Can't Create an LV??? 1

[khalidaaa]

Hi all,

I'm trying to create an lv on a vg on san that is connected to a node in an HACMP cluster. but the following error message appear:

Command: failed        stdout: yes           stderr: no

Before command completion, additional instructions may appear below.

The distribution of this command (111) failed on one or more passive nodes.
0516-822 mklv: Unable to create logical volume.

The only change i made lately is puting a new disk on this vg!

any idea why this is failing?

Regards,
Khalid

 

[khalidaaa]

This is the smitty page that i'm trying to do:

  Logical volume NAME                                [testlv]
* VOLUME GROUP name                                   oravg
* Number of LOGICAL PARTITIONS                       [10]                     #
  PHYSICAL VOLUME names                              []                      +
  Logical volume TYPE                                [jfs2]                  +
  POSITION on physical volume                         middle                 +
  RANGE of physical volumes                           minimum

 

[ogniemi]

if you put a new disk to the oravg via CSPOC then running lspv on any cluster node you should see this disk configured for oravg. can you confirm your new disk is available on all nodes and configured for oravg?

 

[ogniemi]

btw. I met problem adding new disk to shared vg on cluster then the new disk hadn't pvid when configured on both nodes.

to configure it successfuly I had to:

1. discover (configure/cfgmgr) the disk on one node
2. create pvid for new disk: chdev -l hdiskX -a pv=yes
3. discover the disk on the other cluster nodes
4. extend the VG from C-SPOC

 

[khalidaaa]

Thanks ogniemi for the comments

I've already done that and yes the lspv shows that the oravg is having this new disk and it was concurrent! I managed to fail it over to the next node and the databases were accessible as well from the other node!!

But i can't create any new lv on any of the disks in the oravg any more!

 

[KPKIND]

This is a known bug involved with gsclvmd part of the HACMP software. I have faced this problem quite a few times and logged it with IBM, who didn't have any clue as to what was causing this issue,the work around they have suggested was to varyoff the VG and varyon again which did fix the issue but cuased a downtime. Then they gave me a debug ifix to capture more debug information, which I have installed and finally after a couple of months the problme re-occured which capture more info and that enabled IBM to point it towards the gsclvmd and also they have realised this can be seen in two cases and the problem I was facing was case 1 of their explanation. It is an APAR now with number APAR IY91953.

hope this helps...

Cheers
KPKIND

 

[khalidaaa]

Oh My God!!!

I reported this to IBM but they haven't came back to me!

So i guess i will call for downtime!

I will keep you posted with the answer.

Thank you very much KPKIND

A star is deserved

Rerads,
Khalid

 

[KPKIND]

Hi Khalid,

Thanks. But I have really struggled when I had this problem for the first time..

Cheers
KPKIND