Re: Culled Wedged CPUworkers Bug

[marlowe <marlowe@xxxxxxxxxxxxxx>]

Roger Dingledine wrote:
> On Mon, Oct 23, 2006 at 09:56:56PM -0400, marlowe wrote:
> > I have discovered the following error message in my log files:
> >
> > "[notice] cull_wedged_cpuworkers(): Bug: closing wedged cpuworker. Can 
> > somebody find the bug?"
> >
> > Has anyone else seen this and if so what information can I provide to 
> > help track down this bug?
>
> I believe this was fixed in 0.1.1.22, which was released on 2006-07-05.
>
> What Tor version are you running? If it's a later version than that,
> can you provide more details, like how often it occurs, whether you can
> induce it to occur, whether there's high load at the time, whether your
> CPU sucks, whether you're running an obscure operating system, and other
> hints that might be helpful?
>
> --Roger
Roger,

I have installed 0.1.1.24 on CentOS 4.2 with full updates.  The CPU on 
the machine is on the lower end (1 GHz).  As previously noted on this 
list, I was having trouble getting my ORPort to be recognized as 
reachable until I compiled Tor with --disable-threads.  The threads 
problem disappeared with an upgrade to the CentOS kernel RPM, 
kernel-2.6.9-42.0.3.EL.

The error appears over the last couple of days in the log at 1000 and 
2200 EST each day.  I am not able to induce the error myself.  What 
other information (i.e. debug logs, etc) can I provide to help debug 
this issue?

Patrick