Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Flexible cable bugs #55

Open
wormzjl opened this issue May 28, 2017 · 70 comments
Open

Flexible cable bugs #55

wormzjl opened this issue May 28, 2017 · 70 comments

Comments

@wormzjl
Copy link
Contributor

wormzjl commented May 28, 2017

https://pastebin.com/EVhRV2xS

@Foghrye4
Copy link
Owner

I can't recreate that crash. Could you show me your setup?

@wormzjl
Copy link
Contributor Author

wormzjl commented May 28, 2017

It happens a bit random, probably by connecting ihl cable directly to ic2 cables

@Foghrye4
Copy link
Owner

I still could not catch it. I added additional null check in grid calculation in 0.640. This should solve a case.
http://www.mediafire.com/file/hgqj9ekcyagh457/ihl-tools_n_machines-0.640-1.7.10.jar

@wormzjl
Copy link
Contributor Author

wormzjl commented May 28, 2017

ok thx!

@yuyuyzl
Copy link

yuyuyzl commented May 29, 2017

Another stall happened in our server today, luckily we have recorded the stall log and an image of his wire connections.
ihlcablestall
And the stall log:
stall-2017-05-29_20.44.19.txt

Seems it would be a IHL&Uncomplication compatible problem.
And we're using Thermos.
Hoping this would help.

@Foghrye4
Copy link
Owner

Seems to be a fatal flaw of using NBT object as grid cable representative. Fixing this would take some time.

@Foghrye4
Copy link
Owner

Ok, try a new version (0.642), it should not cause stalls, because I get rid of NBT in energy net.

@tony987852
Copy link

It happened again. We have already updated to the latest version.

Here is the stall log:
stall-2017-05-31_23.06.49-2.txt

Thanks.

@Foghrye4
Copy link
Owner

I'm not sure in anything now, but I changed a code a little bit. Maybe this could help.
http://www.mediafire.com/file/5dx5ff93rmq152c/ihl-tools_n_machines-0.0.643-1.7.10.jar
BTW how complex is grid? Do it contain a circles?*


  • that's should be absolutely safe.

@wormzjl
Copy link
Contributor Author

wormzjl commented May 31, 2017

It quiet simple, just power source -> electrolysis bath with length less than 2

@wormzjl
Copy link
Contributor Author

wormzjl commented May 31, 2017

private IEnergyNetNode getHasCable(NBTTagCompound cable, IEnergyNetNode exclude)

Can you add some sort of size limit & debug warning on this hashmap? Might help pinpointing the issue
Also is it possible to print the location of bugged cable?

@Foghrye4
Copy link
Owner

@wormzjl I removed that hashmap. Source updated, check by yourself.

@wormzjl
Copy link
Contributor Author

wormzjl commented May 31, 2017

ok will test

@Foghrye4
Copy link
Owner

@wormzjl ...no. Does it stall?

@wormzjl
Copy link
Contributor Author

wormzjl commented May 31, 2017

Nvm, still testing
Just have a probable cause for the issue coming up in mind.
Thermos/ spigot "deactivates" entities if they are some distance away from players, will this cause issue with the cables?
Also if the nbt is not saved properly on chunk unload, on chunk reload will it crash like this?

@Foghrye4
Copy link
Owner

@wormzjl entities does not involved in enegy transfer unless they are touch ground and uninsulated or voltage in cable above insulation breakdown voltage. In that case they emit smoke particles.
Each time chunk containing energy node unloaded ... wait... I think I found a possible case. When TE is in uloaded chunk it is invalidated and removed from all lists. But in my case it is remains in grid TE list. So, when chunk is loaded again it will generate new instance of TE, which will be added to grid alongside with old invalid instance of TE.

@wormzjl
Copy link
Contributor Author

wormzjl commented May 31, 2017

So boom?

@Foghrye4
Copy link
Owner

Ok, try this: http://www.mediafire.com/file/taojc7ubcicbyke/ihl-tools_n_machines-0.644-1.7.10.jar
Also check fractionator column.

@wormzjl
Copy link
Contributor Author

wormzjl commented May 31, 2017

https://pastebin.com/PSijj7ht
crashed by linking an electrolysis bath with a mfe

@Foghrye4
Copy link
Owner

Foghrye4 commented Jun 1, 2017

Unexpected.
Well this version does not crash: http://www.mediafire.com/file/gw2fbm0yl7j7o5y/ihl-tools_n_machines-0.645-1.7.10.jar
(and as well does not work with uncomplication mod (energy not transferred after first packet))

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 1, 2017

Some weirdness in the enet?

@Foghrye4
Copy link
Owner

Foghrye4 commented Jun 2, 2017

@wormzjl probably demanded energy >0d but <1d bug.

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 3, 2017

Seems that when the electrolysis bath finishes it's work, it refuses to work again unless the cable is reconnected

@Foghrye4
Copy link
Owner

Foghrye4 commented Jun 3, 2017

For a freaking hour I tried one to another options to fix that. Screw it. Ok, IHL officially incompatible with Uncomplication mod until a further notice.

Maybe I should make electrolyzer to support work with normal cables?

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 3, 2017

That will lose some...identity

@Foghrye4
Copy link
Owner

@wormzjl I took my will into fist and found a lot of bugs related to flexible cable energy transfer system (0.646). I did not test fixes with uncomplication, but it was not an uncomplication related bugs at a first place. You may want to check a work with uncomplication now, and I need to sleep.

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 12, 2017

Thank you very much! gonna do some more testing

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 12, 2017

Ok, the power transfer from/to ic2 energy block is fine.
But it seems that the electrolysis bath is not receiving any power?
Also before you can attach wire to desired block face by simple shift + rightclick, but now you need to stand right before that side

@Foghrye4
Copy link
Owner

But it seems that the electrolysis bath is not receiving any power?

Is recipe conditions are fulfilled? Because everything fine at my side (even with uncomplication mod). Try this: launch Minecraft with opened console. Right-click at electrolysis bath with thermometer (from IHL creative inventory). Copy-paste 4 lower lines of console output here. Do a same thing with opposite side of cable connected (there should be invisible but clickable "junction box").

Also before you can attach wire to desired block face by simple shift + rightclick, but now you need to stand right before that side

O.O I did not change anything related to this part of code. At least as I remember.

@yuyuyzl
Copy link

yuyuyzl commented Jun 13, 2017

Also before you can attach wire to desired block face by simple shift + rightclick, but now you need to stand right before that side

That's true. I have not read your code but in game you should stand at the face you want to put wire and click or your wire will connect to other faces.

Because everything fine at my side (even with uncomplication mod).

And i'm sorry about my misleading to uncomplication. IC+IHL+Uncomplication works fine, The problem is Immersive engineering use a strange way to feed energy to IC capacitors, which causes the problem i've mentioned before. An IE->IC->IHL wire connection will cause the problem, but IE->IC->IC->IHL wire works fine.

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 16, 2017

No this is probably not the issue so I deleted it, still investigating

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 16, 2017

Ok this has nothing to do with dragonapi
Ihl cables only conducts power when energy source is facing east, you might want to have a look at the side checks

@Foghrye4
Copy link
Owner

You mean that energy source transfer energy to flexible cables only if it facing east? Why everything works fine without DragonAPI then?

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 16, 2017

yeah, tested without DragonAPI, no power emit when it's not facing east even with MFE->MFE connection, nothing to do with the electrolysis bath

@Foghrye4
Copy link
Owner

Ooops. My bad.

@Foghrye4
Copy link
Owner

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 18, 2017

thx, doing some live testing

@wormzjl
Copy link
Contributor Author

wormzjl commented Jun 27, 2017

no problem occurred yet, seems to be solved

@wormzjl wormzjl closed this as completed Jun 27, 2017
@wormzjl wormzjl changed the title Flexible cable crash with uncomplication Flexible cable bugs Jul 6, 2017
@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 6, 2017

Crashing again...gonna post them later

@wormzjl wormzjl reopened this Jul 6, 2017
@tony987852
Copy link

New bug:
When cable overload, it will crash the server.
I also wonder why it did not happen before today that using the same cable and the same energy network.

I tried 3 times.
crash-2017-07-06_10.02.03-server.txt
crash-2017-07-06_10.20.48-server.txt
crash-2017-07-06_10.31.45-server.txt

Thanks.

@tony987852
Copy link

tony987852 commented Jul 6, 2017

The stall happened again. I don't know which player is using the cable.

stall-2017-07-06_21.18.46.txt
stall-2017-07-06_19.39.39.txt

@Foghrye4
Copy link
Owner

Foghrye4 commented Jul 7, 2017

I wonder how that is happened. I add a code to prevent crash in 0.650 and it will generate some log output while removing corrupted tile-entities. I would like to see that log. As for stalls - if they happened again I would like to see log output as well.

@Foghrye4
Copy link
Owner

Foghrye4 commented Jul 8, 2017

@wormzjl is latest version of IHL fix issue?

@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 8, 2017

Been able to crash it on singleplayer, put down a MFE, connect it to a MV transformer then a diamond cable, then ihl cable -> electrolysis bath. Crash seems to happen on cable meltdown.
crash-2017-07-08_16.22.51-server.txt
fml-client-3.txt

Also got another bug, seems that the length of cable used is calculated by player move distance?
If I do not move during cable placement, the length will be 0

@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 8, 2017

ihl-tools_n_machines-0.648-1.7.10.jar
Umm you have updated to 650? did anything change related to the cable?
Going to test with the latest

@Foghrye4
Copy link
Owner

Foghrye4 commented Jul 8, 2017

@wormzjl yes. I added a code, which should remove nodes with ID == -1 from grid and generate some log output.

@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 8, 2017

Ok, can you take a look at the cable length issue mentioned above?

@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 8, 2017

Got the same result.....
fml-server-3.txt

@Foghrye4
Copy link
Owner

Foghrye4 commented Jul 8, 2017

Damn, I forgot to add a one more line.

@Foghrye4
Copy link
Owner

Foghrye4 commented Jul 8, 2017

@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 8, 2017

[17:05:27] [Server thread/ERROR] [ihl/]: Removing TE with grid ID==-1 ihl.processing.chemistry.ElectrolysisBathTileEntity If it happened more than a single time, report that to mod author.
[17:05:27] [Server thread/WARN] [Uncomplication/]: EnergyNet.emitEnergyFrom: ihl.flexible_cable.AnchorTileEntity@2d2179 is not added to the enet
[17:05:40] [Server thread/ERROR] [ihl/]: Removing TE with grid ID==-1 ihl.processing.chemistry.ElectrolysisBathTileEntity If it happened more than a single time, report that to mod author.
[17:05:40] [Server thread/WARN] [Uncomplication/]: EnergyNet.emitEnergyFrom: ihl.flexible_cable.AnchorTileEntity@2d2179 is not added to the enet
[17:05:53] [Server thread/ERROR] [ihl/]: Removing TE with grid ID==-1 ihl.processing.chemistry.ElectrolysisBathTileEntity If it happened more than a single time, report that to mod author.
[17:05:53] [Server thread/WARN] [Uncomplication/]: EnergyNet.emitEnergyFrom: ihl.flexible_cable.AnchorTileEntity@2d2179 is not added to the enet
[17:05:59] [Server thread/ERROR] [ihl/]: Removing TE with grid ID==-1 ihl.processing.chemistry.ElectrolysisBathTileEntity If it happened more than a single time, report that to mod author.
[17:05:59] [Server thread/WARN] [Uncomplication/]: EnergyNet.emitEnergyFrom: ihl.flexible_cable.AnchorTileEntity@2d2179 is not added to the enet

Got these in the console

Screenshot, the cable can be actually burned
1

2

@Foghrye4
Copy link
Owner

Foghrye4 commented Jul 8, 2017

So happen every time on cable burnout?

@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 8, 2017

Nope, seems to happen only on certain setups

@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 8, 2017

image

Seems somewhat related to direction, the one on the left will not give error on burnout, the one on the right will

@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 8, 2017

image
Also...the cable length bug

@wormzjl
Copy link
Contributor Author

wormzjl commented Jul 8, 2017

This is too difficult to replicate, maybe just let the electrolysis bath accept EU directly (from the upper side)

@Foghrye4
Copy link
Owner

Foghrye4 commented Jul 9, 2017

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants