Tried the latest build. Can't reproduce the memory leaks anymore. The energy cell aren't reverting to zero anymore, but I am seeing an... unusual behavior. I'm not sure if you'll believe me on this, but... every server reset the energy cell in one person's home saves correctly, whereas the ones in mine all reset to 50% at server start. I know that sounds bizarre, but that's what I'm looking at. Tried it a good few times to make sure I wasn't going out of my gourd. The cells in my home are being powered by a prodigious number of industrial steam engines (36), so the network there probably has a lot more activity going on All the cells I'm looking at are under a ChickenChunks Chunk Loader. Not sure if that makes any difference.
At any rate, it seems playable enough for the moment. I suppose this could be something funny going on in TE, rather than directly related to TT, but I don't really know. They saved correctly in my home without TT installed. I know its really odd. Could it be that they were at 50% last time I shut down the server without TT and it isn't overwriting them on server shut down?
Edit: I saw a weird exception in my log when I wasn't on the server, then logged on and saw that all my cells had been cut to about 30%. Here's the stack trace from the log:
2013-02-03 17:42:21 [SEVERE] [TickThreading] Exception during tile entity tick
ticking: c thermalexpansion.transport.tileentity.TileConduitEnergy thermalexpansion.transport.tileentity.TileConduitEnergy@4b97b169 x,y,x: -465, 116, -770 at x,y,z:-465,116,-770
Tick region: rX: -29, rZ: -48, hashCode: -3145757:
java.lang.NullPointerException
at java.util.LinkedList.node(Unknown Source)
at java.util.LinkedList.get(Unknown Source)
at thermalexpansion.transport.GridEnergy.transferEnergy(GridEnergy.java:108)
at thermalexpansion.transport.tileentity.TileConduitEnergy.g(TileConduitEnergy.java:331)
at me.nallar.tickthreading.minecraft.tickregion.TileEntityTickRegion.doTick(TileEntityTickRegion.java:130)
at me.nallar.tickthreading.minecraft.tickregion.TickRegion.run(TickRegion.java:51)
at me.nallar.tickthreading.minecraft.ThreadManager$2.run(ThreadManager.java:89)
at me.nallar.tickthreading.minecraft.ThreadManager$1.run(ThreadManager.java:40)
at me.nallar.tickthreading.minecraft.ThreadManager$ServerWorkThread.run(ThreadManager.java:150)
Another Update: I removed all the conduits coming out of the energy cells - restarting the server in this case does not cause any of the energy cells to misbehave and they all retain the correct charge. I'm going to try breaking and replacing all my conduits and see if that helps at all.
Another, Another Update: I replaced the "extraction" conduits above the cells and it seems like the cells now only lose ~3% of their total charge on server reset. Then, the exception occurred again while the server was still running and all the cells lost 50% once more. Later, it happened again and it dropped completely to zero.
Here's a screenshot of my setup where the problem occurs - I've got two setups just like this, across from one another:
