A change in the invocation of antnode caused upgraded nodes to not start correctly (now needs --no-upnp flag even for custom ports defined nodes).
Halting the upgrade and rolling back the failing nodes was a temporary fix.
Due to time limitations, my next step is likely to reset my Original nodes and restart with the current anm until after IF. The alternative is to patch the node services files and then let the normal upgrade process restart them. This could cause broken nodes as --no-upnp is an invalid flag for 0.3.9.
I had considered a canary process as a future feature when there is a persistent process manager desirable, to monitor spike and settled traffic and logs before upgrading more nodes instead of blindly going forward. xdconsole for antnode/wnm?