Minecraft, even in single-player mode, has a server module and a client module. What you saw was a case of the server-side hanging. This can be cause by many things, but in the end most of them come down to infinite loops in processing. I'm using the term "infinite loop" in a very broad way here.
In your particular case, based on your mention of general load on your machine, I'd say this is most likely due to hitting your memory allocation cap. I'm guessing the memory for MC is running out, causing the java garbage collector to run, which normally would be able to clear up some old memory and then continue running the game again (this is usually what's happening when you get a temporary version like you mentioned, with the "can't keep up" message). But if there is too much RAM that is still in active use, it will never recover. I'd think this would cause the program to terminate, but I could be wrong. I haven't run into this issue myself.
I seem to remember seeing this happen on another thread, though I can't remember for sure if it was triggered by something infinite or not (for some reason "infinite cows" seems to be on the tip of my mind? Was too long ago to remember).
If you can do something to increase the RAM given to MC, or to decrease the RAM it is using, I'd suggest trying to do so. If you're running a texture pack or shaders, I would personally try disabling those first. They don't have an effect on gameplay, so if disabling them fixes the issue, you know it's just RAM, and not some mod breaking (probably).
Yep, I think that's the right log. I see it's not just modded TEs, but also vanilla TEs which are failing to load.
My first guess is world corruption, which you already tested for ofc, but there is always the possibility that you were not loading the corrupted *part* of the world in the SP test? Generally, world corruption is not the entire world, but rather certain chunks, so if those chunks were being loaded in multiplayer mode but not SP mode for some reason...
Other than that, I didn't come up with any good ideas from the log (unless you count "maybe the SF4 team will have a better idea when they see this log").
Just for reference, how much RAM is the server allocated? I think when a player dies and respawns, the chunks at their respawn point load before the chunks where they died are released (though I'm not sure, so don't quote me on that). If this is true, and the server was already near max memory capacity, I could see it possibly causing corruption.
In any case, I'd suggest going to their issue tracker that I linked to in my previous message, and posting a new issue for this, with a link to this thread for reference. Even if it does end up being world corruption and you manage to wipe the offending chunks, it would mean that something somehow caused said world corruption, and it might happen again. And I could also be way off track, so getting them to take a look is probably a good idea.
I don't see anything about "skipping block entity" in either of those, so I'm guessing they're not the logs I'm looking for. Generally, the logs you want are the ones found in the "logs" folder of the server and the client with issues. If you just ran a session and had an issue, it will be the log file named "latest". Otherwise it will be one of the others (or if the problem was too many runs ago, it will be gone).
I took a look through the MC code a bit, and that error is (as you might expect) just a generic "Hey, something went wrong" error. The things it can trigger off of are not specific enough to give me any solid clues. (If anyone is interested, it's in TileEntity.create() )
(TileEntity == BlockEntity; somewhere along the line, the most common name switched from BlockEntity to TileEntity, but they're the same thing)
You could also try the skyfactory 4 issue tracker, but I can almost guarantee they will also want that log file with the error message you mentioned in it.
The filename is wrong. Notice the " (1)" at the end (ie. you downloaded it twice). I'm guessing that's all there is to it (ie. in the manifest, it's probably "CoFHCore-1.12.2-4.6.3.27-universal.jar" and not "CoFHCore-1.12.2-4.6.3.27-universal (1).jar").
But I could be wrong. I haven't actually messed with publishing a modpack yet, as mine is still in need of work before its ready.
Even just the mod listings in the log file are useful, and fairly often there is a little message somewhere that is overlooked (though I'm not saying that's necessarily the case for you here). I'd pastebin/hastebin and link it here just in case. I haven't played SF4 yet, but I'll at least look at the log file and see if I notice anything new, if you post it. I can also scrape through forge's code since I'm a mod dev myself and see where that particular error is triggering from in code (assuming the message lives in MC/Forge, and not in one of the mods in the pack).
I don't know which mod it's from, but if you want to find out, I have a library mod (which is normally just a requirement for my other mods) which goes through the registries and saves reference text files of all items, entities (mobs), potions, etc. You can use it to debug and then uninstall it afterwards if you like.
The text files will appear in config/fantasticlib/reference, and will be created/updated when the game finishes starting up. Just open entitites.txt and search it for "scientist". You'll probably find the full entity name (which includes a mod id).
If you just want to find the actual mob in-game instead, you might be able to do so using LagGoggles, but that depends on how much lag the mob is generating. I'm assuming it's generating enough due to the message spam, but I could be wrong. With LagGoggles installed, press "insert" to open the menu for it, then run a profiling test -> World type. When it's done, open the menu again (if you closed it) and look at the results. If you see the mob on the left you can double click it to teleport to it.
Actually, I think LagGoggles might show the full entity name too, though I can't remember for sure atm.
Oh, and don't move around when running a World-type profiling in LagGoggles. It tends to crash the server if you do.
0
Not sure: I used Notepad++ to do an in-file search of all files within the archive, and it showed 0 results for "optifine" (not case sensitive).
I don't see that directory in the archive either.
Btw, your hyperlink is broken (I copy/pasted)
0
Minecraft, even in single-player mode, has a server module and a client module. What you saw was a case of the server-side hanging. This can be cause by many things, but in the end most of them come down to infinite loops in processing. I'm using the term "infinite loop" in a very broad way here.
In your particular case, based on your mention of general load on your machine, I'd say this is most likely due to hitting your memory allocation cap. I'm guessing the memory for MC is running out, causing the java garbage collector to run, which normally would be able to clear up some old memory and then continue running the game again (this is usually what's happening when you get a temporary version like you mentioned, with the "can't keep up" message). But if there is too much RAM that is still in active use, it will never recover. I'd think this would cause the program to terminate, but I could be wrong. I haven't run into this issue myself.
I seem to remember seeing this happen on another thread, though I can't remember for sure if it was triggered by something infinite or not (for some reason "infinite cows" seems to be on the tip of my mind? Was too long ago to remember).
If you can do something to increase the RAM given to MC, or to decrease the RAM it is using, I'd suggest trying to do so. If you're running a texture pack or shaders, I would personally try disabling those first. They don't have an effect on gameplay, so if disabling them fixes the issue, you know it's just RAM, and not some mod breaking (probably).
0
Yep, I think that's the right log. I see it's not just modded TEs, but also vanilla TEs which are failing to load.
My first guess is world corruption, which you already tested for ofc, but there is always the possibility that you were not loading the corrupted *part* of the world in the SP test? Generally, world corruption is not the entire world, but rather certain chunks, so if those chunks were being loaded in multiplayer mode but not SP mode for some reason...
Other than that, I didn't come up with any good ideas from the log (unless you count "maybe the SF4 team will have a better idea when they see this log").
Just for reference, how much RAM is the server allocated? I think when a player dies and respawns, the chunks at their respawn point load before the chunks where they died are released (though I'm not sure, so don't quote me on that). If this is true, and the server was already near max memory capacity, I could see it possibly causing corruption.
In any case, I'd suggest going to their issue tracker that I linked to in my previous message, and posting a new issue for this, with a link to this thread for reference. Even if it does end up being world corruption and you manage to wipe the offending chunks, it would mean that something somehow caused said world corruption, and it might happen again. And I could also be way off track, so getting them to take a look is probably a good idea.
0
Based on the crash report, this is probably coming from OTGCore (aka Open Terrain Generator). I highly suggest posting it on their issue tracker.
0
I don't see anything about "skipping block entity" in either of those, so I'm guessing they're not the logs I'm looking for. Generally, the logs you want are the ones found in the "logs" folder of the server and the client with issues. If you just ran a session and had an issue, it will be the log file named "latest". Otherwise it will be one of the others (or if the problem was too many runs ago, it will be gone).
I took a look through the MC code a bit, and that error is (as you might expect) just a generic "Hey, something went wrong" error. The things it can trigger off of are not specific enough to give me any solid clues. (If anyone is interested, it's in TileEntity.create() )
(TileEntity == BlockEntity; somewhere along the line, the most common name switched from BlockEntity to TileEntity, but they're the same thing)
You could also try the skyfactory 4 issue tracker, but I can almost guarantee they will also want that log file with the error message you mentioned in it.
0
Glad you got it working (I assume this means the manifest is using modid or something else, and not filename)
0
If you solve the issue, please post your solution for others to reference!
And ofc if neither of our solutions work, say so and maybe someone else will come in with another idea.
0
I unfortunately have no experience with this yet, but I'm interested, so I'll follow this thread and hope someone else can give some useful info :P
0
The filename is wrong. Notice the " (1)" at the end (ie. you downloaded it twice). I'm guessing that's all there is to it (ie. in the manifest, it's probably "CoFHCore-1.12.2-4.6.3.27-universal.jar" and not "CoFHCore-1.12.2-4.6.3.27-universal (1).jar").
But I could be wrong. I haven't actually messed with publishing a modpack yet, as mine is still in need of work before its ready.
0
Even just the mod listings in the log file are useful, and fairly often there is a little message somewhere that is overlooked (though I'm not saying that's necessarily the case for you here). I'd pastebin/hastebin and link it here just in case. I haven't played SF4 yet, but I'll at least look at the log file and see if I notice anything new, if you post it. I can also scrape through forge's code since I'm a mod dev myself and see where that particular error is triggering from in code (assuming the message lives in MC/Forge, and not in one of the mods in the pack).
0
In reply to Forge_User_32669499:
In reply to Forge_User_32669499:
This thread is about a mob causing server lag. You should probably both create your own threads for your problems ;)
0
In reply to Burnflamez:
I don't know which mod it's from, but if you want to find out, I have a library mod (which is normally just a requirement for my other mods) which goes through the registries and saves reference text files of all items, entities (mobs), potions, etc. You can use it to debug and then uninstall it afterwards if you like.
Here's the mod
The text files will appear in config/fantasticlib/reference, and will be created/updated when the game finishes starting up. Just open entitites.txt and search it for "scientist". You'll probably find the full entity name (which includes a mod id).
If you just want to find the actual mob in-game instead, you might be able to do so using LagGoggles, but that depends on how much lag the mob is generating. I'm assuming it's generating enough due to the message spam, but I could be wrong. With LagGoggles installed, press "insert" to open the menu for it, then run a profiling test -> World type. When it's done, open the menu again (if you closed it) and look at the results. If you see the mob on the left you can double click it to teleport to it.
Actually, I think LagGoggles might show the full entity name too, though I can't remember for sure atm.
Oh, and don't move around when running a World-type profiling in LagGoggles. It tends to crash the server if you do.