Tell me more ×
Unix & Linux Stack Exchange is a question and answer site for users of Linux, FreeBSD and other Un*x-like operating systems.. It's 100% free, no registration required.
up vote 0 down vote favorite

I have a Blade Center H and a DS4700 SAN. Today we detected one problem in the mail service in one of the blades. It can't see a particular volume available in the SAN. All other blades can see SAN disks from sdb to sdj , but this one sees only sdb to sdi. The volume that is sdd in all other blades is lost in this one and replaced by the next one.

I can mount sdd1 in every other blade, so sdd1 is working and has content. All servers are running CentOS 5.8. I'm totally lost here.

What thing can hide a disk from that system?

Edit: More info:

The HBAs are Emulex:

> lspci | grep -i emulex
03:05.0 Fibre Channel: Emulex Corporation Helios-X LightPulse Fibre Channel Host Adapter (rev 01)
03:05.1 Fibre Channel: Emulex Corporation Helios-X LightPulse Fibre Channel Host Adapter (rev 01)

The switch is an IBM TotalStorage SAN 16B-2. It was configured by the IBM personnel and it hasn't been touched by me.

The multipath is managed by the Linux RDAC multipath drivers for IBM SystemStorage provided by IBM personnel during the original installation.

> lsmod|egrep 'mpp|lpfc'
mppVhba               165888  11 
lpfc                  515109  18 
scsi_transport_fc      83145  1 lpfc
mppUpper              165532  1 mppVhba
scsi_mod              199001  18         be2iscsi,ib_iser,iscsi_tcp,bnx2i,cxgb3i,libiscsi2,scsi_transport_iscsi2,scsi_dh,mppVhba,usb_storage,lpfc,scsi_transport_fc,mptsas,mptscsih,scsi_transport_sas,mppUpper,sg,sd_mod

This is /proc/scsi

> ls /proc/scsi/
device_info  mpp/  mptsas/  scsi  sg/

> ls /proc/scsi/mpp/
3

In some machines there is a /proc/scsi/lpfc folder too.

share|improve this question

1 Answer

up vote 1 down vote

This could be a temporary problem on your server.

Rescan your devices and see if the missing one appears.

If not this is probably a configuration problem in the SAN.

I guess your SAN uses WWN based zoning. So check your zoning - is the missing disk in the zone for your target HBAs?

BTW - forget sd-names - they are irrelevant (i.e. more or less random). Look at the hardware-paths - they contain the WWNs of the SAN-disks.

share|improve this answer
Rescanning didn't worked. I guess is a config problem, but since it wasn't touched since installation, it's at least strange. There's one difference I see in the config between blades. The blade with problems see the disk using /proc/scsi/mpp/3 and the rest uses /proc/scsi/mpp/4. Why or how to change it, I don't know. – jdinuncio Sep 28 '12 at 14:05
@jdinuncio Can you provide more details about your SAN-storage as update to your question? Hardware/Vendor, how do you use/setup multipathing? What HBAs do you use, which firmware-revision... – Nils Sep 30 '12 at 21:20

Your Answer

 
discard

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.