Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

ASM disk cannot be found

1,347 views
Skip to first unread message

charles

unread,
Nov 12, 2010, 1:11:22 PM11/12/10
to
Dear Group,

I have been working with Oracle without success.

On my box, when i installed 10g, oracle cannot find the ASM disk with
the default path 'ORCL:DISK1'. I have to add a asm_diskstring in
order for Oracle 10g to work. that was not a problem.

Recently, i have wiped out 10g and trying to install 11.2.0.2 grid
infrastructure, i got the error PRVF-5150
PRVF-5150 : Path ORCL:VOL1 is not a valid path on all nodes -
Cause: -
Action:

Oracle asked me to ignore the error and continue with the install,
when running root.sh
i got
.....
Adding daemon to inittab
ohasd failed to start
Failed to start the Clusterware. Last 20 lines of the alert log
follow:
2010-11-12 12:56:55.448
[client(30805)]CRS-2101:The OLR was formatted using version 3.
2010-11-12 12:56:59.167
[client(30807)]CRS-2317:Fatal error: cannot get local GPnP security
keys (wallet).
2010-11-12 12:56:59.167
[client(30807)]CRS-2316:Fatal error: cannot initialize GPnP,
CLSGPNP_ERR (Generic GPnP error).
2010-11-12 12:56:59.169
[client(30807)]CRS-1013:The OCR location in an ASM disk group is
inaccessible. Details in /crs/oracle/product/11.2.0/grid/log/rac1/
client/clscfg.log.

I have problem only on this box, but our SA insist this box is the
same as other boxes. Somehow ASMlib is not configured right.

Could somebody help?

Thanks

John Hurley

unread,
Nov 12, 2010, 5:30:07 PM11/12/10
to

Hows about working on this one ...

charles

unread,
Nov 16, 2010, 3:49:56 PM11/16/10
to

> Hows about working on this one ...
>
> CRS-1013:The OCR location in an ASM disk group is
> inaccessible. Details in /crs/oracle/product/11.2.0/grid/log/rac1/
> client/clscfg.log.

Not much useful info

[ CKSCFG][2575079776]clscfg_main: Configuration type [4]

onedbguru

unread,
Nov 17, 2010, 8:43:30 PM11/17/10
to
On Nov 12, 1:11 pm, charles <dshprope...@gmail.com> wrote:

First, I have never used ASMlib on any of the more than 25+ clusters I
have built (Solaris, AIX and Linux). If you understand disk device
creation/setup (SAN (preferable) and locally connected SCSI), then
ASMlib is unnecessary.

In addition to the disk string, did you remember to change the the
ownership of the oracle devices to oracle:oinstall (or oracle:dba)???
You need to become a SYSADMIN if you are going to be RAC admin, and
rarely trust what your SysAdmin says unless he can prove it. I have
had to train many sys admins to do RAC and even ASM correctly.

ppineda76

unread,
Jan 21, 2011, 12:26:47 PM1/21/11
to
Have you made any headway on this? I was wondering is you were using multipath for your storage


cquimby

unread,
Jul 30, 2012, 11:33:22 AM7/30/12
to
I know this is very old thread but ...
I have exactly the same issue and we are using multipath. Please help


TheBoss

unread,
Jul 30, 2012, 7:47:39 PM7/30/12
to
cquimby <chunsin...@gmail.com> wrote in
news:iI6dndSxDLTfNIvN...@giganews.com:

> I know this is very old thread but ...
> I have exactly the same issue and we are using multipath. Please help
>

You are not giving us much info to work with, but here's some pointers
to Support Notes on MOS that might be of help:

[ID 1053147.1]
11gR2 Clusterware and Grid Home - What You Need to Know

[ID 1312225.1]
Things to Consider Before Upgrading to 11.2.0.2 Grid Infrastructure/ASM

[ID 1050908.1]
Troubleshoot Grid Infrastructure Startup Issues

[ID 1053970.1]
Troubleshooting 11.2 Grid Infrastructure root.sh Issues

[ID 1332725.1]
CRS-2317:Fatal error: cannot get local GPnP security keys (wallet).

I should start with notes 3, 4, and 5 as these are targeted at
troubleshooting specific issues. If they don't offer a solution, the
first 2 notes can give you more general info that might put you in the
right direction where to look further.

--
Jeroen

John Hurley

unread,
Aug 3, 2012, 4:45:56 PM8/3/12
to
On Jul 30, 11:33 am, cquimby <chunsin.qui...@gmail.com> wrote:
> I know this is very old thread but ...
> I have exactly the same issue and we are using multipath.  Please help

My crystal ball is broken ... give us details ... lots of details ...

What kind of server / what os / ASM or RAC or both / what version of
oracle / error messages and log file contents ... etc.

I just ran into and fixed now an interesting issue with 11.2 ASM on OL
6.2 and EMC VNX not seeing disks correct.

Some pretty smart and experienced people out here but we cannot help
without relevant information.
0 new messages