Thursday, February 16, 2012

Cannot get new disks to fail over when I move group - new drives

I have a SQL 2000 Ent server cluster with current drives in place and
working. We are adding two new drives from a new SAN enclosure. The server
sees the drives fine. I have added the drives into the depenecy list under
the SQL server listing under the "SQL group" in cluster administrator. It is
also listed in the resources group. When we fail it over it jumps right back
to the original node. We receive event id#1069 in the system event viewer.
Any input or suggestions would be greatly appreciated.
Thanks,
Jim K.
"Cinci Tech" <CinciTech@.discussions.microsoft.com> wrote in message
news:6C533F4C-3145-485C-98AC-8BE4240BF146@.microsoft.com...
>I have a SQL 2000 Ent server cluster with current drives in place and
> working. We are adding two new drives from a new SAN enclosure. The
> server
> sees the drives fine. I have added the drives into the depenecy list
> under
> the SQL server listing under the "SQL group" in cluster administrator. It
> is
> also listed in the resources group. When we fail it over it jumps right
> back
> to the original node. We receive event id#1069 in the system event
> viewer.
> Any input or suggestions would be greatly appreciated.
Are the LUN masks set up to see the controllers on all nodes?
Russ Kaufmann
MVP - Windows Server - Clustering
ClusterHelp.com, a Microsoft Certified Gold Partner
Web http://www.clusterhelp.com
Blog http://msmvps.com/clusterhelp
|||I am wondering if I have done something wrong... I have only built three
clusters so far. I looked back over some notes and saw something about when
I first added the disks, one node had to be shut down (the "node 2" owned the
disks and corrupted them) we had to pull them at the SAN level and re-present
them to the first node while the 2nd node was powered off. Then we powered
down the first node and rescanned the disk on the 2nd node. Once they were
seen on the 2nd server we brought up the 1st node and failed them over. Then
we added them into the dependencies list. Does this sound vaguely pertinent?
When you say "Lun masks" I am a bit confused by that word lun masks.
I think we are both aiming at the same thing. What are your thoughts?
"Russ Kaufmann [MVP]" wrote:

> "Cinci Tech" <CinciTech@.discussions.microsoft.com> wrote in message
> news:6C533F4C-3145-485C-98AC-8BE4240BF146@.microsoft.com...
> Are the LUN masks set up to see the controllers on all nodes?
>
> --
> Russ Kaufmann
> MVP - Windows Server - Clustering
> ClusterHelp.com, a Microsoft Certified Gold Partner
> Web http://www.clusterhelp.com
> Blog http://msmvps.com/clusterhelp
>
>
|||"Cinci Tech" <CinciTech@.discussions.microsoft.com> wrote in message
news:AB6464C1-0836-4B39-8FA8-954E50E1C60A@.microsoft.com...
>I am wondering if I have done something wrong... I have only built three
> clusters so far. I looked back over some notes and saw something about
> when
> I first added the disks, one node had to be shut down (the "node 2" owned
> the
> disks and corrupted them) we had to pull them at the SAN level and
> re-present
> them to the first node while the 2nd node was powered off.
You could have done that while Node2 was up.

> Then we powered
> down the first node and rescanned the disk on the 2nd node. Once they
> were
> seen on the 2nd server we brought up the 1st node and failed them over.
> Then
> we added them into the dependencies list. Does this sound vaguely
> pertinent?
Absolutely. So, it sounds to me like both nodes have been properly exposed
to the LUNs.
What are the actually errors that you are getting in Event Viewer and the
cluster logs?
Russ Kaufmann
MVP - Windows Server - Clustering
ClusterHelp.com, a Microsoft Certified Gold Partner
Web http://www.clusterhelp.com
Blog http://msmvps.com/clusterhelp
|||I am getting the following: 1069 and 1034 errors.
Event Type:Error
Event Source:ClusSvc
Event Category:Physical Disk Resource
Event ID:1034
Date:11/21/2006
Time:3:31:37 PM
User:N/A
Computer:COR-DB-02
Description:
The disk associated with cluster disk resource 'Disk G:' could not be found.
The expected signature of the disk was B812F4A6. If the disk was removed from
the server cluster, the resource should be deleted. If the disk was replaced,
the resource must be deleted and created again in order to bring the disk
online. If the disk has not been removed or replaced, it may be inaccessible
at this time because it is reserved by another server cluster node.
and...
Event Type:Error
Event Source:ClusSvc
Event Category:Failover Mgr
Event ID:1069
Date:11/21/2006
Time:3:31:37 PM
User:N/A
Computer:COR-DB-02
Description:
Cluster resource 'Disk F:' in Resource Group 'SQL Group' failed.
"Russ Kaufmann [MVP]" wrote:

> "Cinci Tech" <CinciTech@.discussions.microsoft.com> wrote in message
> news:AB6464C1-0836-4B39-8FA8-954E50E1C60A@.microsoft.com...
> You could have done that while Node2 was up.
>
> Absolutely. So, it sounds to me like both nodes have been properly exposed
> to the LUNs.
> What are the actually errors that you are getting in Event Viewer and the
> cluster logs?
>
> --
> Russ Kaufmann
> MVP - Windows Server - Clustering
> ClusterHelp.com, a Microsoft Certified Gold Partner
> Web http://www.clusterhelp.com
> Blog http://msmvps.com/clusterhelp
>
>
|||"Cinci Tech" <CinciTech@.discussions.microsoft.com> wrote in message
news:23E20EFF-9E8C-4047-AC4B-CFD40CA20A51@.microsoft.com...
>I am getting the following: 1069 and 1034 errors.
> Event Type: Error
> Event Source: ClusSvc
> Event Category: Physical Disk Resource
> Event ID: 1034
> Date: 11/21/2006
> Time: 3:31:37 PM
> User: N/A
> Computer: COR-DB-02
> Description:
> The disk associated with cluster disk resource 'Disk G:' could not be
> found.
> The expected signature of the disk was B812F4A6. If the disk was removed
> from
> the server cluster, the resource should be deleted. If the disk was
> replaced,
> the resource must be deleted and created again in order to bring the disk
> online. If the disk has not been removed or replaced, it may be
> inaccessible
> at this time because it is reserved by another server cluster node.
It sounds like the disk might have been formatted on both nodes. The best
way to handle it is to delete the disk, then re-add it from one node and
format it on that node.
Russ Kaufmann
MVP - Windows Server - Clustering
ClusterHelp.com, a Microsoft Certified Gold Partner
Web http://www.clusterhelp.com
Blog http://msmvps.com/clusterhelp
|||I was able to get it to work, I removed them from the "resource" group,
failed each node.. then started on the 1st node in the cluster, rescanned the
disks, removed the drive letter and readded the drive letter back to the
disk. Once I did that one each node. I then added the drive back into CL.
Admin. to a test group I was able to get to fail over in that test group. I
then moved them into the SQL server group and set dependencies. It worked.
I would to thank you for being so willing to help, the community groups have
a really big help to me. Clustering isn't something anyone sent me to
traning on, I watched someone do it, took notes then did it myself, talked to
a few poeple and learned what I could. If you know of any "official" docs
from MS on how to properly setup clustering and have links to them I would be
greatful to see something. Again, thanks for your help.
Jim
"Russ Kaufmann [MVP]" wrote:

> "Cinci Tech" <CinciTech@.discussions.microsoft.com> wrote in message
> news:23E20EFF-9E8C-4047-AC4B-CFD40CA20A51@.microsoft.com...
> It sounds like the disk might have been formatted on both nodes. The best
> way to handle it is to delete the disk, then re-add it from one node and
> format it on that node.
> --
> Russ Kaufmann
> MVP - Windows Server - Clustering
> ClusterHelp.com, a Microsoft Certified Gold Partner
> Web http://www.clusterhelp.com
> Blog http://msmvps.com/clusterhelp
>
>
|||Take Russ' course (www.clusterhelp.com). I did, and I had already set up a
couple of clusters (under duress) before doing so.
Tom
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
..
"Cinci Tech" <CinciTech@.discussions.microsoft.com> wrote in message
news:C4F35D3C-488A-47E1-844C-B999ED7D58EF@.microsoft.com...
I was able to get it to work, I removed them from the "resource" group,
failed each node.. then started on the 1st node in the cluster, rescanned
the
disks, removed the drive letter and readded the drive letter back to the
disk. Once I did that one each node. I then added the drive back into CL.
Admin. to a test group I was able to get to fail over in that test group. I
then moved them into the SQL server group and set dependencies. It worked.
I would to thank you for being so willing to help, the community groups have
a really big help to me. Clustering isn't something anyone sent me to
traning on, I watched someone do it, took notes then did it myself, talked
to
a few poeple and learned what I could. If you know of any "official" docs
from MS on how to properly setup clustering and have links to them I would
be
greatful to see something. Again, thanks for your help.
Jim
"Russ Kaufmann [MVP]" wrote:

> "Cinci Tech" <CinciTech@.discussions.microsoft.com> wrote in message
> news:23E20EFF-9E8C-4047-AC4B-CFD40CA20A51@.microsoft.com...
> It sounds like the disk might have been formatted on both nodes. The best
> way to handle it is to delete the disk, then re-add it from one node and
> format it on that node.
> --
> Russ Kaufmann
> MVP - Windows Server - Clustering
> ClusterHelp.com, a Microsoft Certified Gold Partner
> Web http://www.clusterhelp.com
> Blog http://msmvps.com/clusterhelp
>
>
|||"Cinci Tech" <CinciTech@.discussions.microsoft.com> wrote in message
news:C4F35D3C-488A-47E1-844C-B999ED7D58EF@.microsoft.com...
>I was able to get it to work, I removed them from the "resource" group,
> failed each node.. then started on the 1st node in the cluster, rescanned
> the
> disks, removed the drive letter and readded the drive letter back to the
> disk. Once I did that one each node. I then added the drive back into
> CL.
> Admin. to a test group I was able to get to fail over in that test group.
> I
> then moved them into the SQL server group and set dependencies. It
> worked.
Thank you for telling us all what you did to fix it. Since these newsgroups
are indexed by many companies, it will likely help somebody in the future.

> I would to thank you for being so willing to help, the community groups
> have
> a really big help to me.
Everyone here is a good person. While we may not all agree, we do love the
technology. <G>
Russ Kaufmann
MVP - Windows Server - Clustering
ClusterHelp.com, a Microsoft Certified Gold Partner
Web http://www.clusterhelp.com
Blog http://msmvps.com/clusterhelp
|||"Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
news:%23Ildh5iDHHA.3212@.TK2MSFTNGP04.phx.gbl...
> Take Russ' course (www.clusterhelp.com). I did, and I had already set up
> a
> couple of clusters (under duress) before doing so.
Thanks for the plug, Tom.
Russ Kaufmann
MVP - Windows Server - Clustering
ClusterHelp.com, a Microsoft Certified Gold Partner
Web http://www.clusterhelp.com
Blog http://msmvps.com/clusterhelp

No comments:

Post a Comment