Items to check before updating the Nexthink to a new release

Question:

What to check and how to prepare the Nexthink Appliance for the smooth update and avoid unexpected issues during the update?

Answer:

In order to avoid issues during or after the update of the Nexthink Appliance, we recommend performing a small check.

Steps:

1. Does the root file system have enough free space to perform the update?

Both the Engines and the Portal need space to download the packages and perform the update.
A reasonable amount is at least 10 Gigabyte free space for on-site Appliances. Also, take into account extra space is required for at least one database backup which is performed just before the update.

To get an estimate of what size the Appliance database backup will be you can check the following locations below. Take into account that creating the backup will take roughly twice the amount of space of the backup file size.

Portal:

CODE

/var/nexthink/portal/backup/

Engine:

CODE

/var/nexthink/engine/01/backups/

2. Check if all the Engines and the Portal have access to the following hosts which are required for the update?

updates-v6.nexthink.com

updates-centos-v6.nexthink.com

You can check connectivity with the following commands from the CLI.

CODE

wget -d http://updates-centos-v6.nexthink.com/welcome.html
wget -d https://updates-v6.nexthink.com/welcome.html

Output should look similar to this here below for proper connectivity:

You may also test this from the Web Console of each Appliance.

Under "Cloud services" section, test the connection.

Keep in mind, it is not possible to test access to updates-centos-v6.nexthink.com from there.

3. Check if the Hardware Requirements have changed for the Portal and the Engine

First, compare current hardware and the release you will be updating too.

If anything needs to be increased please do so before updating to prevent any problems.

4. What is the connection status of your collectors?

For the Collectors auto-update via Nexthink to work please check that your endpoints have a recent "Last seen on TCP" status and "Last known connection status" is fully connected.

For the collector update to be automatic the TCP status is important that it's fully connected.

See the following documentation for details:

Querying The Status Of The TCP Connection Of The Collector

5. Check in the Portal WebConsole under “Federated Appliances” if all the Engines are connected and have a green status before updating.

If there are Engine appliances that do not have a good federation status (in “red”) they will not update, resulting in a version mismatch that will require manual intervention.

6. Read the release notes for the version you want to update for new or deprecated features and other important news.