Is this thing still on?

[wholmgren]

It doesn't seem so. Anyone aware of a reason for failure?

[kandersolar]

It was running on a virtual machine at UA. I think I may have lost my login credentials for it, but I'd be surprised if the VM was still online anyway given that the people who set it up aren't there anymore.

Maybe time to migrate to GH Actions? I remember seeing some blog posts about overcoming challenges in using CI for benchmarks.

[wholmgren]

Whoever could you be referring to???

I found my credentials to log into the VM and actually the machine is still running. The script stopped working with pvlib started requiring python >= 3.9. The cronjob installs pvlib into an environment that was using python 3.7. I've updated it to python 3.11 (3.12 doesn't work because of the solarfactors limitation and the cronjob's spec for pip install .[all] - a few solutions there of course but I chose path of least resistance.

It seems the ssh key for the pvlib-benchmarker account may be expired due to inactivity. I cannot find credentials to log into that github account. Do you have them @kandersolar? Perhaps this was an @alorenzo175 job? Maybe @lboeman is ready to chase us off his property and this is irrelevant?

[lboeman]

Maybe @lboeman is ready to chase us off his property and this is irrelevant?

I'm happy to keep this VM running if it's useful as our group is still heavily dependent on PVLib, so I think running that VM here is the least we could do. I would like to do more to actively contribute to PVLib, when more of my time is freed up but I don't know when that will be. Let me know if there's anything needed from me, and if this does go the way of GH actions I can take that VM down.

[kandersolar]

Nice sleuthing! I do indeed have the GH credentials for @pvlib-benchmarker. Let's get those shared so that the bus factor is a little better.

I'm inclined to keep using the VM as long as it's available!