Re: 0xff bug

[Chris Moller <cmoller at redhat dot com>]

Hmm...

Wild guess:  is it possible the handleTerminatingEvent/sendContinue
sequence is another manifestation of the out-of-order stuff observed in
bz 202115?

https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=202115

cm

Chris Moller mumbled something on 09/01/2006 11:19 AM:
> Turns out I actually did get hit by this bug about two weeks ago, but I
> expect wasn't looking for it at the time.  Anyway, my one and only
> personal hit has one thing in common with the only other log of it I've
> seen, the one on the bz.
>
> The thing both of these hits have in common is the following sequence:
>
> {frysk.proc.LinuxTask@10fe4e0,pid=4715,tid=4715,state=running}
> handleTerminatingEvent
> {frysk.proc.LinuxTask@10fe4e0,pid=4715,tid=4715,state=running} sendContinue
> {frysk.proc.LinuxHost$PollWaitOnSigChld@55bb40,sig=Sig_CHLD} execute
> frysk.sys.Wait pid 4715 status 0xff
> frysk.sys.Wait pid 0 errno 0
>
> with different values for the pid, etc. 
>
> This looks to me like a process has been terminated (the
> handleTerminatingEvent thing) but then continued (the sendContinue thing).
>
> I may be mis-reading this, but it would be handy if everyone could check
> their frysk logs for that sequence and let me know if, (a) it occurs,
> and (b) if a handleTerminatingEvent/sendContinue sequence inevitably
> leads to the 0xff.
>
> cm
>
>

Attachment: signature.asc
Description: PGP signature

Attachment: signature.asc
Description: OpenPGP digital signature