Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have just installed a SAS disk into a Debian server. It was detected correctly and everything was fine.

Then I moved the SAS disk to a different Debian server, the same hardware model and running same version of Debian, but here the SAS disk is detected as /dev/sg7 and not /dev/sdb.

smartctl -a /dev/sg7 works fine, but fdisk and cat hang.

I tried putting the SAS disk in another slot: Same problem.

How can I force the SAS disk to be detected as /dev/sdb?

# uname -a
Linux maxwell 3.2.0-4-amd64 #1 SMP Debian 3.2.41-2+deb7u2 x86_64 GNU/Linux

dmesg says:

[171458.833581] scsi 0:1:5:0: Direct-Access     SEAGATE  ST91000640SS     0004 PQ: 0 ANSI: 6
[171458.833795] scsi 0:1:5:0: Attached scsi generic sg7 type 0
share|improve this question
Take a look at your loaded modules and udev rules. – Falcon Momot Jun 25 '13 at 8:35
Good idea. But alas no cigar. The working server has a subset of the modules in the failing server. udev only differs by mac-addresses. – Ole Tange Jun 25 '13 at 10:33
sg7 is the SCSI generic device, and everything you plug in gets one, whether it gets another device name or not. So that's kind of irrelevant. Check dmesg after you plug in the drive to see what's going on. – Michael Hampton Jun 25 '13 at 15:16
What SAS controller is in the new server? If it is a RAID controller, it might need some configuration (even if there was a single-disk array created on the old server with the same controller model, some kind of “import a foreign array” action may be needed to make the disk accessible on another controller). – Sergey Vlasov Jun 27 '13 at 11:28

The two servers are the same model. They have the same modules loaded, but they are not using the same modules.

lspci shows that the two servers of the same model use very different RAID controllers (I hate it when vendors do that: F*cking change the model number if it is not the same model!).

On the server, where the disk works, you do not need to setup anything to access a disk that is not in a RAID. But on the server, where the disk does not work, you need to set the disk up as a volume in the RAID controller before Linux detects it.

So I have now done that, and the disk is now accessible.

Thanks to Falcon Momot and Sergey Vlasov for pointing me in the right direction.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.