When trying to open a console to a VM, I encountered the message;
Unable to connect to the MKS
I tried forcing a reset but this got stuck at 95% and I wasn't allowed to force a power off through the GUI as the reset process was still ongoing.
To get round this I followed the steps below;
1. SSH onto the host running the VM.
2. Run the command;
ps | grep "VMName"
The name is case sensitive so make sure you type it exactly as shown in vSphere. You will get a list of information as below;
2106831 2097959 vmx-mks:VDP01 /bin/vmx
The second number (in this case 2097959) is the process ID (PID).
3. Kill the process by running the command;
kill -9 PID
4. cd to /vmfs/volumes/DataStoreName/VMName
5. Delete the swap file by using the command below;
rm -r swapfilename.vswp
6. In vSphere Client you should see the machine has powered off. Right click it and remove it from the inventory.
7. Browse the datastore, open the VM folder, right click the .vmx file and choose "Add to Inventory".
8. Power on the VM. You may be prompted to answer a question about whether you have moved or copied the VM so if the power on is stuck at 95%, check for a hidden popup.
On trying to backup a newly P2V converted Server 2003 machine, I found that VDP would allow me to create a backup job and initialise it, but wouldn't actually carry out any tasks and would flag up immediately as a failure.
To resolve this, I had to power off the VDP appliance, edit the settings and double the amount of memory VMware had published the machine at (increased memory from 4GB to 8GB).
Once this had been done, I powered the appliance back on and the new machine backed up successfully.
After restoring a VM from backup the VM would not bring up it's network connection, despite all the correct information being in /etc/network/interfaces
Running ifconfig only showed the lo connection, no eth interfaces.
Running ifconfig -a lists all network interfaces, including those that are currently down. There was an eth3 interface, but no eth1 as specified in the /etc/network/interfaces file. I updated this file with the correct interface number and after reboot, the machine was live.
If you have enabled SSH access to a host you will constantly see a warning message informing you that SSH has been enabled. This message can be suppressed by following the steps below;
1. Connect to the Host via vSphere Client
2. Go to the Configuration tab and click on Advanced Settings under the Software section
3. Scroll down and select the UserVars option on the left
4. Find the UserVars.SuppressShellWarning option (should be at the bottom of the list) and set it to 1
After building an ESXi 5.1 host and setting up the networking I found I was unable to vMotion any machines onto the new host once added to vCenter. Whilst I had correctly added a vMotion VMKernel port, on the new host I had created the Virtual Machine Port Groups with the label all in capitals (eg. SERVERS rather than Servers).
In order for vMotion to proceed, the label of the port group must match that on the old ESXi 4 hosts.
By running the command below in the Nexenta Management Console (NMC) you can view live data transfer progress of an auto-sync job to confirm it is running OK.
SSH to the Nexenta box running the auto-sync service you wish to monitor and su to root permissions.
At the NMC level, run the command
show auto-sync ;autosyncname stats -i 1
