CraftBukkit 1.4.5-R0.2 is now available!

A new CraftBukkit Beta Build (1.4.5-R0.2) that provides Minecraft 1.4.5 compatibility is now available! Please only use this build if you'd like to help us get a new Recommended Build out and don't mind running a preview build to help us test things. The previous beta is considered broken and will be marked as such on dl.bukkit.org. This beta contains several critical changes and as such updating to this beta is highly recommended.

Will plugins break with this build?
Provided the developers of the plugins you are using are keeping up with the development of Bukkit, all your plugins should work fine. Unless we specifically have to break plugins or the plugins you are using are employing inappropriate practises, most plugins should work without needing an update whenever we put out a new release.

What is a Beta build?
A Beta Build is in between a developer build and a Recommended Build. These builds simply work and are promoted much more frequently than a Recommended Build. While we will do some testing before promoting a beta build, we will not be running it through our extensive test process. As such, there are no guarantees that they will not contain minor bugs. If we do find out they are broken, we will mark it as such on DLB and promote a new one. A beta build may contain incomplete API and new features but they should not interfere with running the build in any way.

Notes:
The release of Minecraft 1.3 introduced built in auto-save functionality that is completely new to Minecraft: prior to this release, server admins had to provide their own means to save things on their server at a set interval. As a result, the vast majority of servers already have an auto-saving mechanism of some sort, rendering the newly provided built in auto-save largely redundant. On top of this, it was also immediately noticeable that the built in auto-save was the cause of a significant performance hit to the server, even going so far as to lock up the server each time it ran.

Bearing this in mind, we decided the best course of action was to make the auto-save interval configurable and default to being disabled. By doing so we immediately saw a noticeable performance increase on servers and noticed overwhelmingly positive feedback on our Beta builds. While some might argue that this could leave servers without frequent saving, this change simply reverts behaviour back to what server admins are familiar with from Minecraft 1.2.5.

As mentioned before, we have made the auto-save interval configurable and disableable within the bukkit.yml. Simply set "ticks-per.autosave" to the number of ticks you want between each save or set it to 0 to disable it completely and use a script or plugin to manage it instead.

Known Issues:

• Client displays "Unable to locate sign" debug messages under certain circumstances. This is a Minecraft client change that can't be fixed by Bukkit.
• Monsters, villagers, etc. using portals (added in Minecraft 1.4) has been temporarily disabled.

For more detailed information on what is contained in this update, please see the changelog here.

Download CraftBukkit 1.4.5-R0.2 here

 

and what makes you think that ?

 

Well. The main thread is synchronous by design, as are most games, and run on a single thread with the exception of heavy calculations or latency restrictive methods being pushed to helper threads where possible. Since bukkit itself follows this design pattern it would stand to reason that its a fair assumption that he also has this setup. Since all roads lead to the bukkit api, even pushing methods asynchronously or separately on a new thread would intensify the workload via thread creation alone, so even a custom plugin couldn't stray far from what is essentially a built in restriction. If the api were entirely thread safe we would be gloriously consuming octacore mean machines, but java itself isn't the fastest language in the world, and thread safety slows things down even more so. I assume one would still need a high clock to regulate and synchronise all of the data, and probably end up in a worse situation than we are now. The current restrictions require some clever, strict thinking to overcome obstacles and more often than not yield tidy efficient code rather than lazy bloated methods. What I long for is a client side api. The potentials are just as open as bukkit itself, and could even offer the opportunity to push some work to the player instead, he'll you could even push everything player side and use the server as a post office. Synchronisation isn't as important as say a fps. Anyway. Yeah. 20 worlds. 90 plugins. Please tell me your secret

 

As is mine, I don't like to run my server heavy though. I prefer light weight plugins that do multiple things. I'm running 10 worlds. But cool, nice to see that CraftBukkit Beta is working for more than one of us.

It's called ram. Lol. LOTS of RAM.

EDIT by Moderator: merged posts, please use the edit button instead of double posting.

 

If you want a reply I recommend starting a new Forum Thread at the link. I don't know if it resets spawn points for beds.

 

Would you say that this build is stable enough to run it on a gameserver. Or should i wait until thr next beta or even Recommended Build!

 

im running

1.4.5-R0.2 (Build #02488)

And its ROCK solid !

It havent crashed a single time since i installed it on release a few days ago. Server running 24/7

 

Well until all the plugins will be bug free will take a little I had similar situations in 1.4.2 with Orebfuscator and had to disable the caching for it. Patience my friend

 

I notice that craftbukkit 1.4.5 has been released, but on the "Reccomended Builds" section on the home page, the current reccomended build is still 1.3.2, can I be provided with a download link or can this be explained?

 

This is a Beta build, not a Recommended build. The Beta builds are going to be the first builds promoted after a new MC release, as it allows us to promote a mostly stable build without requiring a feature complete API to go along with it.

 

Cool we will be able to go on servers 1.4.5

very very good

 

We've been running 1.4.5 since shortly before the first beta was out and it's worked well so far. I'd say waiting for the next beta or a RB is pointless, but that's just my opinion. If you had plugins working with 1.3.x then odds are they either already work with 1.4.x or have been updated to work already. We run 36 plugins and multiple worlds and everything works great.

 

I was going through the newer development builds in dl.bukkit.org. Some of them says, "There were no changes since the previous build." Why are there builds with no changes? Are the changes very small/irrelevant to comment? Thanks bukkit and keep up the good work.

 

Offtopic posts removed. Conversations on donations are better served in their own threads in the Bukkit Discussion forum.

 

When will you come up with a Recommended Build? Soon, or in a while? A lot of servers are stuck in 1.3.2 because most of the plugins have not updated properly and there isn't a Recommended Build out yet.

 

Plugin devs should get used to updating to beta builds.

 

It's been quite a while since the last recommended build (More than a month lawl), keep up the great work guys! Can't wait for an RB so the plugin devs will update

 

As I previously stated we run 36 plugins and aside from bukkit build #2502 breaking lockette they all have worked. So your statement about plugins not updating "properly" simple isn't accurate. And besides that just because an RB comes out doesn't mean all of your plugins will be magically updated to, that is up to each plugin developer.

 

Thanks! I thought emerald wasn't generating, but I found out that it only generates in the Extreme Hills and Extreme Hills Edge biomes! (It's hard to stay updated with all these new updates)

 

Is it right that teleporting with the command blocks isn't working??

 

It works. Make sure you're not starting the command with a "/" just use for example...

tp @p -418 116 -1134

 

Works great! Thanks to bukkit team!

 

Please don't ask for support on this thread. If you have a problem, make a post in Bukkit Help. Thanks!

 

Im just wondering, what is slowing you guys down? It's been a long time since a last dev build.

 

Beta build is working fine for me, had 37 players on this weekend at peak, all went smooth.

 

They are probably working something that is taking time to check out before they could let us test. On the other hand, this shows that the latest dev-build is very stable.

 

So is the bukkit project even active anymore? Haven't even seen any dev updates recently, which means either, what we have is good enough for a recommended build, or bukkit has been abandoned in anticipation of the eventual API. I love this game, but my server has been dead for a month, waiting on this. Is there a list of outstanding issues somewhere? What is the criteria for a recommended build?

 

https://github.com/Bukkit/CraftBukkit/commits/master
https://github.com/Bukkit/Bukkit/commits/master

Yup, clearly inactive.



...
what.

 

Very active.

Or there is a bunch of work going in to API additions for Bukkit that you don't see because you haven't looked into the bleeding branches.

Run the latest beta build. The recommended build will be out when the team is happy with the API additions being worked on.

 

Hadn't heard about the API additions. Thank you for the explanation. Although I do think it would be best to get at least one RB out there before adding extra functionality. As soon as Mojang puts out the next pre-release, bukkit will be back at square one again (with people not getting the Halloween update until after Christmas). If the beta is "recommended", then it should be promoted to recommended.

 

We use Betas as a way to say "Here's a build that's mostly stable, but likely won't contain any new API additions you may be expecting to see in a RB". We have always stated RBs should be API complete, but due to community demand we have rushed out the initial RBs on every fresh MC build. This stopped with 1.4.2, as we decided to stick to our original intention, and utilize the Beta build system more.

It gives us a couple benefits - if you run a Beta, you understand it may not have all the new APIs developers want to use, as well as help track down some bugs that can sneak into the previous RB builds. It also allows us to use promote builds more frequently as Beta builds, as the community expects RB's to be bug free and API complete.