Quantcast
Channel: High Availability (Clustering) forum
Viewing all 5648 articles
Browse latest View live
↧

Server 2012 R2 Hyper-V Failover Cluster NIC Team "Host Unmanageable" Status

September 18, 2015, 1:58 pm
$
0
0

Hi All.  I'm trying to troubleshoot an apparent network issue on our 2 node failover cluster, and have found that on one of the nodes, the status of the NIC team is "Host Unmanageable" in Server Manager.  In Local Server properties, it shows that NIC Teaming is Disabled.  

Additionally, the NIC Team is actually active and I'm not seeing any obvious connectivity issues (this team is being used by 3 vNICS for Live Migration, CSV and Management networks).  The cluster fully passes the Network portion of the Cluster Validation with no issues.  Additionally, the MsLbfoProvider Operational log has no errors over the last 2 years.  I can't find any logs that pinpoint when exactly this happened.

The issue we're having, and I'm not sure if it's related or not, is that I can't transfer ownership of any of the CSV disks from this node to the other.  If I manually disable the vNIC for the CSV network, I see the quorum disk attempt to move to the other node, but then it goes offline and the entire cluster disappears, as does access to all the CSV disks that are owned by this node.  Again, I'm not sure if this is related to this teaming issue or not.

I've seen a couple of posts regarding the "Host Unmanageable" status, but in those situations, the team isn't functioning and there are blatant connectivity issues.

I'd rather not recreate the team because the current issue with the cluster means the entire cluster has to be shut down (there's 30 VMs on it.)  Of course if I have to I will.

Does anyone have any suggestions on what I can do?  I've attached a few screenshots which might be helpful.  Thanks!

George Moore

↧
Search

Solution for DHCP Server High availability!!!

September 20, 2015, 8:02 pm
$
0
0

Good morning everyone!!!

I'm making plan to deploy DHCP Server with High availability to avoid any risks when one of them has been broken.

My currently system including: 2 DC with DHCP Server and 1 DC without (all is using Windows Server 2008 R2)

- 2 DHCP operates with splitting DHCP Scopes to load balancing. For example: PDCSVR (10.192.6.1 - 10.192.6.111); BDCSVR (10.192.6.112 - 10.192.6.254).

- the number of DHCP clients is around 200 notes.

Now I want to setup HA for them, to ensure that, If PDCSVR or BDCSVR has been broken. 200 clients will be allocated dynamic IP

I'm paying attention to DHCP Failed over cluster.

Please help me some favour:

1. Do I need to use Failed over cluster for DHCP, If needed, What do I need to prepare?

2. Is there any solutions for this situation?

Thanks so much


 


↧
↧

2 Node Hyper-v 2012 R2 Cluster failover issue

September 21, 2015, 2:48 am
$
0
0

Hi All,

i  am having and issue 2 Node Hyper-v 2012 R2 Cluster failover,  when i shut down when of the node the VMs failover without any issues but when i disconnect the network adapter the VMS will not failover, can some one please let me know if it is by design.

thanks

↧

Requirements for similar hardware in a 2012 Hyper-V failover cluster?

May 25, 2013, 5:51 am
$
0
0

Simple question:

How similar does the hardware have to be on the host servers in a Windows Server 2012 Hyper-V failover cluster?

We have 6 dell servers that we are hoping to use for building a Windows 2012 Hyper-V cluster.

As I recall from our existing 2-host Hyper-V cluster, which is running on 2008 R2, there are fairly harsh demands in the cluster validation wizard with regard to similarity of hardware. So I am wondering if it will work?

The servers in question are 3 x R620, 2 x T410 and 1 x T610 Dell servers. So 3 x 12th generation servers and 3 x 11th generation servers

The servers all use Intel CPUs:

R620: Xeon E5-2670/2680

T410/610: Xeon X5650

and broadcom NICs:

T610: Broadcom 5709c

2 x T410: Broadcom 5716c

3 x R620: Broadcom 5720

Windows 2012 is supported by Dell on all the servers.


Shared storage will be on an SMB3 file share.
↧

Windows Server 2012 R2 Failover Cluster - Network(Production + Management)

September 21, 2015, 8:26 am
$
0
0

Hi All,

The our exist setup have one cluster for two Nodes, each node have 4 network interface make two NIC as teaming finally we have for each nodes Two network teaming name [ Management & Production]

Cluster network:

Assigned  "Management" 10.10.8.0/21 as cluster network only.

VLAN setting for Teaming:

"Management" assigned to VLAN ID:10

"Production"     assinged to VLAN ID: default   ( have multi Vlan's like [20,21,30,31,110,120,40,26,27,24])

All our VM's machines connect to "Production" network  in this case if "Production" come down interface that mean all VM's on one node will be stay in the same node without any task of fail-over clustering which mean they doesn't have any connectivity network and it will work as local VM's in one node.

Now my request how can I add the "production" network to be second NIC in cluster network and the same time all VM's which work with different VLAN's likes [20,21,30,31,.......]?

Please if need any more information ?

Thank you.

Hasanain Ghanem

↧
↧

Cluster Network Failed on One Node

September 22, 2015, 7:03 am
$
0
0

Hello Experts,

 

I am looking for your suggestions on below described issues with Multi-Site MSFC on 2012 R2.

We have 2 Physical Fujitsu Primergy  BX920 S4 servers with CNA cards located both at different sites.

We have created 4 x Virtual NIC on each Physical server using OneCommand Manager. further these NICs are teamed.

Now we have 2 x NICs for each server as below.

NIC1 (VLAN-A  IP-X.X.12.24), NIC2 (VLAN-B IP-X.X.20.10) for Node1

NIC3 (VLAN-A IP-X.X.13.24), NIC4 (VLAN-B IP-X.X.111.10) for Node2

VLAN-A is configured for Cluster Heartbeat without default Gateway.

VLAN-B is configured for Production and both NICs have Default Gateway.

While looking at Cluster Manager console for Network Component, we  observed NIC1 is showing DOWN. and getting event in Event Logs as below.

 

Cluster network 'Cluster Network 1' is down. None of the available nodes can communicate using this network. Run the Validate a Configuration wizard to check your network configuration. If the condition persists, check for hardware or software errors related to the network adapter. Also check for failures in any other network components to which the node is connected such as hubs, switches, or bridges.

We have verified the network communication from NIC1 to Switch is working fine. The HeartBeat Network between both node is going through Switch only, there is no firewall.

We have tried to create static routes however both nodes, however while performing tracert to IP-X.X.13.24 from node1, The traffic is going through NIC2 (VLAN-B IP-X.X.20.10), where as it should go through NIC1.

 

Please suggest how can we troubleshoot it further.

 

 

 

 

 

Cluster network 'Cluster Network 1' is down. None of the available nodes can communicate using this network. Run the Validate a Configuration wizard to check your network configuration. If the condition persists, check for hardware or software errors related to the network adapter. Also check for failures in any other network components to which the node is connected such as hubs, switches, or bridges.

Thanks & Regards,
Amit Katkar (MCITP Windows 2008)
------------------------------------------------------------
This posting is provided "AS IS" with no warranties or guarantees and confers no rights.

↧

Failover Clustering with CSV without dedicated SAN problem

November 2, 2012, 5:58 am
$
0
0

I am trying to create a cluster in a box using vmware.

I have two 2012 member servers which will participate in the cluster and a 2012 domain controller.

this is a home lab so I don't have dedicated san.

first I tried to create a storage on the domain controller which actted as an iscsi target but found a bug in 2012 that fails the cluster validation on "iscsi page 83h vpd descriptor"

this failed so I condisidered using cluster shared volumes so I created 3 vhd on one of the members creates storage pool and added it to the cluster but I get status "failed" on cluster pool 1.

I fail to bring the cluster pool online and when I click information details I get error code 0x80070016 the device does not recognize the command.

what am I doing wrong?

thanks for the help Rotem

↧

Theory question on guest cluster storage

September 23, 2015, 2:54 am
$
0
0

Hello!

Would you please tell me is it affordable to use, for instance, the same iSCSI target for creating a guest cluster as the target being used for the "physical" cluster itself:

Thank you in advance,

Michael

↧
Search

SQL 2014 instance Name

September 23, 2015, 4:12 am
$
0
0

Dear All DBA ,

I would like to know some tips related with SQL , Presently we have SQL server 2014 Enterprise Edition and we made cluster

Cluster Name  SQl2014subash.soft.com IP 192.168.150.215 subnet 192.168.150.0  it's working fine there is no problem at all

Can we use same instance name ( SQl2014subash.soft.com ) in another Node with different IP and sub-net ( 192.168.170.0) is it possible ? or what are the best way .. we would like to use same Instance Name ...

Regards

Subash

↧
↧

Getting Below Error on the time of Network Validation Task (Win2k12)

September 23, 2015, 9:44 am
$
0
0

Network interfaces - Heartbeat and  - Heartbeat are on the same cluster network, yet address  is not reachable from  using UDP on port 3343.

Regards

RD


↧

CSV volumes: File Redirection mode

September 24, 2015, 4:01 am
$
0
0

Hello all,

I'm trying to understand what is CSV File Redirection mode and see the following:

1) http://blogs.msdn.com/b/clustering/archive/2013/12/02/10473247.aspx


2) http://blogs.msdn.com/b/clustering/archive/2013/12/05/10474312.aspx

The second article uses exact the same picture to illustrate the File System Redirected mode as the picture illustrating File Metadata redirection in the first one. If the second article is correct then, according to the first article stating"Metadata are always forwarded to NTFS. On the coordinator node CsvFs will forward metadata IO directly to the NTFS volume while other nodes will use SMB to forward the metadata over the network.", all non-coordinator nodes should ALWAYS have their csv volumes in File Redirection mode that contradicts the reasons of the File System Redirected mode described in the second article itself(User request, IncompatibleFileSytemFilter...).

Q1: Can\must metada redirection trigger the File System Redirected mode?

Q2: What requests are redirected in the File System Redirected mode - a) metadata b) read/write or both?

Q3: If "right clicking on a CSV volume and selecting the Turn On Redirected Access option" initiates theFile System Redirected Mode how is it possible to initiate a Block Level Redirected IO?

Q4: When one of the nodes suddenly looses connection to the shared storage will CSV switch toFile System Redirected Mode or Block Level Redirected IO Mode?

Thank you in advance,

Michael


↧

WAN issue adding/creating cluster

September 24, 2015, 1:01 pm
$
0
0

 I have several sites connect with Cisco ASAs all running Windows Server 2012r2 and for example I can setup a failover cluster from site 1(serverX) to site 3 (serverY), from serverX to serverX1 on the same site, and serverZ to serverZ1 on site 2.   The issue is when I try setup a failover cluster from site 2 to the other sites.

example 1

try to setup a failover cluster from site 2 VM serverZ to Site 1 VM ServerX.

Running Cluster Validation with all test will immediately freeze up on "List Disk" At 0%

Running Cluster Validation with all but Storage will finish with only warnings.  example is I have one network connection so it is warning about single point of failure.

when I continue with creating the cluster it will then lock up on getting Network Topology.   Box to enter Cluster name stays grayed out and subnets never show up.

Networks are 

10.19.0.0/12 site 1

10.35.0.0/12 Site 2

10.53.0.0/16 Site 3

10.54.0.0/16 Site 4....

Example 2

created cluster with two server from site 2  serverZ and serverZ1  worked fine

add serverX from site 1  

hangs up at "configure the Cluster"

errors I see here in Event Viewer - FailoverClustering-Client-Diagnostic is:

ApiQueryValue exited with cluster_error = 234 

ApiClusterControl exited with cluster_error = 234

Other then the above I can't find any logs for the cluster Wizard 

My Cisco ASA guy said all the ports are open for site to site communication.

Firewalls on the server are all turned off.

I also get the same result if I run the Cluster wizard from Site 1 serverX  to Site 2 serverZ

I am at my wits end.  I have tried different server on site 2 with the same result.   I am having the same issue with Exchange DAGs on this site.

any suggestion for the issue or ideas to get better log data for the create cluster wizard?

Thanks

↧

Enable Run this resource in a separate Resource Monitor in the VM

September 24, 2015, 7:55 pm
$
0
0

Hi folks,

In my environment I have a Failover Cluster with 5 nodes, with Windows Server 2012 R2.

I would like know if is recommended enable the resource of RHS, because the Microsoft explains that resource is recommended to be debugged or is likely to conflict with other resource type DLLs

In this case is better use Anti-Affinity for the VMs?

Another point is configuring the Deadlock timeout?

The Microsoft explain that RHS are recommended for unstable resources. May I to adopt that this not apply for all VMs?

What is the better recommendation and steps to resolve this trouble?

Thank you very much!

Wilsterman Fernandes


↧
↧

IP Heartbeat DNS Setting

September 24, 2015, 10:40 pm
$
0
0

I have a simple cluster with one ip address and one heartbeat address.

on DNS record are like this

servername = 10.10.10.56  IPaddress of the server.

servername = 10.10.11.46  IP address of the heartbeat.

server name is only one name for both Server and IP Heartbeat.

when i ping the server i'm getting the Heartbeat address.

when i used nslookup , I'm getting the heartbeat address.

I suppose to get the server ip address not the heartbeat.

so my question is 

how to solve this problem?

do i need to change the setting

Server was working fine.

↧

IP Heartbeat DNS Setting

September 24, 2015, 10:39 pm
$
0
0

I have a simple cluster with one ip address and one heartbeat address.

on DNS record are like this

servername = 10.10.10.56  IPaddress of the server.

servername = 10.10.11.46  IP address of the heartbeat.

when i ping the server i'm getting the Heartbeat address.

when i used nslookup , I'm getting the heartbeat address.

I suppose to get the server ip address not the heartbeat.

so my question is 

how to solve this problem?

do i need to change the setting

Server was working fine.

↧
Search

cluster aware updating windows update

September 25, 2015, 5:32 pm
$
0
0

I got a 2 node 2012r2 server running sql 2012. I have cluster aware updating installed but I see windows updates are available via windows updates, I do not see the updates under cluster aware updating.

any idea? 

what does cluster aware update update?

↧

Failover Happened on RTO?

September 2, 2015, 11:16 am
$
0
0

Hello Everyone,

i'm using windows 2012 R2 failover cluster, can anyone help me to understand how manyRTO occurred on the network that initiate the failover to another Node.

↧
↧

CSV volumes: File System Redirected mode

September 24, 2015, 4:01 am
$
0
0

Hello all,

I'm trying to understand what is CSV File System Redirected mode and see the following:

1) http://blogs.msdn.com/b/clustering/archive/2013/12/02/10473247.aspx


2) http://blogs.msdn.com/b/clustering/archive/2013/12/05/10474312.aspx

The second article uses exact the same picture to illustrate the File System Redirected mode as the picture illustrating File Metadata redirection in the first one. If the second article is correct then, according to the first article stating"Metadata are always forwarded to NTFS. On the coordinator node CsvFs will forward metadata IO directly to the NTFS volume while other nodes will use SMB to forward the metadata over the network.", all non-coordinator nodes should ALWAYS have their csv volumes in File Redirection mode that contradicts the reasons of the File System Redirected mode described in the second article itself(User request, IncompatibleFileSytemFilter...).

Q1: Can\must metada redirection trigger the File System Redirected mode?

Q2: What requests are redirected in the File System Redirected mode - a) metadata b) read/write or both?

Q3: If "right clicking on a CSV volume and selecting the Turn On Redirected Access option" initiates theFile System Redirected Mode how is it possible to initiate a Block Level Redirected IO?

Q4: When one of the nodes suddenly looses connection to the shared storage will CSV switch toFile System Redirected Mode or Block Level Redirected IO Mode?

Thank you in advance,

Michael




↧

CentOS 7 VM's losing disk after Live Migration

September 28, 2015, 3:48 am
$
0
0

Hi!

Some of my CentOS 7 VM's are losing disks after live migration (node1 to node2). I have to reset VM to get them back. It happens with 6 of my VM's, about 25 are Ok. I noticed this problem two live migrations ago, before that I had no problems. I have patched and restarted Hyper-V hosts. I'm using CSV (LUNs from SAN). All the VM's are made from same template.
Where should I start researching the problem?



↧

IP address conflict in windows clustring on vmware

September 28, 2015, 2:38 am
$
0
0

I am doing windows clustering on virtual machines ,windows server 2008 r2 as my base machine and has separate network address like 192.168.0.6 is able to access internet.

I am having three vm machines installed having windows server 2008 r2 OS.

My DC machine has ip address as 192.168.103.142

Clusternode1 has 192.168.103.143 and clusternode2 has IP as 192.168.103.144.

All the vm machine's network adapter settings are using "Custom"  and uses "vmnet2" for public network and vmnet3 for private network.I have configured private network also for three nodes and are having IP as 10.0.0.6 ,10.0.07 and 10.0.0.8 for these machines respectively. Now I sucessfully was able to install sql server cluster and was working fine but then I got IP conflict  error message that showed thatmy windowscluster IP adress is duplicate i.e 192.168.103.155.Now I wonder what has gone wrong and when my network adapter settings are based on "custom" how I got this conflict. My sql network IP is 192.168.103.175.

Please if anyone share the knowledge on it as networking knowledge is limited..where things have gone wrong in setup.

srm_2010

↧
Viewing all 5648 articles
Browse latest View live
Search