a lot of Buffer I/O error on device connected to SAN

[KarloChacon]

hi guys

I have a server(x3650 M2) connected to a SAN...it takes a long time to boot with a lot errors like these

any idea how to fix that?
Suse Enterprise 10 SP3


sdi:end_request: I/O error, dev sdh, sector 0
Buffer I/O error on device sdh, logical block 0
end_request: I/O error, dev sdd, sector 209715072
Buffer I/O error on device sdd, logical block 26214384
end_request: I/O error, dev sdi, sector 0
Buffer I/O error on device sdi, logical block 0
end_request: I/O error, dev sdg, sector 209715072
Buffer I/O error on device sdg, logical block 26214384
end_request: I/O error, dev sdh, sector 0
Buffer I/O error on device sdh, logical block 0
end_request: I/O error, dev sdd, sector 209715072
printk: 2 messages suppressed.
Buffer I/O error on device sdh, logical block 0
end_request: I/O error, dev sdd, sector 209715184
unable to read partition table
sd 5:0:1:1: Attached scsi disk sdh
sd 5:0:1:1: Attached scsi generic sg8 type 0
end_request: I/O error, dev sdi, sector 0
unable to read partition table
sd 6:0:1:1: Attached scsi disk sdi
sd 6:0:1:1: Attached scsi generic sg9 type 0


end_request: I/O error, dev sdd, sector 0
Buffer I/O error on device sdd, logical block 0
end_request: I/O error, dev sdd, sector 0
Buffer I/O error on device sdd, logical block 0
end_request: I/O error, dev sdd, sector 0
Buffer I/O error on device sdd, logical block 0
Buffer I/O error on device sdd, logical block 1
Buffer I/O error on device sdd, logical block 2
Buffer I/O error on device sdd, logical block 3
end_request: I/O error, dev sdd, sector 0
Buffer I/O error on device sdd, logical block 0
end_request: I/O error, dev sdg, sector 0
Buffer I/O error on device sdg, logical block 0
end_request: I/O error, dev sdg, sector 0


thanks a lot

[KarloChacon]

update as soon as I disconnect the HBAs boots up normal and fast and no messages

[KarloChacon]

something else

when my Suse 10 SP3 loads when HBA attached I see my LUNs twice

for instance I created LUN1 60GB and I see it twice...? are those 2 related I think the slow load process and the diuplicated LUNs

[Tim Nelson]

two items:

the io errors are probably real, lun was probably removed. I would have thought that a reboot would remove the reference to the devices unless they are still used for something.


Most SAN implementations have multiple paths to the disk array for redundancy. The use of multipath on linux servers allows you to use a single consistant name to access the redundant paths.

on RHEL it is typically /dev/dm-?
The device naming can be modified for nice-names by configuring /etc/multipath.conf and using multipath command.

then all devices are accessed using /dev/mapper/nice-name

I can only assume SUSE is the same.

[Tim Nelson]

correction.
/dev/dm-* is a dynamic multipath name

/dev/mpath is the persistant name

/dev/mapper/ can also be used to reference the persistant name

[KarloChacon]

maybe I need a multipath software for my linux
I am still searching

[KarloChacon]

In fact I get this on fdisk -l


Disk /dev/dm-0: 10.7 GB, 10737418240 bytes
255 heads, 63 sectors/track, 1305 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Disk /dev/dm-0 doesn't contain a valid partition table

Disk /dev/dm-1: 5368 MB, 5368709120 bytes
255 heads, 63 sectors/track, 652 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Disk /dev/dm-1 doesn't contain a valid partition table



and either pvs - lvs - gvs
which are the ones presented on boot up

/dev/sdd: read failed after 0 of 4096 at 0: Input/output error
/dev/sdg: read failed after 0 of 4096 at 0: Input/output error
/dev/sdh: read failed after 0 of 4096 at 0: Input/output error
/dev/sdi: read failed after 0 of 4096 at 0: Input/output error