In case anyone wonders how this played out. At the end of the day, I shutdown the host with the running workloads on it, kept it at BIOS boot so the 8.3 installer would upgrade it cleanly, and then booted it back up. For reasons related to my particular setup, the down time was more than 60 minutes. But that's not related to XCP-ng. The XCP-ng installer was maybe 10 of that. 20 of that was 3 reboots on a server that takes ages to boot. One thing that is a bit difficult is that the 8.3 installer doesn't notice ZFS volumes or help with ZFS. So after the upgrade, I have to follow a modified set of these instructions (modified because the package names have changed over time) to install ZFS and its components into the system. So the sequence is: Shutdown Install via ISO Reboot to single user mode Run yum update to pull in all the updates Run the ZFS commands to install ZFS on the host Reboot again Run normally Of course, I ended up rebooting more than just those 2 times, because I forgot that I would need to install ZFS (even though I did it on 2 different hosts already ) So now the pool is fully upgraded, everybody is happy. I'm now working on creating shared storage across the nodes. It's amateur hour over here. I appreciate all the help.

@olivierlambert @paco Thanks! Will work on it

@paco said in Storage IOPS and throughput dropouts?: @nikade Short answer is that the spikes seem to have smoothed out. I'm not really sure what happened. I don't have them at any resolution now. Perhaps some kind of heisenbug. Happy to hear that

That's correct. There's no more updates, that's why you MUST upgrade to 8.2 (see https://xcp-ng.org/docs/releases.html#all-releases )

Doc now updated with 2 pictures here: https://xen-orchestra.com/docs/proxy.html#architecture

@stormi It does. I put the screen shots of what it looks like when you run it in the pull request.

@Danp As soon as I read that, I thought "I think I've been down this road before." Sure enough. I gave it 2G of RAM and it wouldn't build. I gave it 4G of RAM and I watched using systat and that final step went to about 52% of RAM usage. It has completed. I routinely run Xen Orchestra on a 1G VM because it's rarely used and it runs fine that way. But building it clearly takes a lot more RAM. I'll update my internal notes. It might be worth it to update the building from source docs to mention having enough RAM. Thanks for hitting the nail on the head with that advice!

Glad I was able to help.