scsi_id question

scsi_id question

[lemons_terry]

Hi

We've been observing the behavior of scsi_id, as a component of the udev machinery that creates persistent device names in the /dev/|generic|disk|tape/by-id directories.  When scsi_id is run against a tape device that has more than one useful item in the VPD pages, scsi_id selects one of these items as the basis for the name that is generated.  There is obviously some sort of hierarchy in play here.  For instance, if a device has both a WWNN and a serial number, scsi_id will use the WWNN.  That's a good choice, IMHO, given that modern tape libraries have WWNN assigned to the 'sled' that tape drives fit into  When a tape drive is replaced, the serial number changes but the 'sled' WWNN stays the same (so I'm told).

But if a device has both an IENN and a serial number, the IENN is used.  And that's a bad choice, because the IENN is not unique.  I have a tape library in my lab that has two tape drives with the same IENN and different serial numbers.  udev does not create a persistent device name for both of these tape drives.  This is, to me, a bug.

Can I/we get a peek at the hierarchy that scsi_id uses?  Methinks some fine-tuning is necessary.

Thanks!
tl

Terry Lemons
Backup Platforms Group
EMC² 
where information lives
4400 Computer Drive, MS D239
Westboro MA 01580
Phone: 508 898 7312
Email: Lemons_Terry@emc.com


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: scsi_id question

[Andrey Borzenkov]

[-- Attachment #1.1: Type: text/plain, Size: 1597 bytes --]

On Friday 02 November 2007, lemons_terry@emc.com wrote:
> Hi
>
> We've been observing the behavior of scsi_id, as a component of the udev
> machinery that creates persistent device names in the
> /dev/|generic|disk|tape/by-id directories.  When scsi_id is run against a
> tape device that has more than one useful item in the VPD pages, scsi_id
> selects one of these items as the basis for the name that is generated. 
> There is obviously some sort of hierarchy in play here.  For instance, if a
> device has both a WWNN and a serial number, scsi_id will use the WWNN. 
> That's a good choice, IMHO, given that modern tape libraries have WWNN
> assigned to the 'sled' that tape drives fit into  When a tape drive is
> replaced, the serial number changes but the 'sled' WWNN stays the same (so
> I'm told).
>
> But if a device has both an IENN and a serial number, the IENN is used. 
> And that's a bad choice, because the IENN is not unique. 

But is not duplicated EUI-64 actually a bug that should be reported to vendor?

> I have a tape 
> library in my lab that has two tape drives with the same IENN and different
> serial numbers.  udev does not create a persistent device name for both of
> these tape drives.  This is, to me, a bug.
>
> Can I/we get a peek at the hierarchy that scsi_id uses? 

Sure. In udev sources :)

> Methinks some 
> fine-tuning is necessary.
>

By default page 0x83 is tried first if available. You can force page 0x80 
(serial number) when calling scsi_id or per vendor or per model 
in /etc/scsi_id.config.

HTH

-andrey

[-- Attachment #1.2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

[-- Attachment #2: Type: text/plain, Size: 314 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

[-- Attachment #3: Type: text/plain, Size: 226 bytes --]

_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel