reactor and everything it need to work properly must be inside same chunk, this is mustknow info everyone must know, not a bug and cannot be fixed.
reactor and everything it need to work properly must be inside same chunk, this is mustknow info everyone must know, not a bug and cannot be fixed.
Note that i build EVERYTHING (except the CF-Outerwalls) in ONE Chunk and the signal comes from bottom so its definetly in the same chunk.
in your case it's possibly bug, reactor dont store it's run state in NBT and always loaded as running, if redstone not properly updated it will possible bugout.
same bug with repeateronly clockgenerators, on chunk reload both repeaters will be locked in ON state ever if this is not possible in runtime.
to fix froblem build 5clock clockgenerator and link to your main system with XOR gate, it will refresh redstone state and wont affect reactor directly.
also chunkreload detector viable option (2 repeaters linked on each other with pulse starter, on chunk reload they will be locken in ON state both)
in your case it's possibly bug, reactor dont store it's run state in NBT and always loaded as running, if redstone not properly updated it will possible bugout.
That's exactly what i wanted to describe. And i know Chunkloaddetectors. I just don't want to build a massive overhaul around the Reactor because of a small critical Bug.
Edit: I know that Computercraft is Ideal for that case, but that Bug had probably destroyed more Bases, as we can imagine. It would be nice if someone (looking in direction of RichardG) would notice that and reply, so that i know that it's even noticed.
Will look into storing the state in NBT. Perhaps this bug caused Direwolf's reactor to explode 
Perhaps this bug caused Direwolf's reactor to explode
Uhhhm not perhaps. I saw his Version and that is definetly exploded, because of that (timing was right, Controlcircuit was applied, BOOM)
steal some spounge (vanilla block) code for AOE blockupdate onload, this will be nice addition to storing state in NBT (it shoud fix no eu after chunkreload problem)
Jeb removed the code which makes sponges work
Jeb removed the code which makes sponges work
That was Notch long time ago. Actually i'm using a CC-StartUp-program as Chunkloaddetector and Reactorshutoff.
Asides from that: YAY, this is the first Post i made from my new, better and used Computer, which i got for free! If that isn't an effective way to get one then i dont know.
I can confirm this bug 100%. I even hacked in some debug statements into the reactor code to hunt this bug down.
Unfortunately, I could not figure out for certain why the call to check if a block is getting redstone was failing. (bukkit, the vanilla server, and the vanilla client all have a very similar call to base code)
Eventually I just gave up and made my reactors have automatic shutdown when they overheat.
I mean the code which updates all nearby blocks was removed by Jeb in 1.2.
Sadly we don't have enough time to get that fix in the 1.2.5 version.
Sadly we don't have enough time to get that fix in the 1.2.5 version.
You save variables to NBT hundreds of places in the code. What's a cntrl-C and a cntrl-v, and changing the name of the variable going to cost you in terms of time?
In fact, if you want to be clever, save the value in the upper bits of the heat variable. Thus no file format changes at all, just a bit shift operation.
NBT value is croach, not complete fix.
Sadly we don't have enough time to get that fix in the 1.2.5 version.
I'd rather have a delayed update with this fix rather than a faster update without the fix, but if not, I think this bug is big enough to deserve a hotfix.
It is possible to work around this bug. Just never stop cooling you reactor, and it will not blow up. You do need an active cooling system for this to work, of course....
It is possible to work around this bug. Just never stop cooling you reactor, and it will not blow up. You do need an active cooling system for this to work, of course....
yeah.. even on SSP I never turn the cooling system off on the ice CAUSC I build. Don't see the point when RP2 tubes are intelligent enough to not send more into a full reactor.
