[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [tor-bugs] #3701 [Torctl]: PyTorCtl shouldn't os.kill() itself

#3701: PyTorCtl shouldn't os.kill() itself
-----------------------+----------------------------------------------------
 Reporter:  mikeperry  |          Owner:     
     Type:  defect     |         Status:  new
 Priority:  major      |      Milestone:     
Component:  Torctl     |        Version:     
 Keywords:             |         Parent:     
   Points:             |   Actualpoints:     
-----------------------+----------------------------------------------------

Comment(by aagbsn):

 Here are some options I've thought of:

 1. Make main thread aware of the child thread state (via signals) and exit
 when appropriate.
   - pro: Could be implemented with a custom signal/handler and os.kill()
 fairly trivially.
   - con: not generic; may require programs that use TorCtl to implement
 signal handling.

 2. Allow BwAuthority to die when TorCtl fails. This has been the default
 behavior.
   - pro: current behavior, requires no change
   - con: BwAuthority dies.

 3. Do not exit BwAuthority when bwauthority_child is killed with SIGTERM.
 (Restart child)
   - pro: trivial to implement
   - con: makes bwauthority a pain to kill, though could modify run_scan.sh
 to use kill -9

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/3701#comment:1>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs