Hi @CamDavidsonPilon ,
Hope all is well 
We’re updating our cluster from 25.2.20 to the latest. I’ve scp-ed the releast_xxx.zip to the root and started using pio update --source release_xxx.zip. I have tried updating incrementally, and encountered some issues, so then tried skipping problematic installations if the subsequent installation(s) allowed (i.e. the minimum version was <= my current version). I still encountered some issues and so tried to rollback to what I started with. To rollback, I’ve been doing this:
sudo pip uninstall pioreactor
sudo pip install pioreactor==xxx --no-cache-dir
But it occurred to me that I’m not reversing the commands in update.sh for each update and so have ended up in rather a muddle. Here are two examples of the issues I’m encountering when running pio version -v back on my original version:
Even if we solve these particular issues, I’m worried about the time it will take to solve issues with each update across our 26 pioreactors, and that we won’t have time to discover the non-superficial issues. I understand this is why it’s good to keep up with the versions! Do you have any recommendations? One option I’m considering is re-flashing the leader and workers with the most recent version and re-loading all the data into the database and the cache. I’ve backed up our database to the cloud and I’ve taken a copy of the whole filesystem locally. Any other suggestions most welcome
Many thanks,
Vicky