03-28-2011 07:58 PM
Hi,
I am facing issue with DMP paths in RHEL 5. After some SAN issue, many path got diable.
I have tried to enable then via, #vxdmpadm enable pathname=path, but it didnt help.
Please refer the below snapshotm which may help to understant the issue.
[root@lme0024 ~]# vxdmpadm listenclosure all
ENCLR_NAME ENCLR_TYPE ENCLR_SNO STATUS ARRAY_TYPE LUN_COUNT
===================================================================================
other_disks OTHER_DISKS OTHER_DISKS CONNECTED OTHER_DISKS 1
xp12k-N2-2240 XP10K-12K 056E0 CONNECTED A/A 44
xp12k-W5-2290 XP10K-12K 05712 CONNECTED A/A 44
fas31701 FAS3170 200000078584 CONNECTED A/P-C-NETAPP 98
[root@lme0024 ~]# vxdmpadm listctlr all
CTLR-NAME ENCLR-TYPE STATE ENCLR-NAME
=====================================================
c1 FAS3170 ENABLED fas31701
c0 FAS3170 ENABLED fas31701
c360 OTHER_DISKS ENABLED other_disks
c0 XP10K-12K ENABLED xp12k-W5-2290
c1 XP10K-12K ENABLED xp12k-W5-2290
c0 XP10K-12K ENABLED xp12k-N2-2240
c1 XP10K-12K ENABLED xp12k-N2-2240
[root@lme0024 ~]#
[root@lme0024 ~]# vxdmpadm getsubpaths all|grep -v -i enable|more
NAME STATE[A] PATH-TYPE[M] DMPNODENAME ENCLR-NAME CTLR ATTRS
================================================================================
sdfk DISABLED PRIMARY fas31701_0 fas31701 c0 -
sdfl DISABLED PRIMARY fas31701_1 fas31701 c0 -
sdfy DISABLED PRIMARY fas31701_10 fas31701 c0 -
sdfz DISABLED PRIMARY fas31701_11 fas31701 c0 -
sdga DISABLED PRIMARY fas31701_12 fas31701 c0 -
sdgb DISABLED PRIMARY fas31701_13 fas31701 c0 -
sdgc DISABLED PRIMARY fas31701_14 fas31701 c0 -
sdge DISABLED PRIMARY fas31701_15 fas31701 c0 -
sdgd DISABLED PRIMARY fas31701_16 fas31701 c0 -
sdie DISABLED PRIMARY fas31701_17 fas31701 c0 -
[root@lme0024 ~]# vxdmpadm getsubpaths all|grep -v -i enable|more|wc -l
100
[root@lme0024 ~]#
Thnaks,
Viral Ahire
Solved! Go to Solution.
03-29-2011 08:15 AM
If you are unable to see the devices from even OS, then issue would be below veritas layer ... remember DMP sites above SCSI/FC layer... just by experience, I would say get the SAN checked, I am certain that DMP should pickup paths again if they are available .. very rare I needed to enable paths manually..
G
03-28-2011 10:33 PM
Which SF version?
What 'recovery' steps did you perform after the failure (e.g. vxdctl enable, vxreattach, etc)?
03-28-2011 11:32 PM
SF version is 5.0 mp4.
No recovery steps from unix end. Some issue @ SAN end. Still one path of each netapp disk is disable.
Is there anything that we need to check with SAN team??
Thanks,
Viral Ahire
03-28-2011 11:38 PM
If the restored daemon of vxdmp is running (vxdmpadm stat restored) then if paths are ok, it should automatically be scanned & marked enabled by DMP. In case running vxdctl enable dosn't help, I believe fix would come from SAN end only.. need to recheck the connectivity etc..
how do you see those paths from OS ? are you able to see those Luns from OS ? can you read from them ?
Gaurav
03-28-2011 11:59 PM
status of resoted daemon.
[root@lme0024 ~]# vxdmpadm stat restored
The number of daemons running : 1
The interval of daemon: 300
The policy of daemon: check_disabled
[root@lme0024 ~]#
I can see these Luns from OS but seems it is not active.
[root@lme0024 ~]# vxdmpadm getsubpaths all|grep sdhz
sdhz DISABLED SECONDARY fas31701_94 fas31701 c0 -
[root@lme0024 ~]# fdisk -l /dev/sdhz
[root@lme0024 ~]# fdisk -l /dev/sdhz|wc -l
0
[root@lme0024 ~]#
Viral Ahire
03-29-2011 08:15 AM
If you are unable to see the devices from even OS, then issue would be below veritas layer ... remember DMP sites above SCSI/FC layer... just by experience, I would say get the SAN checked, I am certain that DMP should pickup paths again if they are available .. very rare I needed to enable paths manually..
G