QUICK HELP - Computers, Cluster Admin

[NickFRP]

Can any one else with Cluster Administration?

I have a cluster group called "LONFS01" and since moving offices it wont update with the new IP address..

I have taken all the resources off line and brought them back on-line, but still no luck

Any ideas?

- a reboot wont solve this!

[Puggit]

Turn the passive node off first before changing the active nodes IP? Just a guess, I don't know much about clusters!

[Captain Volvo]

NickFRP said:

Can any one else with Cluster Administration?

I have a cluster group called "LONFS01" and since moving offices it wont update with the new IP address..

I have taken all the resources off line and brought them back on-line, but still no luck

Any ideas?

- a reboot wont solve this!

Posting this in P&P - can't wait to see the responses... How long before the Gubmint is blamed.....

;-)

[NickFRP]

in the resources, i have lonfs01 ip. in the properties, it has an ip (for example which is the correct IP) 10.111.111.20. but if i ping lonfs01 i get the ip address of 19.20.20.17, the old office ip.

I can ping the correct IP and i it works, but it wont resolve the name to the correct ip.

any ideas?

Edited by NickFRP on Saturday 2nd August 08:53

[zadumbreion]

Is this Windows or Novell?

Is it caching old IP addresses - I guess not if a reboot has been tried. Where are you pinging FROM? A cluster server or a PC or something?

If it's Windows, worst case - can you not just run DNS administration and remove the old records?

[carerra g tea]

somewhere there's a table that is resolving the name to the old ip address. possibly on your router or nameserver. it's either updated manually, in which case you will have to find it and edit it, or it will update eventually, in which case it's a matter of pushing the update or waiting it out.

disabling the fire alarm and then setting a nice medium-sized blaze in the appropriate racks in the server room should solve your problem.

Edited by carerra g tea on Saturday 2nd August 09:17

[Simpo Two]

Computing forum

[zadumbreion]

I presume he posted in P&P because he wants help urgently, and his question is more likely to be spotted here than in the computing forum

[MackemPete]

as a bodge if you only need to connect to you cluster from a few machines you can manually add an entry to the hosts table on each client. But as everyone said old DNS records are more than likely the culprit here. Or the PC your trying to ping from has old records in its DNS cache. Try flushing it (ipconfig /flushdns)

[buggalugs]

NickFRP said:

in the resources, i have lonfs01 ip. in the properties, it has an ip (for example which is the correct IP) 10.111.111.20. but if i ping lonfs01 i get the ip address of 19.20.20.17, the old office ip.

I can ping the correct IP and i it works, but it wont resolve the name to the correct ip.

any ideas?

Edited by NickFRP on Saturday 2nd August 08:53

As said, sounds like you just need to update the DNS records. Find out what the machine you're using is using for it's DNS, and change the records on that DNS server.

[theboss]

zadumbreion said:

Is this Windows or Novell?

Is it caching old IP addresses - I guess not if a reboot has been tried. Where are you pinging FROM? A cluster server or a PC or something?

If it's Windows, worst case - can you not just run DNS administration and remove the old records?

Your cluster group will include an IP address resouce and a Network Name resource, for which a dynamic DNS registration is attempted whenever the network name resource comes online. It sounds to me like this process is failing. I would check that the correct DNS servers have been specified for the public network on each cluster node, then on the primary DNS server remove any existing records for the cluster before taking the resources offline/online. Then check to see if a DNS record has been created. If not check the event logs on the cluster node and the DNS server and troubleshoot (plus post results on here). If the DNS record is created correctly but the problem still occurs then you may have a DNS replication or stale caching issue. Also check that you don't have any DNS servers or default gateways defined for the private cluster network interface on any nodes - this can cause problems.

[The_Jackal]

Which company is this, that runs their systems on an unsupported cluster?
I would bite the bullet and ring the company that installed it.