StoneLund's Posts

Hi Really old thread but i have the exact same problem. We have replaced one of our domain controller, and we moved the roles temporary. During the process of replacing the domain controller, the vcenter server lost contact with a couple of the hosts. When i try to reconnect them i get the "Authenticity of the host´s SSL certificate is not verified". I can can remove and rejoin the host to the vcenter, but since we are using dvSwitch its not prefferable. (need to manually add the host to the vcenter dvswitch). Is there some solution to this problem? The hosts are running fine, but vcenter cannot connect.

HI Ok.. But, in my case, with a host in another location, only one vCenter and vSphere Replication enabled. What could be a appropriate soultion for automated failover with vSphere Replication? Since the UPS has around 1h battery time, failover to loss of power would be unlikely, and could be done manually if needed. However, at offours and holidays etc. it would be nice with an automated soultion.

Hi, that is correct and what i´m trying to do. But, thats a shame that there is now way to script this. It would be very helpful. I should work though to shutdown the machine in case of powerfailure by PowerCLI. Rename the machine vmx, nvram and so on files on the destination datastore with PowerCLI so the become usable. Then register and PowerON the machine with PowerCLI on the destination host. I found a blog post on the net with the commands... Next question is though, will vSphere Replication contunue to replicate the powered off machine. Or does it stop automaticly when the machine is powered off?

Hi Yea, we use Eaton UPS and they also have integration with vCenter. Though, only to shutdown host etc and to trigger scripts. I would like to let the Eaton UPS trigger the script to force a replication of the VM and to restart it on the new host while vCenter is running.

Hi I could succefully uninstall the SSO and everything else (keeping the database) and reinstall. I did however have to add the datasource manually since the vCenter installtion removed that. Also, the problem with the "Unable to connect to one or more vCenter server systemes when logging in to the web client" error was not only the missing pointer, but the name i had on my vCenter server. I had _ (underscore) in the name. When i renamed the server before reinstalling, the Web Client found the vCenter server without any problem. So, lessons learned. You can uninstall vCenter safely, as long as you keep the database. (backup recommended anyway). And, dont have underscore in the server name.

This is some error logs i got from vcenter. Typing $error[0] $error[0].Exception in the powercli did not return any data. Maybe there is some variable that sets the timeout to 50000ms somewhere?? 2012-04-26T07:58:39.617+02:00 [04652 verbose 'Default'] Timed out reading between HTTP requests. : Read timeout after approximately 50000ms. Closing stream SSL(TCPStreamWin32(socket=TCP(fd=4328) local=127.0.0.1:8089,  peer=127.0.0.1:50251)) 2012-04-26T07:59:42.035+02:00 [05552 warning 'Default' opID=f58b1661] [VpxdMoVAppPowerOp] Got exception during vApp start/stop: vim.fault.Timedout 2012-04-26T07:59:42.035+02:00 [03712 error 'Default' opID=f58b1661-4c] [VpxLRO] Unexpected Exception: vim.fault.Timedout 2012-04-26T07:59:42.035+02:00 [03712 info 'Default' opID=f58b1661-4c] [VpxLRO] -- FINISH task-1664 -- vm-146 -- VirtualMachine.powerOff -- 2012-04-26T07:59:42.035+02:00 [03712 info 'Default' opID=f58b1661-4c] [VpxLRO] -- ERROR task-1664 -- vm-146 -- VirtualMachine.powerOff: vim.fault.Timedout: --> Result: --> (vim.fault.Timedout) { -->    dynamicType = <unset>, -->    faultCause = (vmodl.MethodFault) null, -->    msg = "", --> } --> Args: --> 2012-04-26T07:59:42.035+02:00 [05552 error 'Default' opID=f58b1661] (Log recursion level 2) vim.fault.Timedout

I get the time out error after like 30 seconds or something after it has started the shutdown of the first group of machines in the VAPP. I will check more logfiles tomorrow! Sent from Lotus Traveler LucD --- New message: "Stop-Vapp returns operation timed out." --- Från: "LucD" <communities-emailer@vmware.com> Till andreas.stenlund@boliden.com Datum: ons, 25 apr 2012 18:13 Ämne New message: "Stop-Vapp returns operation timed out."

Hi Okey, so i found the VAPP tab. I have not used that before. Basicly, would I be able to create a new vapp in the cluster, add all the virtual machines that require a specific order to that VAPP. Set the start order to that wich i require. (This machine before this one, etc. etc.). And then, start upp the VAPP and the cluster will respect that order when it starts up after a power failure? So, the only question left if that would work is how to call the scripts automaticly, the one for the vcenter. And, the one for the vapp?

Hi Did you managed to solve this? I have quite the same problem. I enter the credentials to the vsphere client, though i get the error on the vmware connection. In the vsphere client, i get this strange error saying: User cannot logon since the user is already logged on. I have created a new user in our active directory with the administrator role for the eaton connection,

Hi Thank you for the reply. I found a command reference to the start-vapp, but i could not find the startup delay option. Next question, How do I call the script after the virtualcenter machine has started, and how do I get the virtualcenter server to start first, if it has been gracefully shutdown. If all the virtual machines get gracefully shutdown by the cluster on power failure, my understanding is that they will not startup since the ha cluster thinks that they are supposed to be shut down.