Automatic Container Updates
We regularly release new versions of Container images when new security patches or bug fixes are released. Automatic updates make sure that your Containers are always running the latest stable version. All you have to do is set a weekly maintenance window for each server, and we take care of the rest.
Every time your server’s weekly maintenance window opens we check whether any new images are available. If so, we apply the update and restart the Container (which involves a few seconds of downtime).
Setting Your Servers’ Maintenance Window
If you don’t have Business or Premium Support and Management, the first step is to upgrade. This is a Team setting within the Console.
Once your new Support and Management plan is applied, it only takes a minute or two to set a maintenance window for each of your servers. Open your server’s Settings tab to find the controls.
It’s best to pick a day and time that’s typically quiet for the websites on the server. Image updates require a reboot of the Container, which means a small amount of downtime.
Opting Individual Containers In or Out
By default, your maintenance window will apply to all the Containers on your server. If you want to retain manual control over some image updates, there’s a switch for that. You’ll find it in each Container’s settings.
You can always opt the Container back in again later.
Config Changes, System Checks and Technical Info
Configuration Changes Are Retained
If you’ve changed your Container configuration by editing files in the container/config
directory, that’s no problem. The Webslice Container platform is designed to put you in control.
During automatic container upgrades, we’ll merge your changes with the new configurations. If for some reason that doesn’t work, then we’ll revert the update and leave your Container as it was.
The most important thing is that automatic Container updates will never change your custom configuration.
Basic Built-In Health Check
After updating your container, we run a simple check by sending a request to the homepage of the Container’s main domain. A successful result of 200 OK
tells us that everything is good to go.
Any other result is taken as a sign to revert the changes and leave your Container as it was before maintenance began.
Optional Health Check
If the built-in health check isn’t thorough enough for your Container, that’s no problem. You can build out a more robust check easily enough.
All you need to do is implement a custom JSON endpoint at /.sitehost/health
which returns the object {“status”: “healthy”}
as the response. Any other response will be considered unhealthy.
If the custom JSON endpoint returns 404 or redirects to another page, we will fallback to the homepage.
Redirects, Aliases & SSL
The health check is designed to honor HTTPS requests by setting the appropriate headers.
It’s also built to follow 301 and 302 redirects as long as the redirect target is a virtual host on the Container. Multiple virtual hosts are not a problem so long as users are redirected to your primary domain.
Redirects outside of your Container will fail the health check.