[chennuri]

Hi All,

I have encountered the following error when performing mkdev on vio servers.

Some error messages may contain invalid information
for the Virtual I/O Server environment.
Method error (/usr/lib/methods/cfg_vt_scdisk):
0514-047 Cannot access a device.

We have four vio servers vio1 vio2 vio3 vio4,as per our request storage team allocated a lun and mapped to all vio servers. storage (10072) Hitachi.

we followed the steps given below.
1.cfgmgr on all vio servers to sense the new and were able to sense without any issues.
2.lspv to identify the hdisk
3. login as padmin on vio and setup env oem_setup_env.
4. chdev -l hdiskx -a pv=yes -a reserve_policy=no_reserve -P
5./usr/ios/cli/ioscli mkvdev -vdev hdiskx -vadapter vhost0 -dev Vxyz
I'm able to success on vio1 server1 and encountered the problem on other vioservers (2|3|4).
Some error messages may contain invalid information
for the Virtual I/O Server environment.
Method error (/usr/lib/methods/cfg_vt_scdisk):
0514-047 Cannot access a device.

To fix the error i have tried with below commands but not helped much.
1. Delted hdisk with rmdev -dl hdiskx
2. executed cfgmgr
3. after step 2 pvid disappeared
4. Again tried to create pv and set the policy but emcountered the given probs.

Some error messages may contain invalid information
for the Virtual I/O Server environment.
Method error (/usr/lib/methods/cfg_vt_scdisk):
0514-047 Cannot access a device.
5./usr/ios/cli/ioscli mkvdev -vdev hdiskx -vadapter vhost0 -dev Vxyz

Help is highly appreciated.

Due to this enter work got held.

[ross.mather]

You've set the permanent reserve_policy - which takes place after a reboot, but you need to repeat that command for the current running system. There is also no need to do that using the oem_set_env command, you can do it as:

chdev -dev hdiskx -attr reserve_policy=no_reserve pv=yes
chdev -dev hdiskx -attr reserve_policy=no_reserve -perm

If you have got this wrong you need to:

Remove all the disks for all the VIO Servers
Deallocate the disks from the VIO Servers on the EMC
Wait a few minutes
Reallocate the disks to the VIO Servers
and run cfgmgr

[chennuri]

Problem got fixed after setting Lun hostgroup for Ldev in array.

Originally Posted by ross.mather

You've set the permanent reserve_policy - which takes place after a reboot, but you need to repeat that command for the current running system. There is also no need to do that using the oem_set_env command, you can do it as:

chdev -dev hdiskx -attr reserve_policy=no_reserve pv=yes
chdev -dev hdiskx -attr reserve_policy=no_reserve -perm

If you have got this wrong you need to:

Remove all the disks for all the VIO Servers
Deallocate the disks from the VIO Servers on the EMC
Wait a few minutes
Reallocate the disks to the VIO Servers
and run cfgmgr

[GFGMojo]

You did it all correctly - you just need to reboot the first VIO server that allocated the disk - the first VIO will lock up the disk on the SAN so the others will not see it - after making the no_reserve changes - reboot the first server which will release the locks on the SAN and you will be able to see the disks from all servers if you have your MPIO software and device drivers installed on the VIO servers.