Anonymous_User Absent Member.
Absent Member.
1266 views

Monitoring Clusters

Is there any way to monitor what is happening with a clustered volume? I
just had a cluster go comatose and would like to know why.

Joe


Labels (1)
0 Likes
7 Replies
Anonymous_User Absent Member.
Absent Member.

Re: Monitoring Clusters

On Mon, 18 Jul 2005 11:51:32 +0000, Joe Regular wrote:

Checking server health in general should give you some information to
work with, there are no special log files for cluster services.
Usuallywhen a resource goes comatose it's caused by a DS problem.
So start off by checking the DS is healthy as per TID 10060600.


--
//Niclas Ekstedt
____________________________________________
Niclas Ekstedt, CNA/CNE/CNS/CLS
Network Consultant/NSC Sysop
InfraSystems Solutions
____________________________________________
(Sorry, support is not provided via email)

0 Likes
Anonymous_User Absent Member.
Absent Member.

Re: Monitoring Clusters

SNMP monitoring or a regular PING would show the loss of a resource.

However resources going comatose is an unusual problem and should be
resolved ASAP.



On Mon, 18 Jul 2005 11:51:32 GMT, "Joe Regular" <joeregular@gov.nl.ca>
wrote:

>Is there any way to monitor what is happening with a clustered volume? I
>just had a cluster go comatose and would like to know why.
>
>Joe
>


Tim
___________________
Tim Heywood (SYSOP)
NDS8
Scotland
(God's Country)
www.nds8.co.uk
___________________

http://support.novell.com/forums/

In theory, practice and theory are the same
In Practice, they are different
0 Likes
Anonymous_User Absent Member.
Absent Member.

Re: Monitoring Clusters

Normally comatose results when the resource cannot complete the startup
script.

You can copy out the script and hand execute the commands to figure out
which does not work.

Did this resource ever work?
Does it mount on any server in the cluster?

--
Timothy Leerhoff
Novell Support Forum Volunteer Sysop

The future comes slowly, the present flies and the past stands still
forever
0 Likes
Anonymous_User Absent Member.
Absent Member.

Re: Monitoring Clusters

"Timothy Leerhoff" wrote
> Normally comatose results when the resource cannot complete the startup
> script.
>
> You can copy out the script and hand execute the commands to figure out
> which does not work.
>
> Did this resource ever work?
> Does it mount on any server in the cluster?



The service works. There was just one night when the night that the master
node failed. Instead of failing over to the second node like it usually
does, the cluster went comatose.

Joe


0 Likes
Anonymous_User Absent Member.
Absent Member.

Re: Monitoring Clusters

> The service works. There was just one night when the night that the
> master node failed. Instead of failing over to the second node like
> it usually does, the cluster went comatose.


One of the items I saw that can cause a resource to go comatose is when
the server tries to bind the secondary IP address and it is already in
use on another server (not released yet) the resource will go comatose.

Maybe the abended server still had the secondary ip address still bound
when the second server tried to bind it.

--
Timothy Leerhoff
Novell Support Forum Volunteer Sysop

The future comes slowly, the present flies and the past stands still
forever
0 Likes
Anonymous_User Absent Member.
Absent Member.

Re: Monitoring Clusters

Particularly when a 2 node cluster is used.



On Fri, 22 Jul 2005 02:02:35 GMT, "Timothy Leerhoff"
<tleerhoffNO@SPAMqwest.net> wrote:

>> The service works. There was just one night when the night that the
>> master node failed. Instead of failing over to the second node like
>> it usually does, the cluster went comatose.

>
>One of the items I saw that can cause a resource to go comatose is when
>the server tries to bind the secondary IP address and it is already in
>use on another server (not released yet) the resource will go comatose.
>
>Maybe the abended server still had the secondary ip address still bound
>when the second server tried to bind it.


Tim
___________________
Tim Heywood (SYSOP)
NDS8
Scotland
(God's Country)
www.nds8.co.uk
___________________

http://support.novell.com/forums/

In theory, practice and theory are the same
In Practice, they are different
0 Likes
Anonymous_User Absent Member.
Absent Member.

Re: Monitoring Clusters

> Particularly when a 2 node cluster is used.

Normally when I see this issue it is a 2 node scsi shared drive array.

--
Timothy Leerhoff
Novell Support Forum Volunteer Sysop

The future comes slowly, the present flies and the past stands still
forever
0 Likes
The opinions expressed above are the personal opinions of the authors, not of Micro Focus. By using this site, you accept the Terms of Use and Rules of Participation. Certain versions of content ("Material") accessible here may contain branding from Hewlett-Packard Company (now HP Inc.) and Hewlett Packard Enterprise Company. As of September 1, 2017, the Material is now offered by Micro Focus, a separately owned and operated company. Any reference to the HP and Hewlett Packard Enterprise/HPE marks is historical in nature, and the HP and Hewlett Packard Enterprise/HPE marks are the property of their respective owners.