Multipath not showing correct RO/RW status
Issue
-
The SAN device was put into read only mode by storage team. Then we had rescanned the devices reload multipath. Kernel logging confirms read only mode (
Write protect
ison
). But multipath still shows RW status and doesn't change to RO.$ cat /sys/block/sdac/ro = 1 $ cat /sys/block/sded/ro = 1 $ cat /sys/block/sdif/ro = 1 $ cat /sys/block/sdmg/ro = 1 $ multipath -ll mpathec mpathec (123456789000292601779533031351222) dm-19 EMC,SYMMETRIX size=16G features='0' hwhandler='0' wp=rw <--- `-+- policy='queue-length 0' prio=1 status=active |- 0:0:0:16 sdac 65:0 active ready running |- 0:0:1:16 sded 70:128 active ready running |- 1:0:0:16 sdif 132:16 active ready running `- 1:0:1:16 sdmg 65:400 active ready running
-
The same issue is observed when device status is changed from RO to RW, i.e. dm-multipath was still listing the devices in RO state even when the SAN team had changed path status to RW
Environment
- Red Hat Enterprise Linux 6.7
kernel 2.6.32-573.el6
or laterdevice-mapper-multipath-0.4.9-72.el6
or later
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.