DMP paths are not being enabled automatically after storage connection has been restored

Article: 100012120
Last Published: 2014-03-18
Ratings: 0 0
Product(s): InfoScale & Storage Foundation

Problem

In the event of a failure and restoration of a storage connection, the paths may not become enabled automatically and manual intervention will be required to re-enable them. Even running commands such as 'vxdctl enable' and 'vxdisk scandisks' did not re-enable the paths. In the end it was necessary to run the 'vxdmpadm enable' command for each of these paths.

 

Error Message

Breaking the Storage connection:

 
vxvm-udev:: device sdd with devno 8:48 has been removed. 
vxvm-udev:: device sdf with devno 8:80 has been removed. 
vxvm-udev:: device sdc with devno 8:32 has been removed. 
vxvm-udev:: device sde with devno 8:64 has been removed. 
vxvm-udev:: device sdb with devno 8:16 has been removed. 
vxvm-udev:: device sdg with devno 8:96 has been removed. 
kernel: VxVM vxdmp V-5-0-112 [Info] disabled path 8/0x40 belonging to the dmpnode 201/0x10 due to admin CLI kernel: 
kernel: VxVM vxdmp V-5-0-112 [Info] disabled path 8/0x20 belonging to the dmpnode 201/0x30 due to admin CLI kernel: 
kernel: VxVM vxdmp V-5-0-112 [Info] disabled path 8/0x60 belonging to the dmpnode 201/0x10 due to admin CLI kernel: 
kernel: VxVM vxdmp V-5-0-112 [Info] disabled path 8/0x30 belonging to the dmpnode 201/0x30 due to admin CLI kernel: 
kernel: VxVM vxdmp V-5-0-112 [Info] disabled path 8/0x10 belonging to the dmpnode 201/0x30 due to admin CLI kernel: 
 
 
After restoring the connection to the Storage, paths are seeing coming back online at the OS layer:
 
kernel: sd 1:0:0:0: [sdb] Write Protect is off 
kernel: sd 1:0:1:0: [sdc] Write Protect is off 
kernel: sde: 
kernel: sd 1:0:0:0: [sdb] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA 
kernel: sdf3 sdf8 
kernel: sd 1:0:1:0: [sdc] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA 
kernel: sde3 sde8 
kernel: sdc: 
kernel: sdb: 
kernel: sd 1:0:3:0: [sde] Attached SCSI disk 
kernel: sdd3 sdd8 
kernel: sdb3 sdb8 
kernel: sdc3 sdc8 
kernel: sd 1:0:4:0: [sdf] Attached SCSI disk 
kernel: sd 1:0:2:0: [sdd] Attached SCSI disk 
kernel: sd 1:0:0:0: [sdb] Attached SCSI disk 
kernel: sd 1:0:1:0: [sdc] Attached SCSI disk 
vxvm-udev: device sde with devno 8:64 has been added. 
vxvm-udev: device sdf with devno 8:80 has been added. 
vxvm-udev: device sdg with devno 8:96 has been added. 
vxvm-udev: device sdc with devno 8:32 has been added. 
vxvm-udev: device sdd with devno 8:48 has been added. 
vxvm-udev: device sdb with devno 8:16 has been added.
 
 
However if upon checking a disk using 'vxdisk list', the paths are still showing up as disabled:
eg.
numpaths: 6 
sdm state=enabled type=secondary 
sdk state=enabled type=primary 
sdl state=enabled type=secondary 
sdf state=disabled type=primary 
sdg state=disabled type=secondary 
sde state=disabled type=primary
 

Solution

Solution:

Once the latest VRTSaslapm was installed and the test repeated, the paths were then re-enabled automatically.

At the time of this issue, version 6.0.400.101 of the VRTSaslapm package was the latest available version on https://sort.Veritas.com/asl/latest


Applies To

RedHat 6.5

StorageFoundation 6.0.3 with HF 6.0.300.200

VRTSaslapm - 6.0.100.201

Was this content helpful?