https://bugs.mojang.com/browse/MC-100524
Log files were previously overwritten when more than 7 were created on the same day. This is caused by Log4J's default behavior with DefaultRolloverStrategy, which defaults to a max of 7.
While a max of 1000 doesn't fully stop this problem from happening, for 1000 log files in a single day to be reached the server would have to restart faster than once every 1.5 minutes, which is unlikely to happen. So 1000 seems like a good limit. A higher max isn't used because when it gets higher, there are performance hits due to the way Log4J checks for the next file.
Something the log4j ConsoleAppender does makes the console work correctly
on Windows. After trying to pull pieces of it out and run them manually
I decided to just put the appender back. We now once again start with the
ConsoleAppender then remove it immediately after starting.
When Minecraft first introduced an auto-save feature, we
were taken by surprise by how much of an impact it actually had on the performance
of the server. After investigating the potential causes of the significant
slow-downs we saw at the time, we came to the conclusion that it was a
combination of the auto-save interval being incredibly frequent and
servers already having an auto-save solution that was conflicting with the
newly added built-in one.
Since we noticed that most servers already had their own auto-save
solution, we decided to completely disable the built in auto-save by
default. In hindsight, however, we were so happy that we discovered and
squashed the cause of the performance issues that we forgot to consider
the future and, as a result, some servers have unfortunately been caught
by surprise when they ran their servers without any auto-save plugins.
Without the auto-save plugin conflict, however, Minecraft's default save
interval of 45 seconds is not suitable for the types of servers that run
Bukkit, to the point where it was negatively impacting performance. As
such, we've decided to re-enable the built in auto-save at an interval of
5 minutes for newly created servers.
This adds two settings to bukkit.yml, allowing activation and control of
two chunk garbage collection triggering conditions:
chunk-gc/period-in-ticks controls a periodic GC, run once every N ticks
(default is 600); chunk-gc/load-threshold causes the GC to run once
after every N calls to loadChunk() on a given world (this call is an API
call used by plugins, and is distinct from the path taken for routine
player movement-based loading). In both cases, setting to zero will
disable the given GC scheduling strategy.
In either case, the act of doing the GC is simply one of scanning the
loaded chunks, seeing which are NOT being used by one or more players
(due to view-distance) and which are not already queued for unload, and
queueing them for a normal unload. Ultimately, the unload is then
processed the same as if the chunk were unloaded due to leaving the
view-distance range of all players, so the impact on plugins should be
no different (and strategies such as handling the ChunkUnloadEvent in
order to prevent unload will still work).
The initial interval for the periodic GC is randomized on a per-world
basis, in order to avoid all world being GCed at the same time -
minimizing potential lag spikes.
The new setting is located at "ticks-per.autosave". By changing this
value, it affects how often a full save is automatically executed,
measured in ticks.
This value is defaulting to 0 (off) because we believe that the vast
majority of servers already have a third-party solution to automatically
saving the server at set intervals. Having the built in auto-save disabled
by default ensures that we are not saving things twice; doing so leads to
absolutely no benefits, but results in detrimental and noticeable
unnecessary performance decrease.
For servers that do not use an automated external script to perform saves,
this setting can be turned on by setting the value higher than 0, with 900
being the value used in vanilla.
useExactLoginLocation() looks for settings.use-exact-login-location within bukkit.yml.
If true, we will bypass Vanilla's behaviour of checking for collisions and moving the player if needed when they login.
If false, we will continue to follow Vanilla's behaviour and move players that 'collide' with objects when they login.
