Welcome to Microsoft's inability to handle SANs!

Seriously, Windows has no facility to persistently bind devices and
definitions on a fiber-connected device. That's what you're
seeing--whenever there is a discontinuity between a Windows server and
its fiber-connected devices, there may (probably will) be a renaming.

My understanding is that the TSM 5.3 server suffers much less from this
issue.

--
Mark Stapleton (stapleton AT berbee DOT com)
IBM Certified Advanced Deployment Professional
  Tivoli Storage Management Solutions 2005
IBM Certified Advanced Technical Expert (CATE) AIX
Office 262.521.5627

 

>-----Original Message-----
>From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On 
>Behalf Of Prather, Wanda
>Sent: Friday, June 10, 2005 2:10 PM
>To: ADSM-L AT VM.MARIST DOT EDU
>Subject: Windows and device addressing for TSM (Make it stop!!!)
>
>Could use some help from any Windows gurus out there:
>HOW do I keep Windows from changing the addresses of the TSM devices?
>
>This server is Win2K3 (but the same thing has occurred on 
>Win2K), TSM is 5.2.2.5.
>
>There are fibre connections from 2 HBA's to 2 different SAN switches.
>1 Fibre cable from each SAN switch to 1 LTO drive in a 3583 
>(that way if we lose a switch, we can still access at least 1 drive).
>
>Yesterday in order to clean up a big tangle of fibre cables, 
>the Windows admin disconnected 1 fibre cable from the HBA, 
>untangled things and plugged the cable back into the HBA, 
>EXACTLY the same connection.  Repeat for the other cable.
>
>Windows RENUMBERED the library and tape drives  (For example, 
>mt2.0.0.2 became mt1.0.0.2).  
>I saw the same thing once after a power outage, when the 
>server came up before the switch was powered up. 
>
>Now, I know that to get TSM working again, all I have to do is 
>install the LTO drivers for these "new" devices, and update 
>the TSM path for each device.
>
>But I would really like to understand WHY this happens, and 
>how we can avoid it.  I would like to go on vacation one day, 
>without worrying that this should happen while I'm gone!
>
>Any insight appreciated!
>
>Wanda Prather
>"* I/O, I/O, It's all about I/O *"  -(me)
> 
>