Linux RHEL qla2xxx loop status

[Thierry D]

Hi all,

Got a RHEL server connected to SAN. Yesterday received error messages telling me some of the qla2xxx where dead.

# for i in 0 1 2 3 ; do echo "HBA: `pwd`/$i" ; cat $i | grep -i state; done
HBA: /proc/scsi/qla2xxx/0
Host adapter:loop state = , flags = 0x1a03
HBA: /proc/scsi/qla2xxx/1
Host adapter:loop state = , flags = 0x1a03
HBA: /proc/scsi/qla2xxx/2
Host adapter:loop state = , flags = 0x1a03
HBA: /proc/scsi/qla2xxx/3
Host adapter:loop state = , flags = 0x1a03
#

What I am missing / needing would be to know the different status possible for the loop state and the corresponding error description?

READY : all is good
DEAD : ?
....

Regards,
Thierry

[Ivan Ferreira · Accepted solution]

I think that the information that you need is available in the source code for the qlogic driver:

http://lxr.oss.org.cn/source/drivers/scsi/qla2xxx/qla_os.c


Whenever, a port is in the LOST state we start decrementing its port down timer every second until it reaches zero. Once it reaches zero
the port it marked DEAD.


loop_state) == LOOP_DOWN)
loop_state) == LOOP_UP)
loop_state) == LOOP_READY)
loop_state) == LOOP_TIMEOUT)
loop_state) == LOOP_UPDATE)
loop_state = "UNKNOWN";

[Benoy Daniel]

If you have hp_fibreutils installed, then verify the /opt/hp/hp_fibreutils/adapter_info output. Looks like you lost the link.

[Thierry D]

Info on Loop state received and link that were marked as DEAD where disconnected because not in use.

Thierry