Ocfs users - Jul 2004 - My node is restarted when running OCFS

Hi,
 
I am trying to install Oracle 9i RAC on SLES 8 SP3
with a shared firewire disk.
 
I am using the following OCFS:
OCFS Release 1.0.11-1 for suse sp3 2.4.21-138 smp kernel
ocfs-2.4.21-138-smp-1.0.11-1.i586.rpm 
ocfs-tools-1.0.10-1.i386.rpm 
ocfs-support-1.0.10-1.i386.rpm 
 
1/ I have configured OCFS and a shared file system /opt/oracle/oradata
2/ I have created a raw device for use as quorum file   
    (/opt/oracle/oradata/CMQuorumFile)
    This file is referenced by a symbolic link on the above shared OCFS  
    file system.  The quorum file has the 660 permission and is owned by 
    oracle:oinstall.
 
3/ I have installed Oracle Cluster manager patchset 9.2.0.4. 
    And I can start the cluster manager on both nodes 
    using Suse "rcoacle start" command provide by the orarun package.
 
My problem is: when I start Oracle cluster manager on one node (or both nodes),
the node is restarted after a few minutes.
When I don't start Oracm, I don't have this problem.
 
And verifying the content of the cluster manager log file 
(/opt/oracle/product/9ir2/oracm/cm.log) does not show me any problem.
 
Where can I find the system's log file?
Any idea??? 
That is my last step before installing Oracle RAC
 
Thanks.
Jean




---------------------------------
L?che-vitrine ou l?che-?cran ? Yahoo! Magasinage.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://oss.oracle.com/pipermail/ocfs-users/attachments/20040723/e0ab3a75/attachment.htm

As I recall, the hangcheck-timer module that comes with SLES 8 is
buggy.  We had this same problem.  Get the hangcheck-timer source from
oss.oracle.com and compile your own - this should fix the problem.  The
Cluster Manager loads the hangcheck-timer module when it starts, if I
recall correctly.

Eric Hensley
Director, IT
Instill Corp.

On Fri, 2004-07-23 at 13:32, Jean-Hilaire Yapi wrote:
> Hi,
>  
> I am trying to install Oracle 9i RAC on SLES 8 SP3
> with a shared firewire disk.
>  
> I am using the following OCFS:
> OCFS Release 1.0.11-1 for suse sp3 2.4.21-138 smp kernel
> ocfs-2.4.21-138-smp-1.0.11-1.i586.rpm 
> ocfs-tools-1.0.10-1.i386.rpm 
> ocfs-support-1.0.10-1.i386.rpm 
>  
> 1/ I have configured OCFS and a shared file system /opt/oracle/oradata
> 2/ I have created a raw device for use as quorum file   
>     (/opt/oracle/oradata/CMQuorumFile)
>     This file is referenced by a symbolic link on the above shared OCFS  
>     file system.  The quorum file has the 660 permission and is owned by 
>     oracle:oinstall.
>  
> 3/ I have installed Oracle Cluster manager patchset 9.2.0.4. 
>     And I can start the cluster manager on both nodes 
>     using Suse "rcoacle start" command provide by the orarun
package.
>  
> My problem is: when I start Oracle cluster manager on one node (or both
nodes), the node is restarted after a few minutes.
> When I don't start Oracm, I don't have this problem.
>  
> And verifying the content of the cluster manager log file 
> (/opt/oracle/product/9ir2/oracm/cm.log) does not show me any problem.
>  
> Where can I find the system's log file?
> Any idea??? 
> That is my last step before installing Oracle RAC
>  
> Thanks.
> Jean
> 
> 
> 
> 
> ---------------------------------
> L?che-vitrine ou l?che-?cran ? Yahoo! Magasinage.
> 
> ______________________________________________________________________
> _______________________________________________
> Ocfs-users mailing list
> Ocfs-users@oss.oracle.com
> http://oss.oracle.com/mailman/listinfo/ocfs-users