Unfortunately you are not the first to report a problem since the “thread is dead” issue was reported as fixed in. @Dave reported an issue in the EmonPi Crash thread but there doesn’t appear to have been any updates there.
If it happens again can you attach your emonhub.log file to this thread and also there should be a log file at… well I give in! I’ve spent some time looking for some reference to it but I had to give up, I recall it being talked about and I can see that @Dave has attached one to his post but I failed to find where that might be.
To be fair this was never a “fix”! the route cause has never been located or tackled, the “fix” was to stop one thread bringing down another thread when it crashed, add a traceback to help debug crashed threads and to add an automated thread restart (if you look at @Dave’s log it restarted many times in the duration of that one log), the intended result of the first part would be that only the faulty thread would crash, the second part would report info on why it crashed so it could be fixed and then the thread would get restarted, However the 2nd part doesn’t appear to work looking at Dave’s log, so when the third part works it simply “papers over the cracks” until a major stoppage occurs and then without the second part we are none the wiser.
It’s quite interesting that you say the emonhub service had stopped as that is a new symptom, emonhub itself has never crashed out when the interfacer threads have died previously, that combined with the fact you had to reboot (something else @Dave probably had to do as his feedwriter service had stopped/crashed) suggests this issue has evolved to affect other services rather than having been fixed.
If you post the logs and as much info as you can it would help debugging. (perhaps @Dave can tell us where he found the log file)