azonenberg changed the topic of #scopehal to: libscopehal, libscopeprotocols, and glscopeclient development and testing | https://github.com/glscopeclient/scopehal-apps | Logs: https://libera.irclog.whitequark.org/scopehal
Darius has quit [Quit: Bye]
Darius has joined #scopehal
Degi_ has joined #scopehal
Degi has quit [Ping timeout: 255 seconds]
Degi_ is now known as Degi
bvernoux has joined #scopehal
juri_ has quit [Ping timeout: 240 seconds]
juri_ has joined #scopehal
nelgau has joined #scopehal
bvernoux has quit [Read error: Connection reset by peer]
<d1b2> <johnsel> sorry for the suspense I fell asleep
<d1b2> <johnsel> I found out that there are changes on the XCP-ng side that are causing these issues
<d1b2> <johnsel> XOA has been patched to solve this issue. The terraform scripts have not
<d1b2> <johnsel> It is still not clear where specifically things are going wrong, but I do have a workaround that unfortunately does depend on manually creating templates that aren't dynamically configured after deployment but at least gives us a way to continue work while I post the issue on the terraform github and let it percolate back up the pipeline
<d1b2> <johnsel> it looks like something in the API changed related to how the cloud-config scripts are sent to the VM
<d1b2> <johnsel> I guessed that already, but I confirmed it
<d1b2> <johnsel> What I found was a check to see if the header of the cloud-config was correct, I had hoped that that + multiline with an <<EOF notation would bec ausing it but it wasn't
<d1b2> <johnsel> it's something else
<azonenberg> Ok so tl;dr upstream bug in terraform caused by changes in xcp-ng?
<azonenberg> And you have a workaround for the near term?
<d1b2> <johnsel> well, upstream changes in XCP-ng and/or XOA caused a bug or at least caused the API calls to no longer work as they did before
<d1b2> <johnsel> and then that is the the terraform plugin for XOA
<d1b2> <johnsel> terraform itself is fine
<d1b2> <johnsel> but yes other than those details, correct
<d1b2> <johnsel> I will just set up templates manually with the github runner installation script either finished or ran on startup
<d1b2> <johnsel> then we do most of the package management through the runner already anyway
<d1b2> <johnsel> for windows same deal
<d1b2> <johnsel> that does mean we won't have infra-as-code which is a bummer but that is what it is
<d1b2> <johnsel> we can go back when they figure out how to make it work again
<azonenberg> Yeah getting CI working at all is priority IMO
<azonenberg> then we can fine tune
sgstair_ has joined #scopehal
sgstair has quit [Ping timeout: 240 seconds]
sgstair has joined #scopehal
sgstair_ has quit [Read error: Connection reset by peer]
<d1b2> <johnsel> azonenberg: ping discord
<d1b2> <johnsel> vpn cert needs renewal
<d1b2> <johnsel> I may have found a fix for the actual problem
<d1b2> <johnsel> or at least some way to keep using existing scripts
sgstair has quit [Read error: Connection reset by peer]
sgstair has joined #scopehal
sgstair has quit [Read error: Connection reset by peer]
sgstair_ has joined #scopehal
sgstair has joined #scopehal
sgstair_ has quit [Read error: Connection reset by peer]
sgstair has quit [Read error: Connection reset by peer]
sgstair_ has joined #scopehal
sgstair_ has quit [Ping timeout: 258 seconds]
sgstair has joined #scopehal
nelgau has quit [Ping timeout: 255 seconds]
nelgau has joined #scopehal