using cluster administrator console
I have never had a chance to install a cluster environment, but I have started in a company where they have cluster environment, I have been reading about the concept of clustering and how to set it up, but for now I think I need to know much about how to use the cluster administrator console, since most of the wrk is done from this console.
I can see
Groups
Resources
cluster configuration
1- I would like to knwo how/when to use those nodes.
2- I would like also to know where to go first when the cluster is down, assuming it's NOT a connectivity or hardware problem.
Thanks
I can see
Groups
Resources
cluster configuration
1- I would like to knwo how/when to use those nodes.
2- I would like also to know where to go first when the cluster is down, assuming it's NOT a connectivity or hardware problem.
Thanks
Depending on the xersion of the OS 2000 or 2003 will control this to a degree. 2000 allows 2 nodes; 2003 allows up to 8 nodes in the cluster.
Lets make it simple and assume you only have thw two nodes.
You will have identified a disk group and make that available using the same drive letter(s) visible by both nodes. You will then have installed clustering on the first node only. You will have created a cluster name and chosent how the nodes will be talking to each other (heartbeat) as well as how the users will connect to the cluster. This is normally over two different network cards.
You will assign a virtual IP address that will be used for the cluster.
You will have selected an disk area that both nodes will be able to see which holds the 'quorum' for the cluster, lets sy its your w: drive. (This is cluster information)
In your cluster admin console, you will now have cluster groups(s) based on the group you created and Resources that are included within the group. The group may be called printers for example and the resources required for the group may included printer shares, the virtual IP address, the W: and so on. ie all these resources put together make the printer cluster group.
Further down the conse, you will see the names of the nodes that form the cluster. As you look in the groups & resources within them, this is showing which mode is currently running the service at this minute in time. If the service fails, it will transfer itself to the other node in the list. By viewing each of these listings you will see its current state. For example, online, offline, pending, near-online etc.
As mentioned in your other cluster question. The first port of call if the whole cluster has failed is the event log on each node plus the cluster service logs in c:\ windows\cluster or c:\winnt\cluster depending on your installation.
To actually test clustering on a working system, there is an option in the admin console to move group. This forces the node to issue an interruption to the heartbeats and tells the other node to transfer the services to the satndby node. Remember, whatever you have heard, Microsoft does NOT support Active/Active clusters; only Active Passive.
Lets make it simple and assume you only have thw two nodes.
You will have identified a disk group and make that available using the same drive letter(s) visible by both nodes. You will then have installed clustering on the first node only. You will have created a cluster name and chosent how the nodes will be talking to each other (heartbeat) as well as how the users will connect to the cluster. This is normally over two different network cards.
You will assign a virtual IP address that will be used for the cluster.
You will have selected an disk area that both nodes will be able to see which holds the 'quorum' for the cluster, lets sy its your w: drive. (This is cluster information)
In your cluster admin console, you will now have cluster groups(s) based on the group you created and Resources that are included within the group. The group may be called printers for example and the resources required for the group may included printer shares, the virtual IP address, the W: and so on. ie all these resources put together make the printer cluster group.
Further down the conse, you will see the names of the nodes that form the cluster. As you look in the groups & resources within them, this is showing which mode is currently running the service at this minute in time. If the service fails, it will transfer itself to the other node in the list. By viewing each of these listings you will see its current state. For example, online, offline, pending, near-online etc.
As mentioned in your other cluster question. The first port of call if the whole cluster has failed is the event log on each node plus the cluster service logs in c:\ windows\cluster or c:\winnt\cluster depending on your installation.
To actually test clustering on a working system, there is an option in the admin console to move group. This forces the node to issue an interruption to the heartbeats and tells the other node to transfer the services to the satndby node. Remember, whatever you have heard, Microsoft does NOT support Active/Active clusters; only Active Passive.
What OS are you talking about? Most manufacturers understand something different when they talk about clusters.
I have tried to instal a virtual cluster environment in WXP, using virtual PC free trial, in order to have a good knowledge of it, but could not get through it successfully.
Our company has cluster environment W2000 - 2 nodes. I would like to learn how to use cluster administrator in order to add/remove resources/ and bring up the cluster when it goes down.
Our company has cluster environment W2000 - 2 nodes. I would like to learn how to use cluster administrator in order to add/remove resources/ and bring up the cluster when it goes down.
The failover process is automatic and to ne honest, the cluster cosole is not used if a cluster fails. It is more for identifying the status of the cluster and adding/removing groups and resources. The event logs/cluster logs are used to identify the issues with utilities such as ping and arp.
To add a resource, is different for each type. For example, adding a disk group is different from adding a print spooler which again is totally different to adding an application. Even applications break down into different types; those that are cluster-aware and those that are not. These forums are not th right place to go over all the different aspects. For that, you need a training course (Clustering is course 2087A by the way and covers clustering and NLB, Network Load Balancing, its a good course as well).
If you wish to pick one of your clusters and tell me what is in the group/resources I'll go through it with you. Alternatively, pick ONE application, and I'll go through a walk-through for you. What I cannot do is explain all the variances here though as I am sure you can appreciate.
To add a resource, is different for each type. For example, adding a disk group is different from adding a print spooler which again is totally different to adding an application. Even applications break down into different types; those that are cluster-aware and those that are not. These forums are not th right place to go over all the different aspects. For that, you need a training course (Clustering is course 2087A by the way and covers clustering and NLB, Network Load Balancing, its a good course as well).
If you wish to pick one of your clusters and tell me what is in the group/resources I'll go through it with you. Alternatively, pick ONE application, and I'll go through a walk-through for you. What I cannot do is explain all the variances here though as I am sure you can appreciate.
This one is on us!
(Get your first solution completely free - no credit card required)
UNLOCK SOLUTION
O, and one item to add to my last post. Clustering can be rebuilt but as I menion, this is hardly ever anything to do with the console. Backups: When rou restore a cluster, the quorum, cluster group etc they expect to go back onto the same disk drive as they were backed up from. You have the ability to export the drive signatures that hold the quorum and the cluster information. Do so otherwise you can resore the files/database bu the group will not come back online. You would use the clusrest.exe to restore the quorum but if it does not recognise the disk signature then no play.
let me expand the cluster administrator, sorry it's kind of too much data.
ClusterName
Groups Name STATE OWNER
CLuster Group online ServerName2
Roaming Profiles online ServerName1
Cluster Group Applic Disk Res online ServerName2
Cluster IP Addre online ServerName2
Logon log online ServerName2
Quorum drive online ServerName2
Print spooler online ServerName2
Users online ServerName2
Roaming Profiles Disk G online ServerName1
ServerName3-Profiles1 online ServerName1
ServerName3-Profiles2 online ServerName1
Roaming Profiles online ServerName1
TS Profiles online ServerName1
Resources Cluster IP addr online ServerName2
Quorum Drive online ServerName2
Cluster Name online ServerName2
Print Spooler online ServerName2
......
......
Cluster Configurations
Resource Types DHCP Service xxxx.dll
file share vvvv.dll
IP addr gggg.dll
.......
Networks
Private Network ServerName1 Private Network UP
ServerNAme2 Private Network UP
Public Network ServerName1 Public Network UP
ServerName2 Public Network UP
Network Interfaces ServerNAme1 Private Network UP
ServerName2 Private Network UP
ServerName1 Public Network UP
ServerName2 Public Network UP
ServerName1
Active Groups Roaming Profiles online ServerName1
Active Resources Disk G:
Roaming Profiles
TS Profiles
Network Interfaces ServerName1 Private Network UP
ServerName2 Public Network UP
ServerName2
Active Groups Cluster Group online ServerName2
Active Resources Cluster IP addr online ServerName2
Quorum Drive online ServerName2
Cluster Name online ServerName2
Print Spooler online ServerName2
Network Interfaces ServerName2 Private Network UP
ServerName2 Public Network UP
ClusterName
Groups Name STATE OWNER
CLuster Group online ServerName2
Roaming Profiles online ServerName1
Cluster Group Applic Disk Res online ServerName2
Cluster IP Addre online ServerName2
Logon log online ServerName2
Quorum drive online ServerName2
Print spooler online ServerName2
Users online ServerName2
Roaming Profiles Disk G online ServerName1
ServerName3-Profiles1 online ServerName1
ServerName3-Profiles2 online ServerName1
Roaming Profiles online ServerName1
TS Profiles online ServerName1
Resources Cluster IP addr online ServerName2
Quorum Drive online ServerName2
Cluster Name online ServerName2
Print Spooler online ServerName2
......
......
Cluster Configurations
Resource Types DHCP Service xxxx.dll
file share vvvv.dll
IP addr gggg.dll
.......
Networks
Private Network ServerName1 Private Network UP
ServerNAme2 Private Network UP
Public Network ServerName1 Public Network UP
ServerName2 Public Network UP
Network Interfaces ServerNAme1 Private Network UP
ServerName2 Private Network UP
ServerName1 Public Network UP
ServerName2 Public Network UP
ServerName1
Active Groups Roaming Profiles online ServerName1
Active Resources Disk G:
Roaming Profiles
TS Profiles
Network Interfaces ServerName1 Private Network UP
ServerName2 Public Network UP
ServerName2
Active Groups Cluster Group online ServerName2
Active Resources Cluster IP addr online ServerName2
Quorum Drive online ServerName2
Cluster Name online ServerName2
Print Spooler online ServerName2
Network Interfaces ServerName2 Private Network UP
ServerName2 Public Network UP
This one is on us!
(Get your first solution completely free - no credit card required)
UNLOCK SOLUTION
Only close it if it has helped. If you need something more (short of a course on Clustering) then let me know. i can talk about it for hours but need specific questions to respond to. Is there a part of your config you are not sure about? Is there a part you want clarifying?
Would it help to pull the clustering 'how to do anything/everything from my ftp site'? plus all the MS technotes for it?
keith_alabaster
I haven't closed the question otherwise I will forget to print it out. I have been busy, but I will do all my best to print your explanations above as well as some screen shots from the cluster administrator in order to follow the way the cluster is set up.
I will close the question tomorrow. I might open new other questions.
thanks
I haven't closed the question otherwise I will forget to print it out. I have been busy, but I will do all my best to print your explanations above as well as some screen shots from the cluster administrator in order to follow the way the cluster is set up.
I will close the question tomorrow. I might open new other questions.
thanks
Join our community and discover your potential
Experts Exchange is the only place where you can interact directly with leading experts in the technology field. Become a member today and access the collective knowledge of thousands of technology experts.
*This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.