"Help I tried to `trellis up` an older project and it's not working!" and other stories

“All I did was run trellis up what happened??”

A couple of recent upstream changes, to macOS Monterey’s nfs settings and GitHub’s public keys specifically, have broken older Trellis instances, causing previously working Trellis environments to break seemingly on their own. While we’ve corrected these issues in the latest versions of Trellis, trying to figure out which changes to back port into your specific project without breaking something else can be a challenge. In this post I hope to help you fix a few of the more common issues, and consolidate a known-good fixes for breaking Trellis changes.

This is not a thread to report new issues with Trellis.

I upgraded to macOS Monterey and now nfs won’t mount!

A change in how macOS Monterey handles IP addresses locally broke some older Trellis projects. If your Trellis project hangs at:


==> default: Mounting NFS shared folders...

And eventually times out:


mount.nfs: Connection timed out

Here’s the fix:

Open trellis/vagrant.default.yml find the following around line 2:

vagrant_ip: '192.168.50.5'

Replace with:

vagrant_ip: '192.168.56.5'

Open trellis/hosts/development find the following around line 38:

192.168.50.5 ansible_connection=local

Replace with:

192.168.56.5 ansible_connection=local

Also find the following around line 41:

192.168.50.5 ansible_connection=local

Replace with:

192.168.56.5 ansible_connection=local

Destroy your Trellis VM and re-create it:

# From your trellis directory
vagrant destroy && trellis up
7 Likes

Now it says the key for GitHub has changed?

GitHub recently changed their public key. If you see something like the following when you’re trying to install Composer dependencies:


@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ 
@ WARNING: POSSIBLE DNS SPOOFING DETECTED! @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ 
The ED25519 host key for github.com has changed, and the key for the corresponding IP address 140.82.114.3 is unknown. This could either mean that DNS SPOOFING is happening or the IP address for the host and its host key have changed at the same time.

Here’s the fix:

Open trellis/group_vars/all/known_hosts.yml
Replace the contents of this file with the latest from the Trellis repository:

Destroy your Trellis VM and re-create it:

 # From your trellis directory 
vagrant destroy && trellis up
6 Likes

I just used this thread, so I’m calling this a smashing success.

4 Likes

Yes that helped me! Thank you! :heart: