vCloud Availability 3.5 appliance DNS issues

Recently I have been deploying a few new vCloud Availability 3.5 environments and I really like the product. The ability to connect multiple vCloud Director sites together and provide a migration path between them as well as in and out of the hosted platform is something IaaS providers have needed for a long time. That being said the one thing I keep seeing with the vCloud Availability 3.5 appliance is that it does not seem to use the DNS settings you specify during deployment.

When you are installing the appliance you are prompted to select a deployment configuration. You have the option of selecting Cloud Replication Management, Cloud Replicator, Cloud Tunnel or the Combined Appliance. I have found that this DNS issue occurs with all configuration types. You will first notice the issue when trying to specify the Lookup Server Address. Instead of prompting you to accept the certificate you will be greeted with the following error as it simply cannot resolve the name. If you change it to the IP address here it will prompt with the cert straight away but that’s not a fix.

To look into this further ssh to the appliance address (if you enabled ssh during the install) otherwise just launch the web console. Login as root and attempt to ping the lookup server address which for most will be your vCenter server (if combined) otherwise it will be your separate Platform Services Controller.

You should see the following error – Temporary failure in name resolution

Then run nslookup with the server name like below. You will notice it returns the 127.0.0.53 address which is not the one you specified during the install.

To resolve this we need to modify the network configuration file. First browse to the correct directory,

cd /etc/systemd/network

Then enter the following to edit the file,

vi 10-eth0.network

Note – If you see that the file is blank then press ESC and type :q and press enter to exit out. Then list the contents of the directory with the ls command. Your file name may have been created as 10-eth0-static.network instead. I have seen this with both entries.

Press i to allow you to insert the address and then type in the DNS settings in the same format as below. You can add multiple DNS servers just use a space in between each one.

To save the changes press ESC then type in :wq and press Enter

Then restart the network service using the following command

systemctl restart systemd-networkd

Now try and ping your PSC again by its hostname and the issue will be resolved.

LINKEDIN

Related Post

LINKEDIN