datacenter.QueryConnectionInfo error when connecting ESX host

I had a bit of a problem in my home lab. I have 2 hosts, and one crashed. Unfortunately this was the host running AD & vCenter. When I brought the host back, I managed to restart the VMs, but looking in the vSphere client when I connected back onto vCenter, both hosts were disconnected. When I tried to reconnect, I got this Datacenter.QueryConnectionInfo error message. I have a feeling that name resolution was to blame here, but couldn't work out what was wrong. I could ping the hosts from vCenter by IP & name, I could add the hosts by IP address, but not by name. The only work around I could come up with was to add the ESX host details into the hosts file to avoid using DNS. As soon as I did that, I could reconnect the hosts. I've no idea what the actual problem is, although i wonder if some connection information is stored in the database and was corrupted when the host died.

Useful vMA commands

NFS Storage

If you are connecting an NFS share to your ESX(i) server, use the command vicfg-nas

To see what NFS datastores are connected to a host, use

• vicfg-nas -h esxhostname/ip address -l
• e.g. vicfg-nas -h esx.test.com -l
  

To add a datastore to a host

• vicfg-nas -h ESXhostname/ip address -a "Datastore name" -o NFS Servername/ip address -s sharename
• vicfg-nas -h esx.test.com -a "NFS datastore" -o nfs.test.com -s /mnt/nfs
  

To delete a datastore from a host

• vicfg-nas -h ESXhostname/ip address -d "Datastore name"
• e.g. vicfg-nas -h esx.test.com -d "NFS datastore"

Virtual Machine vSwitch Configuration

To list vSwitches and portgroups use:

• vicfg-vswitch -h ESXhostname/ip address -l
• e.g. vicfg-vswitch -h esx.test.com -l

To create a vSwitch

• vicfg-vswitch -h ESXhostname/ip address -a vSwitchname
  
• e.g. vicfg-vswitch -h esx.test.com -a esxvSwitch
  

To delete a vSwitch

• vicfg-vswitch -h ESXhostname/ip address -d vSwitchname
  
• e.g. vicfg-vswitch -h esx.test.com -d esxvSwitch

To create a portgroup in a vSwitch

• vicfg-vswitch -h ESXhostname/ip address -A portgroupname vSwitchname
• e.g. vicfg-vswitch -h esx.test.com -A portgroup1 esxvSwitch

To configure a portgroup with a VLAN ID (this can't be done when the portgroup is created

• vicfg-vswitch -h ESXhostname/ip address -v vlanid -p portgroupname vSwitchname
• e.g. vicfg-vswitch -h esx.test.com -v 32 -p portgroup1 esxvSwitch

To delete a portgroup from a vSwitch

• vicfg-vswitch -h ESXhostname/ip address -D portgroupname
• e.g. vicfg-vswitch -h esx.test.com -D portgroup1

vmkernel vSwitch Configuration

The format is very similar to the vicfg-vswitch command
To list vmkernel portgroup details use:

• vicfg-vmknic -h ESXhostname/ip address -l
• e.g. vicfg-vmknic -h esx.test.com -l

To create a vmkernel portgroup (i.e. convert the portgroup from a Virtual Machine portgroup to a vmkernel portgroup)

• vicfg-vmknic -h ESXhostname/ip address -a -i ipaddress -n netmask -p portgroupname
  
• e.g. vicfg-vmknic -h esx.test.com -a -i 10.10.10.1 -n 255.255.255.0 -p portgroup1
  

To delete a vmkernel portgroup (i.e. convert it back to a Virtual Machine portgroup)

• vicfg-vmknic -h ESXhostname/ip address -d -p portgroupname
  
• e.g. vicfg-vmknic -h esx.test.com -d -p portgroup1

To enable a vmkernel portgroup for VMotion

• vicfg-vmknic -h ESXhostname/ip address -E -p portgroupname
  
• e.g. vicfg-vmknic -h esx.test.com -E -p portgroup1

VCD Reservation Pool Capacity

I’ve been playing with some capacity stuff in the cloud, and have used PowerCLI to gather some allocation information for each of the Org vDCs that exist (such as reservation allocated to each vDC & reservations used by the VMs in it).

This has gone swimmingly for PAYG & Allocation Pools, but for Reservation Pools I made the assumption that no VMs in the vDC would have any reservations. I thought I should get some data to validate my script, so created a couple of random vApps in a test Reservation Pool vDC. I ran my script and found that actually reservations existed for some of the VMs.

The reason is that for Reservation Pools only, the resource settings come directly from the template VM, so if the template was created on an Allocation Pool vDC, those settings will be transferred when this template is cloned.

I believe that this is a bug rather than a feature.