TorCtl Concurrency Bugs
To quote #tor-dev... 22:33 < atagar> mikeperry: There seems to be a couple concurrency bugs with shutting down torctl: (1) when connect fails to authenticate it needs to close the failed torctl instance to stop its main loop from continuing and (2) close joins on the event thread, but not self._thread causing occasional shutdown issues like [1]
14:14 < mikeperry> atagar: is there a trac entry for this, or when the scrollback is gone, or when this pastebin expires, this will be history?
14:16 < atagar> mikeperry: Heh, can do if you'd like. I'd be happy to make patches but I can't reliably repro the issue so testing would be a pita. I was hoping that you'd be able to either say "yup, there's a obvious bug here" or "nah, you're missing X".
14:17 < atagar> the second issue seems to occure around 1% of the time or less (guess I could try a script that starts/stops torctl a bunch and see if that does the trick...)
14:19 < atagar> NightMonkey seemed to be able to reliably repro the first issue, but he's usually not around and his repro seems to be platform dependent (gentoo)
[1] atagar@fenrir:~/Desktop/arm$ ./arm -e 2 Exception in thread TorLoop (most likely raised during interpreter shutdown): Traceback (most recent call last): File "/usr/lib/python2.6/threading.py", line 525, in __bootstrap_inner File "/usr/lib/python2.6/threading.py", line 477, in run File "/home/atagar/Desktop/arm/src/TorCtl/TorCtl.py", line 675, in _loop File "/home/atagar/Desktop/arm/src/TorCtl/TorCtl.py", line 824, in _read_reply <type 'exceptions.TypeError'>: 'NoneType' object is not callable