Yeah, something happen to the server last night right after sawine got off lol.
there is a tradeoff in how we run the server.
If we ran the server and never saved then we could have a smooth running server under this setup with and absence of lag. A ramdrive, 16gigs of ram, and the high end processor should run the entire thing lag free if we never had to worry about saving.
because the ramdrive is truly one of the ideal aspects of the setup we are in greatest need for saving. having the world running on ram means that if at any point the server turns off or requires a restart the entire world would be wiped. Also any failures in the ramdrive means that the map might think it needs to generate space that already exists such as what happened on the 3-17 rollback.
The best way to avoid either of these issues is to save more often. this means that every time it saves or copies a save the server divides its attention between running the server program and doing the proper saving and copying process. This means that while we lower our chances of rollback we do increase the chances of small single issues like block lag and location lag.
It's finding a balance in between the two that really matters. We have to let the server run on it's own enough of the time to make gameplay smooth, but at the same time we have to save in order to avoid rollback, where rollback on an unsaved ramdrive means that it goes all the way back to square one.
If we ran the server and never saved then we could have a smooth running server under this setup with and absence of lag. A ramdrive, 16gigs of ram, and the high end processor should run the entire thing lag free if we never had to worry about saving.
because the ramdrive is truly one of the ideal aspects of the setup we are in greatest need for saving. having the world running on ram means that if at any point the server turns off or requires a restart the entire world would be wiped. Also any failures in the ramdrive means that the map might think it needs to generate space that already exists such as what happened on the 3-17 rollback.
The best way to avoid either of these issues is to save more often. this means that every time it saves or copies a save the server divides its attention between running the server program and doing the proper saving and copying process. This means that while we lower our chances of rollback we do increase the chances of small single issues like block lag and location lag.
It's finding a balance in between the two that really matters. We have to let the server run on it's own enough of the time to make gameplay smooth, but at the same time we have to save in order to avoid rollback, where rollback on an unsaved ramdrive means that it goes all the way back to square one.
it could be due to plugins
mob tamer and rider
mob tamer and rider
Timeouts are when the server get's stuck on a process that is often completable but takes a large amount of time. because of how things are prioritized the server focuses on that task and then forgets to respond to the players and therefore your clients never log in.
Those are the donations for 2 months, i forgot to reset it last month.
And Im looking into the issue, ofc im not happy the server crashing, Im trying to keep the quality standards high and to give the best server possible.(considering its p2p)
But even if we had a 3-4x more powerfull server. It woulnd change much because the server is not crashing because it misses power. This box is able to handle 50-100 players. There is probally a plugin that is not stable and compatible and makes the server processes work forever on it, probally crash. I would also need to set back those automatic restarts...
Anyway like I said, im reviewing the logs atm and looking into it.
And Im looking into the issue, ofc im not happy the server crashing, Im trying to keep the quality standards high and to give the best server possible.(considering its p2p)
But even if we had a 3-4x more powerfull server. It woulnd change much because the server is not crashing because it misses power. This box is able to handle 50-100 players. There is probally a plugin that is not stable and compatible and makes the server processes work forever on it, probally crash. I would also need to set back those automatic restarts...
Anyway like I said, im reviewing the logs atm and looking into it.
thanks sawine <object class="emojione" data="https://resources.enjin.com/1489581540/themes/core/images/emojione/svg/1f642.svg?0" type="image/svg+xml" standby=":)">:)</object>
between cryston and sawine's explanation i understand alot more thank you guys now im not as mad as i was earlier about block respawn
I hate when people bitch about stuff they have absolutely no clue about. locked
sawine and cryston unlock if you wanna continue explaining stuff.
sawine and cryston unlock if you wanna continue explaining stuff.