Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations Chris Miller on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

moving to none HACMP Environment

Status
Not open for further replies.

takheen

IS-IT--Management
Jan 21, 2016
1
SY
I have created HACMP environment years ago on 2 X p570 servers connected to DS4700 Storage subsystem

Each node will get its resources when staring the cluster services ( which is mainly a volume group )

And now finally we decided to stop using HACMP anymore and move the application VG manually between the nodes in case of failure ,

When cluster services in down on both nodes , I tried to export the VG from the first but not able to import it on the other node where I get this error :

0516-022 :Illegal parameter or structure error
0516-780 : importvg , unable to import volume group

By the way all VGs created n to be part of HACMP resource group should be set to “ don’t varyon VG at startup “ , when I tried to change this parameter it gave me :

0516-1328 : auto vary-on settings are not supported for an extended concurrent mode enabled VG
0516-732 : chvg unable to change VG


Noting that I am able to export/import VGs located on the system shared storage between these nodes but they are not tagged as a high availability HACMP resources

I guess that this VG is in a mode where only cluster services can move it from system to other , what do you think , do you have an idea ?

My second question , if a volume group has file systems and (some raw devices and logical volumes created inside the Database not from AIX and offcourse they appear in the lsvg – i vgname command output ) , and I want to run savevg command on this VG , I am going to have all LVs on backup media ?

Many Thanks
Kris

 
I guess you don't want to set VG auto varyon if you plan to have possibility to start resource on second node manually too
if you set auto-on on one, you will not be able vary it on on second one at the same time - unless your first node is down or VG is varied off manually beofore
also, please consider if you will be able to vary on VG on second node if first one is crashed - if you'll be able to remove scsi reservation locks (which are usually done by HACMP automatically - it has some preconfigured methods for this)

your errors with importing VG might be related to it is already varied on on apposite node (in such case maybe importvg -L .. will work --- -L stands for learn VGDA only I think).

besides, your VG is converted to enhanced concurrent mode which us usefull only when you use HACMP (only HACMP can manage it). That is why you can't setup auto varyon. To change autovaryon you would need to convert VG back to normal VG - see chvg command and -l switch

-l Changes the volume group into a Non-Concurrent Capable volume group. The volume group must be varied on in non-concurrent mode for this command to take effect.

 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top