Quantcast
Channel: High Availability (Clustering) forum
Viewing all 2306 articles
Browse latest View live

Is NetShareAdd still valid for Windows Server 2012 within a Failover Cluster?

$
0
0

We are trying to use NetShareAdd to create a share on the system and have it recognized in the clustered resource.  So the clustered resource name is "ExileOnMainSt" and thus:

SHARE_INFO_503 si503 = { 0 };
si503.shi503_servername = L"EXILEONMAINST";
si503.shi503_netname = L"TSV"
si503.shi503_type = STYPE_DISKTREE;
si503.shi503_remark = L"";
si503.shi503_permissions = ACCESS_READ;
si503.shi503_max_uses = 0xFFFFFFFF;
si503.shi503_path = L"S:\\TSV";
DWORD dwParmErr = 0;
NET_API_STATUS eNetErr = NetShareAdd(L"EXILEONMAINST", 503, (LPBYTE) &si503, &dwParmErr);

I'm receiving ERROR_INVALID_PARAMETER but dwParmErr is zero when the spec says it should be one of the following:

#define SHARE_NETNAME_PARMNUM         1
#define SHARE_TYPE_PARMNUM            3
#define SHARE_REMARK_PARMNUM          4
#define SHARE_PERMISSIONS_PARMNUM     5
#define SHARE_MAX_USES_PARMNUM        6
#define SHARE_CURRENT_USES_PARMNUM    7
#define SHARE_PATH_PARMNUM            8
#define SHARE_PASSWD_PARMNUM          9
#define SHARE_FILE_SD_PARMNUM       501
#define SHARE_SERVER_PARMNUM        503

This works in Windows Server 2008 R2. Has the functionality changed?


Adam



Clustered Storage Spaces architecture

$
0
0

Hi!

During recent TechEd sessions there were some talks about Storage Spaces and how to scale out SoFS cluster (for example, http://channel9.msdn.com/Events/TechEd/Europe/2013/MDC-B218#fbid=F3Zyz8qjfcO, 28 — 35 min). It was shown that we can build a cluster using 4 servers and a common JBOD - it makes up 1 unit. Then we can add more such units in the same storage cluster.

The interconnection inside one unit is SAS, the interconnection between units is Ethernet. And it works as one file server cluster, it can survive the failure of the whole unit. That means that data can be mirrored across different units. If so, what prevents us from having one server and one JBOD per unit? And if so, why can't we use local SAS disks instead of JBOD for each server?

The picture at 15 min shows this 1-server-per-unit scenario but all the documentation I could found implies that it's infeasible.



Failovercluster BSOD

$
0
0

We just got our second BSOD on a failovercluster box in 2 weeks. The setup is 3 server 2012 boxes (dell r610) with a 10 gig nic for san and the onboard for the public.

the minidump shows netft.sys as the culprit 

ADDITIONAL_DEBUG_TEXT:  
You can run '.symfix; .reload' to try to fix the symbol path and load symbols.

MODULE_NAME: netft

FAULTING_MODULE: fffff802f5683000 nt

DEBUG_FLR_IMAGE_TIMESTAMP:  5010aa07

PROCESS_OBJECT: fffffa80ac6e46c0

CUSTOMER_CRASH_COUNT:  1

DEFAULT_BUCKET_ID:  WIN8_DRIVER_FAULT_SERVER

BUGCHECK_STR:  0x9E

CURRENT_IRQL:  0

LAST_CONTROL_TRANSFER:  from fffff8800660d845 to fffff802f56dd440

STACK_TEXT:  
fffff880`0316a938 fffff880`0660d845 : 00000000`0000009e fffffa80`ac6e46c0 00000000`0000003c 00000000`00000000 : nt+0x5a440
fffff880`0316a940 00000000`0000009e : fffffa80`ac6e46c0 00000000`0000003c 00000000`00000000 00000000`00000000 : netft+0x2845
fffff880`0316a948 fffffa80`ac6e46c0 : 00000000`0000003c 00000000`00000000 00000000`00000000 fffff802`f5636dd4 : 0x9e
fffff880`0316a950 00000000`0000003c : 00000000`00000000 00000000`00000000 fffff802`f5636dd4 fffffa80`4cdecc30 : 0xfffffa80`ac6e46c0
fffff880`0316a958 00000000`00000000 : 00000000`00000000 fffff802`f5636dd4 fffffa80`4cdecc30 fffff880`0660d516 : 0x3c


STACK_COMMAND:  kb

FOLLOWUP_IP: 
netft+2845
fffff880`0660d845 ??              ???

SYMBOL_STACK_INDEX:  1

SYMBOL_NAME:  netft+2845

FOLLOWUP_NAME:  MachineOwner

IMAGE_NAME:  netft.sys

Any thoughts on what i should do?

Keep Repair CNO to Live Migrate VM

$
0
0
I've 2 node cluster -Win 2012 Hyper-V. Cluster is working fine and can live migrate without any problem. But after a while (few days) saw error message on dashboard mentioned CNO -kerberos security error. When this happen, i cannot perform live migration. Normally what i did was go to CNO, Simulate Failure and then repair. Well, it fixed the problem and able to live migrate. However this problem keep appearing and i use back the same method. Any idea why this happen and prevent to repair. CNO is located in the default Computer container. Please advise

Lai (My blog:- http://www.ms4u.info)

Use legacy CSV LUNs in new Hyper-V 2008 R2 Cluster

$
0
0

Just wondering if it's possible/supported to move several CSV LUNs from an existing 2008 R2 Hyper-V Cluster to a new one that is going to be built?

The exisitng SAN will remain but new Hyper-V Hosts will be provisioned to replace the current ones and all existing VMs need to be migrated.

Cheers for now

Russell

3 Nodes SQL 2008 cluster failover scenario.

$
0
0

I want to create a 3 node cluster of SQL 2008 over windows 2008 R2 cluster.

If I use Majority Nodes with Disk as the quorum then:

1) Cluster will continue working after the failure of 1 Node.

2) Cluster will continue to work after the failure of disk-witness.

Question:

1) Will cluster continue to work if Disk-witness is online and 2 Nodes are down.

2) If NO, then How am I achieving the high-availability of SQL database even when I am having One node and one Disk-witness available. Because even if I have all 3 nodes available still the SQL database service will be served by only one node. The fail-over cluster does not load balance the user load among the participating servers.


Nitesh Pathak

Clustered CA failover but the CA appl not started

$
0
0
After changing the Clustered CA servicename, and testing the CA failover, I see that both the storage and the CA servicename failed over successfully but the CA apllication itself did not start, unless I  manually start it in the new node.  Am I missing something?   I have the auto-start option selected in the cluster Service and Applications.   It seems the servicename is decoupled from the actually CA application.  Any clue will help.

How to set private properties of a cluster resource using WMI?

$
0
0
Hi,

I wanted to know if there is an easier way to set or modify private properties of cluster resource using WMI from managed code (C#). One way I can think of is using ExecuteResourceControl Method of the MSCluster_Resource class to invoke CLUSCTL_RESOURCE_SET_PRIVATE_PROPERTIES control code. Is there a simpler way?

Thanks in advance for any help,
Anagha

Cluster event log errors

$
0
0
Not sure where the problem is but my cluster, services and applications in the cluster seem to be running properly however I keep seeing the same event id errors in my log: 1069, 1205, and 1214 over and over about every hour. Is this normal? or is there something going on that i'm not aware of. This also happens on 2 other isolated separate systems in my lab.

Windows 2012 Hyper-v Cluster Live Migration Failure

$
0
0

This is a fairly new deployment.  When I first started having this issue, I foundhttp://support.microsoft.com/kb/2779204, which suggests that gpupdate /force will temporarily resolve the issue.  That worked for me for several weeks, but no longer.  It was always temporary and I had to do it at least once a day, but even that doesn’t work anymore and my Hyper-V guests are stuck.  I also suspect if I have a Hyper-V host failure, the guests on the failed system will not come back up on the operational host.

kb2779204 suggest adding NT Virtual Machine\Virtual Machines in the entries for Log on as a Service.  I have created a dedicated OU for the virtual hosts (as suggested by the kb article) and moved them to the new OU. I then addedNT Virtual Machine\Virtual Machines in the entries for Log on as a Service, then gpupdate /force and then wait minutes, then hours. It still doesn’t work.  I get 2 slightly different depending on which server I initiate the live migration.

If from the server that currently is running the guest (HV1):

Live migration of 'Virtual Machine vg1' failed.

Virtual machine migration operation for 'dc4' failed at migration source 'HV1'. (Virtual machine ID 140C8893-44EC-481B-B8D5-52FCB8D422DC)

The Virtual Machine Management Service failed to establish a connection for a Virtual Machine migration with host 'HV2': General access denied error (0x80070005).

The Virtual Machine Management Service failed to establish a connection for a Virtual Machine migration because the destination host rejected the request: General access denied error (0x80070005).

If from the server attempting to move the guest to (HV2):

Live migration of 'Virtual Machine vg1' failed.

Virtual machine migration operation for 'vg1' failed at migration destination 'HV1'. (Virtual machine ID 140C8893-44EC-481B-B8D5-52FCB8D422DC)

'vg1' Failed to create Planned Virtual Machine at migration destination: Logon failure: the user has not been granted the requested logon type at this computer. (0x80070569). (Virtual machine ID 140C8893-44EC-481B-B8D5-52FCB8D422DC)

how to delete a failed file share witness

$
0
0

I have a 2 node 2008SP2 (Geographically separated) cluster. I use file and node majority for quorum. There is a file share at third site and both nodes can ping/open the share and the nodes have full controll of the share. Also, there are other cluster/nodes using the same share for quorum. This has worked for several years. After the last round of MS patches the file share witness resource would not come back online for 1 cluster. I tried forcing it online several times but I cancelled after 10 minutes or so. I found that I can add another folder in the original location and reran the qurom wizard, and it came online. I am guessing that maybe the multiple clusters using the same single folder is not good, not sure why it worked for so long though. Anyway, the delete function is gone from the failed file share witness, how can I get rid of the old failed resource? 

 

Thanks

~M

Network Adapter Question (Network Location)

$
0
0

I have a W2K8R2 SP1 (Datacenter Edition) Failover Cluster. Each node has 8 network interfaces: 4 External Virtual Networks connected to company network switches, 2 iSCSI interfaces (CSV) and 2 management interfaces on different subnets. NW3 (subnet 10, same subnet as the domain controller) is listed as a "Domain" network. NW4 (subnet 20) is listed as a "Public" network. With the exception of the External Virtual Networks, all cables are routed to a pair of HP V1810G 48 port gigabit switches trunked together. Each switch has 2 VLANs, one for iSCSI only and another for all others.

I was expecting NW4 to be listed as a domain network as well. Perhaps I have misconfigured something? Should I create another VLAN for subnet 20? Use more switches? Add a NIC to the domain controller on subnet 20? Maybe I need to register the subnet 20 IP addresses in DNS?

I am not a network engineer so I don't understand all of the intricacies involved. The network diagrams I have seen in researching this are line diagrams and not a physical representation of the network.

Thanks for your assistance, Vint



Unable to add Node to cluster - Windows Server 2012

$
0
0

We are having troubles trying to add a node to a cluster where there is an AlwaysOn Availability Group role, all systems are using Windows Server 2012. This node was previously working without problems, but was evicted during an operation, and we are trying again to add this node to the claster.

The cluster is working OK, because we have other avaibility groups working in the same cluster. Both cluster and node are in the same network and domain.

The cluster validation was completed without  errors. This is the report generated after the unsuccessfull adding process:

Node:
VM_name.domain
Started
24/07/2013 15:35:08
Completed
24/07/2013 15:41:30
Adding VM_name.domain to the cluster.
Validating cluster state on node VM_name.
Getting current node membership of cluster Cluster_name_.
Adding node VM_name to Cluster configuration data.
Validating installation of the Network FT Driver on node VM_name.
Validating installation of the Cluster Disk Driver on node VM_name.
Configuring Cluster Service on node VM_name.
Waiting for notification that Cluster service on node VM_name.domain has started.
Waiting for notification that node VM_name is a fully functional member of the cluster.
Unable to successfully cleanup.
The server 'VM_name.domain' could not be added to the cluster.
An error occurred while adding node 'VM_name.domain' to cluster 'Cluster_name'.

This operation returned because the timeout period expired


Any hint about how to resolve this problem would be very appreciated.

VM_name.domain
VM_name.domain

Creating share in cluster wih 2 noeds

$
0
0

Hi

I have two Windows 2008R2  in clustering with all disk on SAN. I need to create Hight avaibility share.

I create a access point. Create a object and ip address in the network. I define a share and the service seems to be ok but 

when I try to swith to share to cluster servers, it says that the Identity cluster do not have enough right to update the object.  Any ideas ?

Best regard

How to set private properties of a cluster resource using WMI?

$
0
0
Hi,

I wanted to know if there is an easier way to set or modify private properties of cluster resource using WMI from managed code (C#). One way I can think of is using ExecuteResourceControl Method of the MSCluster_Resource class to invoke CLUSCTL_RESOURCE_SET_PRIVATE_PROPERTIES control code. Is there a simpler way?

Thanks in advance for any help,
Anagha

Cluster event log errors

$
0
0
Not sure where the problem is but my cluster, services and applications in the cluster seem to be running properly however I keep seeing the same event id errors in my log: 1069, 1205, and 1214 over and over about every hour. Is this normal? or is there something going on that i'm not aware of. This also happens on 2 other isolated separate systems in my lab.

Windows cluster witness server problem

$
0
0

Hi,

One of our client has Exchange server 2010 with Hyper-v and they are running two site DC and DR. Witness server are located in DC site with other physical server, this server is also running file server service. When they down DC's Exchange server then Exchange server are running fine from DR site. But when they down witness server after 1 to 2 hour's later then DR Exchange Database is dismounted state and Exchange server is not working. Their High availability of exchange server with DAG for mailbox. Why this problem occur.

Windows CA 2012 in geographic Cluster ... How to

$
0
0
Helloto all.

I have toinstall 2windows server2012with Certification Authority,in ageographic cluster(2different sites,butthe same domain).

I read that youneedredundanthardware(servers and storage).

Moreover, in caseofnetworkfailure,the Cluster servicerequiresa third "vote" tobe able to start.

I wanted to askif everything isright, and ifthere isdocumentationon the internet thatexplainshow to implement it.

Thank you all forthe help.

/Mino

Cluster node addition fails on cleanup

$
0
0

We have a 2 node cluster setup already

(2) HP BL460c G8 servers connected to a VNX5300 SAN (Nodes 1 & 2)

Server 2012 Datacenter installed

Quarum: Node + Disk

all failover tests went perfectly and all VMs are healthy

Verification on the cluster show some warnings but no failures

We have rebuilt a server (node 3) renamed it and have run a single machine verification test to see if it is suitable for clustering. it succeeded with minor warnings

We ran verification on all three machines and received the formentioned warnings but no game stoppers, however when trying to add the host to the cluster we get the following error in the logs:

WARN mscs::ListenerWorker::operator (): ERROR_TIMEOUT(1460)' because of '[FTI][Initiator] Aborting connection because NetFT route to node <machine name> on virtual IP fe80::cdf2:f6ea:5ce:5f9c:~3343~ has failed to come up.'

This happens after the node is added to the cluster but reports a failure on cleanup processes and reverts everything back. I have done all of this under my domain_admin account.

before and after the attempt to add the NetFT adapter is in media disconnect, during the attempts it does pull down a 169 address as it is supposed to

Node 3 Networking breakdown

The new host uses an Intel/HP NC365T Quard port adaptor

port 1: Mgmt : Static assignment subnet 1

port 2: VM net: Static assignment sibmet 2

port 3: Heartbeat: assigned via DHCP subnet 1 pool (we have attempted the above with this disabled as well)

NCU is not installed for the adapter and bridging in server 2012 is not enabled.

I am at a loss, and would appreciate any additional help as i have spent 3 days researching this to try and find the cause.

Error: The computer is joined to cluster when creating the Cluster

$
0
0

Hello Guys,

I have created a cluster to configure Hyper-V for 2 Nods, everything was greate and works perfectly, next day the storage hang and the cluster didn't work any more, I have destroyed the cluster the removed the cluster feature from both nods, deleted the cluster-computer from AD and the deleted the storage. after we fixed the storage, I have reconnect the storage, installed the cluster service on both nods, then I have validate the configuration and I had everything green 100%.

while creating the cluster, I faced an issue Unable "to successfully cleanup" I kept trying and removed the anti-virus, restarted the servers manytime, then I ended up to have another error, direclty when I add the server name on the creat cluster wizard, its telling me that the computer I'm adding is joined to cluster.

I think I need to do some cleaning to the previous cluster, can I have some help here ?

Regards..

Nour


Nour

Viewing all 2306 articles
Browse latest View live


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>