's the case with mixed Vendors using FCP and iSCSI so my
guess would be to use the Servers seperatley. I'm sure if you were to connect
up the Servers to the Filer and map the LUN's to one nose at a time before
bringing the 2nd node online it would work, alternatively you could add the
2nd node to the iGroup after the Cluster Service has taken ownership of the
disks as this is the same thing.

Another consideration is that if you are using Multistore (vFiler) from
NetApps that Snap Drive does not support connections to vFiler partitions,
only vFiler0 which is the hosting Filer itself.

Cheers,
Steve.

"Eli Asulin" wrote:

	CASTalk.com Forum Index -> Storage System	All times are GMT
Page 1 of 1