Configure Let's Encrypt SSL certificate Failed on Configuring server

[oliverati]

At the end of this task Certificate is renewed, but its not automated for the next renewal.

i tried with this solution but, not luck.

Let's Encrypt renewal never finishing

I’m running Octopus 2020.3.2 on an Azure VM, and I have Let’s Encrypt setup which has been working fine. But now it’s time for renewal and the server is failing to install a new cert. It seems to get one successfully, but then fail to install it. [image] The server seems to be running the Let’s Encrypt renewal twice simultaneously, and I’m wondering if that’s causing the issue? [image] I’ve tried rebooting the VM, but the issue persists. Any ideas how to fix this?

[clare.martin]

Good afternoon @oliverati,

Welcome to the Octopus Forums and sorry to hear you are having issues surrounding Let’s Encrypt.

This particular issue was a bit of a nightmare. There are two GitHub Issues in for this which I will post below:

Issue 1
Follow up Issue

Are these similar to what you are experiencing? Are you able to send us your raw task logs for the Lets Encrypt task please so we can see it and see if we can get to the bottom of it.

I have provided you with a secure link here which you can upload them to.

Also, what version of Octopus are you running please.

Kind Regards,

Clare Martin

[oliverati]

Hi Clare, thanks for the reply, i uploaded the logs for my issue. My certificate is renewing at the end of the task, but there isn’t setup for next renewals as you have posted. The task is finished for 44 seconds. At the end i have 2 more errors: node are draining and network lost to the server.
My version is 2021.3 (12217)

[clare.martin]

Hey @oliverati,

Thanks for the logs and for confirming your server version. I have had a look at the logs and it does look like you are experiencing the second GitHub issue I posted. As you can see from your logs it does actually request the Let’s Encrypt certificate but then fails to restart the server afterwards. The restart is where it actually completes the task and therefore allows the certificate to auto-renew.

Your logs (certificate name taken out):

== Failed: Configuring server ==
 Adding 'https://xxxxxxx' to listen prefixes
Draining connections from this server
Entering maintenance mode to drain all web connections
Waiting for other tasks to complete
Restarting server...
If the restart does not automatically occur within 10 minutes, please manually restart the server and the take the node out of maintenance mode (Configuration -> Nodes)
If the server does not start automatically, please log into the server machine and start the services

Your logs are identical to the ones I posted in that second GitHub Issue (towards the end of the logs where it tries to drain the server). Also, note it’s the same error at the start of the logs…

Object reference not set to an instance of an object.

You also said you see two other errors after trying to renew ‘nodes are draining and network lost to the server.’ Which is exactly what happens when that issue in the GitHub document occurs.

As you can see unfortunately the workaround posted only works for server version 2021.3.8815 so it will not work for your instance.

The only way around this I am afraid is to upgrade to version 2021.3.12459 or later. You can see all the different versions on our Octopus Downloads page.

Once you have upgraded you will then see your Let’s Encrypt certificate will auto-renew.

I know upgrading might be an issue for you at the moment but, unfortunately, it is the only way you will get your Let’s Encrypt certificate to renew.

Let me know how you get on,

Kind Regards,

Clare Martin