[tor-bugs] #4345 [Tor Relay]: Bug: closing wedged cpuworker.

[Tor Bug Tracker & Wiki]

#4345: Bug: closing wedged cpuworker.
-----------------------+----------------------------------------------------
 Reporter:  jn         |          Owner:                   
     Type:  defect     |         Status:  needs_information
 Priority:  normal     |      Milestone:                   
Component:  Tor Relay  |        Version:  Tor: 0.2.2.34    
 Keywords:             |         Parent:                   
   Points:             |   Actualpoints:                   
-----------------------+----------------------------------------------------

Comment(by nickm):

 So, this message occurs when Tor has sent a question to a cpuworker, and
 hasn't gotten a response for an insanely long time.

 One way to try to fix this, if you know your way around gdb, is to use gdb
 -p to attach to the tor process after the bug occurs, and look at each
 thread's stack in turn to see where each one is.  If one of them really is
 stuck (on a lock or something), that should tell you where.  Or you could
 try it after the process has been running for an hour or two: if this
 happens reliably, then odds are one or more cpuworkers will be stuck by
 then.

 To provoke this message into occurring more quickly, we might lower
 CPUWORKER_BUSY_TIMEOUT; it is ridiculously high at the moment.

 Any other ideas?

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/4345#comment:3>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online