zzz
back, read scrollback, dr|z3d was it on you, or you think it's on me, or you don't know?
dr|z3d
not entirely sure, zzz, I double checked the merge, nothing seemed amiss. so either I've missed a previous merge that enabled super somewhere..
dr|z3d
I'm assuming you're running the patch without issue.. because if you had issues, you'd know about it.
dr|z3d
was the "super" part of the deal something to do with the whole segmented client stuff, because that was a huge headache to merge.. that's the only thing I can think might be slightly off.
zzz
dr|z3d, pretty sure I tested it, and I'm running it again right now
zzz
the thing is, your merge rev diff looks _nothing_ like my diff for my rev
zzz
if it were one-for-one I'd be looking at my stuff. but yours is like hundreds of lines of diff
dr|z3d
it's not immediately obvious what the issue is, which is why I needed T3s|4 to prod me a couple of times and not_bob_afk to mention ramble.i2p to take note.
dr|z3d
diff is all whitespace apart from the changed lines, sorry about that.
zzz
if IMD is broken the router is broken; logging there should make things apparent
dr|z3d
yeah, you'd think.. I don't recall seeing anything obvious there in the logs.
zzz
no need for sorry, nobody died, and you do you, but it makes it hard to help from here
dr|z3d
and affected routers seemed to run mostly fine.. issue seemed related to service hosting.
zzz
the 'super' thing is moving setReceivedBy() from LeaseSet to DatabaseEntry (its superclass) which was a part of the subdb stuff
zzz
last year
dr|z3d
yeah, figured. all I got when I grepped for super was: super.setReceivedBy(localClient); in Leaseset.java
zzz
in theory my rev was just a microoptimization you can ignore, but you may wish to track it down anyway, and I will be building on all this for the LS spam stuff
zzz
so atm I'm assuming its a plus-only issue but will keep an ear out for reports
dr|z3d
yeah, ignoring for now, but I'll dive deeper and see if I can find the root cause, might be some sudb stuff that I didn't merge.
dr|z3d
thanks
zzz
you or anybody else could also just eyeball/review the diff for that canon rev line by line in about two minutes, which I've done at least 4 times
dr|z3d
yup, I've reviewed my revoked merge several times, looked absolutely fine relative to your commit.
zzz
I make mistakes all the time, may have here, but don't see it atm
dr|z3d
I'll give it another looksee later, just in case.
zzz
okey dokey. It's not a forgiving piece of code, as you see, bugs are pretty fatal