Hi

I am trying to add a third node to an existing 2 node 2008 R2 SP1 Hyper-v cluster. The validation passes with 1 warning that concerns me.

Under 'List all Disks' check one of my CSV disks is not 'read' by server that i am trying to join to the cluster. The existing nodes read it fine. The error I get is

Failed to get information on PhysicalDrive 7 from node 'node_name' status 2

I have a total of 13 luns provisioned from my storage and none of the other drives encounter this problem.

The above drive gets assigned a drive letter and appears like a local disk with no regards for the existing reservations. In fact the server kept auto assigning a drive letter to it till I used diskpart to disable this. Now it just appears as volume under the new server even though I have not joined it to the cluster. Temporarily i have disabled iscsi communication from this server till i can figure out why it is behaving this way. Just for additional information, when it was assigned a drive letter Event id 55 from source ntfs was logged repeatedly in eventviewer.

Any ideas on how to figure how what the status codes mean and how i can stop this behaviour?

Hi all,

I am trying to install a SQL Failover Cluster nodes for our infrastructure. The idea is to provide 2-NLB SharePoint Servers, with 2-Failover-Clustered SQL. I am now in the testing environment, in which I am using a software called StarWind to emulate the iSCSI Hard-Drive.

I have finished adding the Failover Service, running smoothly. However, when I try to install the SQL Server Cluster node it gives an error:

[[[

Rule "Cluster shared disk available check" Failed.

The cluster on this computer does not have a shared disk available. To continue, at least one shared disk must be available.

]]]

if anyone has an idea, pls assist??

many thanks to all of you.. :)

Greetings. I am trying setup a HyperV environment to meet similar demands on my VMware Environment. I have VRanger, SRM. We do not use Failover at this time. An unplanned failover looks like the server rebooted and takes 30 seconds or so typically. I will be using Fiber Channel Storage. Onto a plan for HyperV.

I hear the terms "Continuously Available" and "no perceived downtime" for SOFS. It is implied that this is better than a CSV being only highly available.  Does it really take a few minutes for a guest to failover with a CSV in an unplanned failover? Does it really take just a few seconds on a SOFS? If so how do we manage no perceived downtime with SOFS in case of a host failure? Do guest VMs on different hosts keep their memory and CPU states in sync? Does SMB introduce latency and extra complexity? Would DPM be the best backup solution? Are there any third part tools to extend failover beyond high, medium, and low? Boot these 10 servers first....etc?

Hello,

I'm discovering virtualization, and I would like to know if the schema described below is realistic or not.

1) I have 2 physical servers (P1 and P2), each having its own internal SATA disks, and running W2012 standard; on 1 site

2) On P1, I create 2 VM : Exchange server (EXCH-SRV) and application server (APPL-SRV)

3) On P2, I would like to create 2 identical VM, and use them as failover of the VM of P1

4) Would it be possible to replicate in real-time the disks of EXCH-SRV and APPL-SRV from P1 to P2 ?

5) Would it be possible to replicate in real-time the memory state of EXCH-SRV and APPL-SRV from P1 to P2 ?

6) And, would this allow to have an automated failover (invisible for users) from P1 to P2 when P1 is down ?

7) Do I need specific additional components (NIC, specific licenses...) ?

Thanks in advance for your help.

We're running a two node MSFC with FC attached storage. I have presented all new LUNS to the nodes and need to move all the old storage assigned to the MSFC over to the new nodes hopefully without breaking any of the currently configured Services and Applications. Down time is ok, just don't want the SAP folks to have to rebuild things.

Is this possible?

Here's the order I'm thinking of doing this;

1. Shutdown the cluster

2. Re-assign the volume drive letters to the new volumes

3. Start the cluster

Will I need to do a re-validate? Will the change of the UUID be a problem?

Jason Wilcox

I have a Server 2008 2-node multi-site cluster, using node and file majority. The network connection between one node and the file has some kind very short problem (our network team is denying anything is wrong), the other node and other multi-site clusters use the exact same file resource and do not have a problem.
Is there a timeout value that I can increase, as I think the problem is very short?

The exact error is:

File share witness resource 'File Share Witness' failed to arbitrate for the file share '\\servername\witness\clustername.
Please ensure that file share '\\servername\witness\clustername' exists and is accessible by the cluster.

Thanks,

M

Have 2 servers running Server 2008 R2 service pack 1.  Both have failover cluster feature installed.  When I go to validate a configurations, I try to add the other server and it always replies with the same error. 

Event id 4631.

Failover cluster manager detected a parameter that was not in the expected range: the computer (name) does not have the failover clustering feature installed.

But both have the feature installed.  They can ping each other, on the same domain, same hardware and software setup.  Ive tried removing the feature and re-adding it and rebooting, but nothing seems to correct the issue.

Any help would be great.

THank you

I'm looking for some clarification around setting up what should be a fairly basic 2-node cluster configuration and hoping that some of the clever minds around here can assist.  The client is small office with a small amount of stored data (~100GB) and running a few apps (nothing unusual - no DBs, no virtualization or anything of the sort).  As they only have a single server currently, and maintaining uptime is fairly critical, it was suggested to purchase a second one and configure a 2-node cluster to provide some redundancy.

1. The existing server is currently running on Server 2008 Std which appears not to support Failover Clustering, so it looks like it'll have to be upgraded to Server 2012, correct?

2. The new server (and the current one) are both validated for Server 2012, but are different models from the same manufacturer. I've read in some posts that others have successfully configured clustering on different hardware though it's not a best practice. Given that the primary intention of having the second server is just to provide temporary service allowing users to keep working until the primary server can be brought back online, is having different models in the cluster likely to be an issue? If so, what would those issues likely be?

3. There is currently no shared storage in the environment and given the small amount of data being currently stored on the server, there's no plan to purchase any. Though I've read in various design guides that shared storage is necessary, can someone please confirm this as I'm not sure that I'm interpreting the guides correctly given the simple nature of this client's environment.

4. The configuration guides I've read (or watched) have all been for more complex environments than this client's.  Does anyone know of a guide for setting up a basic 2-node cluster or make any recommendations? It would be greatly appreciated.

Thanks for any assistance.

JL

Hello!

Organization use Exchange 2007 SP3 mailbox servers in SCC.
There are 4 active and 1 passive node in cluster.
All OS are Windows server 2008 Ent SP2.
All domain controllers OS Windows server 2008 Std SP2.
Cluster and domain controllers placed in the same AD site.
On all active nodes every day occures an error, registered in "System" journal:
____________________________________
Log Name:      System
Source:        Microsoft-Windows-FailoverClustering
Date:          2/5/2013 8:33:34 PM
Event ID:      1207
Task Category: Network Name Resource
Level:         Error
Keywords:     
User:          SYSTEM
Computer:      node1.domainname.com
Description:
Cluster network name resource 'Network Name (App1)' cannot be brought online. The computer object associated with the resource could not be updated in domain 'domainname.com' for the following reason:
Unable to update password for computer account.

The text for the associated error code is: Access is denied.
 
The cluster identity 'Cluster$' may lack permissions required to update the object. Please work with your domain administrator to ensure that the cluster identity can update computer objects in the domain.
____________________________________

And on 3 active nodes periodical occurs event, registered in "FailoverClustering/Operational" journal:
____________________________________
Log Name:      Microsoft-Windows-FailoverClustering/Operational
Source:        Microsoft-Windows-FailoverClustering
Date:          2/5/2013 2:27:54 PM
Event ID:      1201
Task Category: Resource Control Manager
Level:         Information
Keywords:     
User:          SYSTEM
Computer:      node1.domainname.com
Description:
The Cluster service successfully brought the clustered service or application 'App1' online.
____________________________________

I don't understand - why contradictory events occurs on cluster nodes.

I looked for some articles to resolve problem and find these:
http://technet.microsoft.com/en-us/library/cc773451(v=ws.10).aspx
http://support.microsoft.com/kb/947049?wa=wsignin1.0
..and a lot of forums, same topics....

All cluster accounts, include CNO, computers (nodes), applications (services), placed in default location - "Computers" container.

I check permissions for container and for accounts:
____
For container "Computers"
 - CNO "Cluster$" have permissions "Full Control" - this object only.
____
For all accounts (CNO, computers, applications)
 - CNO "Cluster$" have special permissions: list contents + read all attributes + change password
____

Now cluster and application run correct. I moved applications between nodes, and all also working correct. But..
I can't find information how to "Check that the domain-wide quota for creating computer objects (by default, 10) has not been reached"
I think, that the error may be related with incorrect permissions for cluster accounts.

The problem I have is that I can't to make change in permissions (and other any changes), before argues the correct solution.
Also, I can't experiment in prodaction environment, and I haven't got a test environment to emulate same errors and trying resolve them.
And I must guarantee, that after I make changes - cluster will continue working properly, and I must argue the solution.

Someone have any expirience with successfull resolve this error?
Maybe there are reasons that I missed?
Please help me to fix this issue.
Thanks!

One of our clients has a Failover Cluster running on two Windows 2008 Datacenter R2 Servers and EqualLogic SAN. There are two VMs and they work great. They also have a SQL server on a physical server. With limit budge, they wonder if it is possible just converting SQL server to a VM in Failover Cluster. In other words, they don’t have money to hire consultant to create SQL Failover and buy new hardware. If yes, any disadvantages? If not, what are the other suggestions?

Bob Lin, MVP, MCSE & CNE Networking, Internet, Routing, VPN Troubleshooting on

http://www.ChicagoTech.net

http://www.howtonetworking.com

I thought I had posted this, but the system appears to have eaten it. I will try again.

I am setting up a 2-node Server 2012 Failover Cluster. I have installed the tools on each server, and I have shared storage waiting in the form of a Dell Compellent 8000 system. Each of the 2 nodes has 4 NICs:  #1 is External (they are both on the same public subnet, and can ping each other successfully), #2 is Internal (a private gigabit switch intended for the heartbeat, they are 10.0.0.101 and 10.0.0.102 on a 255.255.0.0 network and can ping each other), #3 and #4 are iSCSI switches that are part of the Compellent System.

When I run the Failover Cluster  Manager > Validate Configuration Wizard, and enter the same of the first node (The one I'm running it on), that's fine.  When I enter the second node, it says "The node cannot be contacted. Ensure that the node is powered on and is connected to the network."

Obviously it is, as I can ping between them on both external and internal connections. I also tested by creating a share on CL-01 (Node 1) and was able to access it via UNC path on Node 2.

Both servers are in our Active DIrectory, and when I Browse through the Validate Wizard, both servers appear in the OU. 

Is there something I may be missing? Can you tell me what port(s) this wizard may be using to communicate so that I can verify they're not being blocked somehow? It should not be the case, but I am at a loss to explain why these nodes can communicate in every sense except that one.

2003 cluster issue

Issue : MSDTC' in Resource Group 'Cluster Group' failed

cluster logs

=====================================================

FmpEnumerateGroupResources: Exit for group <a539af5c-9838-445f-8f21-a96806f0e706>....

000011ec.000011f8::2013/02/04-15:25:38.647 ERR  Distributed Transaction Coordinator <MSDTC>: Failed the IsAlive test. Current State is 1.

00000c6c.00001200::2013/02/04-15:25:38.663 INFO [FM] NotifyCallBackRoutine: enqueuing event

00000c6c.00001200::2013/02/04-15:25:38.663 INFO [FM] Calling RmNotifyChanges in monitor 11ec.

00000c6c.00000e04::2013/02/04-15:25:38.663 INFO [CP] CppResourceNotify for resource MSDTC

00000c6c.00000e08::2013/02/04-15:25:38.694 WARN [FM] FmpHandleResourceTransition: Resource Name = 96b27365-dcf4-49ca-bfff-7bef8cc313b8 [MSDTC] old state=2 new state=4

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [GUM] GumSendUpdate:  Locker waiting                      type 0 context 8       

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [GUM] Thread 0xe08 UpdateLock wait on Type 0

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [GUM] GumpDoLockingUpdate: lock was free, granted to 1

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [GUM] GumpDoLockingUpdate successful, Sequence=29752 Generation=0

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [GUM] GumSendUpdate: Locker dispatching seq 29752          type 0 context 8

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [GUM] GumSendUpdate: Dispatching seq 29752        type 0 context 8 to node 2

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [GUM] GumSendUpdate: Locker updating seq 29752               type 0 context 8

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [GUM] GumpDoUnlockingUpdate releasing lock ownership

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [GUM] GumSendUpdate: completed update seq 29752         type 0 context 8

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [FM] FmpPropagateResourceState: resource 96b27365-dcf4-49ca-bfff-7bef8cc313b8 failed event.

00000c6c.00000e08::2013/02/04-15:25:38.694 INFO [FM] FmpHandleResourceFailure: taking resource 96b27365-dcf4-49ca-bfff-7bef8cc313b8 and dependents offline

000011ec.00000e68::2013/02/04-15:25:38.694 INFO Distributed Transaction Coordinator <MSDTC>: In call to CommonTerminate

000011ec.00000e68::2013/02/04-15:25:38.694 INFO Distributed Transaction Coordinator <MSDTC>: Offline request.

000011ec.00000e68::2013/02/04-15:25:38.694 INFO Distributed Transaction Coordinator <MSDTC>: Service died; status = 1062.

000011ec.00000ca4::2013/02/04-15:25:42.038 INFO Distributed Transaction Coordinator <MSDTC>: Service is now on line

Event Type:        Error

Event Source:    ClusSvc

Event Category:                Failover Mgr

Event ID:              1069

Date:                     2/4/2013

Time:                     9:25:38 AM

User:                     N/A

Computer:          SNNHOUSQL1

Description:

Cluster resource 'MSDTC' in Resource Group 'Cluster Group' failed.

=====================================================

Hello,

I have read through talks about this, but I am not sure what I am missing. 

My cluster name is controlmDev.na.sas.com.  I have two services/applications:

controlmdevem

controlmdevsrv

These are both configured with different ip addresses and should register in dns.  If I look in active directory, CONTROLMDEVEM and CONTROLMDEVSRV have controlmDev$ listed with full control.

The log shows:

05:44.438 INFO  [RES] Network Name <Cluster Name>: adapter Local Area Connection 3 (2)
00000da8.000019d0::2013/02/06-14:05:44.438 WARN  [RES] Network Name <Cluster Name>: Trying to remove credentials for LocalSystem returned status C0000225, STATUS_NOT_FOUND is a non-critical failure for a remove operation
00000da8.000019d0::2013/02/06-14:05:44.453 INFO  [RES] Network Name <Cluster Name>: Initiating the Network Name operation : 'Verifying computer object associated with network name resource controlmDev'
00000da8.000019d0::2013/02/06-14:05:44.453 INFO  [RES] Network Name <Cluster Name>: Trying to find computer account CONTROLMDEV object GUID(f869e38fd6bb6140b2699224e577894f) on any available domain controller.
00000da8.000019d0::2013/02/06-14:05:44.453 INFO  [RES] Network Name <Cluster Name>: Found computer account CONTROLMDEV on domain controller\\shqnadc07.na.SAS.COM.
00000da8.000019d0::2013/02/06-14:05:44.453 INFO  [RES] Network Name <Cluster Name>: Trying to obtain the VSToken for Core Cluster Name resource
00000da8.000019d0::2013/02/06-14:05:44.469 INFO  [RES] Network Name <Cluster Name>: GetCoreNetnameObject_VSToken returning status 0
00000da8.000019d0::2013/02/06-14:05:44.469 INFO  [RES] Network Name <Cluster Name>: Obtained the security token for cluster name account.
00000da8.000019d0::2013/02/06-14:05:44.625 INFO  [RES] Network Name <Cluster Name>: Got new Logon Session.
00000da8.000019d0::2013/02/06-14:05:44.625 INFO  [RES] Network Name <Cluster Name>: Checking necessary attributes for computer account CONTROLMDEV
00000da8.000019d0::2013/02/06-14:05:44.640 INFO  [RES] Network Name <Cluster Name>: Successfully validated the attributes for computer account CONTROLMDEV
00000da8.000019d0::2013/02/06-14:05:44.640 INFO  [RES] Network Name <Cluster Name>: Initiating the Network Name operation : 'Updating alternate credentials associated with network name resource controlmDev'
00000da8.000019d0::2013/02/06-14:05:44.640 INFO  [RES] Network Name <Cluster Name>: Updating LSA credentials for Network Name resource CONTROLMDEV
00000da8.000019d0::2013/02/06-14:05:44.640 INFO  [RES] Network Name <Cluster Name>: Re-registering DNS records time period (151059 secs ) between last registration and now is greater than 4097398886784
00000da8.000019d0::2013/02/06-14:05:44.640 INFO  [RES] Network Name <Cluster Name>: Initiating the Network Name operation : 'Registering network name resource controlmDev DNS records'
00000da8.000019d0::2013/02/06-14:05:44.640 INFO  [RES] Network Name <Cluster Name>: Trying to obtain the VSToken for Core Cluster Name resource
00000da8.000019d0::2013/02/06-14:05:44.640 INFO  [RES] Network Name <Cluster Name>: GetCoreNetnameObject_VSToken returning status 0
00000da8.000019d0::2013/02/06-14:05:44.640 INFO  [RES] Network Name <Cluster Name>: DNS name controlmDev flushed from DNS resolver Cache
00000da8.000019d0::2013/02/06-14:05:44.671 ERR   [RES] Network Name <Cluster Name>: Failed to replace DNS A records for owner controlmDev.na.SAS.com over adapter 'Local Area Connection 3', status 9017
00000da8.000019d0::2013/02/06-14:05:44.671 ERR   [RES] Network Name <Cluster Name>: Failed to modify DNS A records for owner controlmDev.na.SAS.com over adapter 'Local Area Connection 3', status 9017
00000da8.000019d0::2013/02/06-14:05:44.687 INFO  [RES] Network Name <Cluster Name>: Initiating the Network Name operation : 'Performing NetBT related operations for the network name resource controlmDev'
00000da8.000019d0::2013/02/06-14:05:44.687 INFO  [RES] Network Name <Cluster Name>: Registered server name CONTROLMDEV on transport \Device\NetbiosSmb.
00000da8.000019d0::2013/02/06-14:05:44.687 INFO  [RES] Network Name <Cluster Name>: Initiating the Network Name operation : 'Registering name CONTROLMDEV on transport \Device\NetBt_If3'
00000da8.00000c74::2013/02/06-14:05:45.295 WARN  [RES] Physical Disk <Cluster Disk 1>: OnlineThread: Failed to set volguid \??\Volume{ca0a563e-5e91-11e2-8717-005056914599}. Error: 183.

No, I do not have Symatec running.  :)

I see DNS bad key messages related to dns when I look at the DNS Name properites for controlmDevEM and controlmDevSrv.

How do I fix this so it can replace th DNS A records?

Hi All,

I am trying to build a 2 node SQL cluster with 2008R2 servers in a vmware workstation environment using openfiler as shared storage.

1) installed application server role on both nodes.

2) installed failover clustering role on both nodes.

3) present 4 LUNS to both nodes.

4) while validating the cluster by adding both the nodes its getting failed , validating server A and B together fails whereas(report says access denied) , A on A passes and B on B passes.

vice versa A on B fails and B on A also fails . i am not even able to manage the servers fro each other as i am getting WINRM error , please help me to resolve this.

Also please answer the following queries

* after presenting the LUNS to both the nodes should i bring the disks online on both the nodes or just on the node where i am configuring the cluster.

*why is that if i create a folder on one node and it does show up on the shared disk on the shared disk.

Thanks in advance

Thanks

Shiv Ram

Hi Guys

I have a three node Windows Server 2008 R2 SP1 CSV cluster. I am interested in upgrading the environment to Windows Server 2012. 

1. Can this be done with an in-place upgrade????
2. Do I have to break this CSV and rebuild from scratch???
3. What is the recommended approach???

jk

I'm running Server 2012 Datacenter on VMWare 5.1.  I'm working on installing SQL 2012, and I want to use local storage for the AlwaysOn feature.

The problem I'm having is that when I try to add the disks, they will not come online.  I get an error that says Incorrect Function, Error Code: 0x80070001.

This article details the problem exactly, but there is no real solution here: http://support.microsoft.com/kb/2813005

Has anyone run across this before?  Is there an issue with these local disk appearing that they're on the same SCSI bus as the OS disk?

Hi,

I'm looking at running af 2-node failover-cluster in Server 2012 and would like to make use of the new cluster scheduled tasks ability.
Reading through a lot of guides online, i'm onboard with managing the tasks from powershell. So far i created tasks and bound them to a ressource.

But all the guides mention, that tasks should be visible in the Task Scheduler snap-in for MMC. I just cant find them.
I've checked the Failover Cluster section on both nodes (locally and on the role name) and the tasks is nowhere to be found.

What am i doing wrong?

Also. anyone knows how to manually start a scheduled task from powershell?

# This schedules a tasks...
# The batch file pipes text to a file

$action = New-ScheduledTaskAction –Exec E:\Jobs\test\Runme.bat
$trigger = New-ScheduledTaskTrigger –At 10:13 –Daily
$trigger.RepetitionInterval = (New-TimeSpan -Minutes 5)
$trigger.RepetitionDuration = (New-TimeSpan -Days 1)
Register-ClusteredScheduledTask –Cluster “srv2012cl1n1” –TaskName kasper-test –TaskType ResourceSpecific –Resource “Cluster Disk 1” –Action $action –Trigger $trigger

Kasper