XO server loses pool and hosts momentarily, timeout error
-
@felibb said in XO server loses pool and hosts momentarily, timeout error:
upgrade my old XO to bookworm + latest commit in master
Welp, that didn't help much, still seeing timeouts. Also neither XO nor XOA show the VM's own IP in the GUI anymore. dist-upgrade renamed interface from
eth0toetX0, and I had to edit/etc/network/interfacesto get the network back up, and I can connect, but GUI still says "No IP record". Management agent 8.0.50-1 detected, in case it matters.Fresh VM setup to be tested another day.
-
(Replying to my previous post, a bit off-topic for the thread, but having installed https://github.com/xenserver/xe-guest-utilities/releases/tag/v8.4.0 manually, I see the IP in GUI now, but XOA says "Management agent 8.3.60-1 detected")
-
So I booted a pre-bookworm upgrade XO VM I had created earlier, moved its network to
ManagementLAN, and installed 6444f88 (latest as of writing this). No timeouts anymore. Seems that Debian version doesn't matter here, and the code is fine (as we sort of determined already), but my networks are perhaps not routed fast enough for XO? It makes sense to put a mgmt appliance into the mgmt LAN, of course, however older code (that I tested, pre-bfb8d3b) did not have this issue. So maybe a combo of that newundicipiece and slightly higher latency is causing it for me? -
That's a great feedback @felibb , thanks!
We'll try to dig on this to see if that lead takes us to something
