.silence.
Vault Hunter
I heard we were having more or the same issues with lag on our FTB server (it's currently down). As far as I understand it, eventually the server decides to use an exorbitant amount of resources thus creating unplayable lag. Also from what I can tell, we don't know what is causing that pull on resources (through no fault of any of our staff.) I get that you are looking for answers, but in the meantime, we deal with server lag or complete lack of a MC server. As a donator, this is frustrating. Really frustrating. And I know you're frustrated too.
A proven fix for this situation is scheduled restarts for the server, via use of a script. I've done some searching and that seems to be a very common practice amongst server operators, to prevent the java process from being a COMPLETE drain on box resources. Does this find what is using all of those resources? No. Does this prevent the server from getting to that point, where the it's trying to use 1200% of the CPU? Yes. Look at it this way: it seems like the lag happens within a few days or a week of a restart (from updates and the like.) If we restart once a day when no one is on, and never get that point again, we have eliminated the problem.
I would much rather wait through a 15-30 second reboot, than deal with large outages and ridiculous lag.
Would something like a reboot every 24 hours kill us? An automated 5am reboot seems like a no-brainer to me, rather than beating our heads against the desk trying to find the issue, and in the meantime deprive our generous supporters of what can be a very enjoyable FTB server. You know, when it's up.
A proven fix for this situation is scheduled restarts for the server, via use of a script. I've done some searching and that seems to be a very common practice amongst server operators, to prevent the java process from being a COMPLETE drain on box resources. Does this find what is using all of those resources? No. Does this prevent the server from getting to that point, where the it's trying to use 1200% of the CPU? Yes. Look at it this way: it seems like the lag happens within a few days or a week of a restart (from updates and the like.) If we restart once a day when no one is on, and never get that point again, we have eliminated the problem.
I would much rather wait through a 15-30 second reboot, than deal with large outages and ridiculous lag.
Would something like a reboot every 24 hours kill us? An automated 5am reboot seems like a no-brainer to me, rather than beating our heads against the desk trying to find the issue, and in the meantime deprive our generous supporters of what can be a very enjoyable FTB server. You know, when it's up.