On 16-03-2005 18:06, "Adam R. B. Jack" <[EMAIL PROTECTED]> wrote: >> I just killed off 100+ processes on brutus again. Main offenders seemed to >> be mono (invoking NAnt), kaffe and 'cvs up', which of course all hang > around >> 'cuz the parent python process has trouble cleaning up after itself. There >> were a few python processes left hanging around as well. > > So is Mono "working" on Brutus now? Last I recalled (a ways back) we > couldn't do simple compiles. I'm vaguely recalling that Stefan (?) built > Mono from scratch to get it working. Is this the case? Just curious.
I have no idea. >> Hehehe. Adam, if you're looking for a challenge, it'd be real nice if you >> could get gump to clean up child processes after itself :-D > > Surely you've heard me whine about this over and over. Oh yeah, and over > and over. :-) :-) > It is the main thing that makes me think Python was not quite > the best tool for the Gump job. I tried 5 or 6 different ways of trying to > write it/fix it -- and clearly none worked. That said, I'm pretty sure it > can be done w/ use of process groups & am game for a rematch one of these > days. It must be possible. I know os.system() or one of the other calls returns a process id, so using that we should at least be able to kill all the processes we spawn directly. > BTW: What is odd is that I no longer see use of "timeout" (Sam's OSS find > that looks like it is either part of the OS distro now, or nicely installed, > see "man timeout"). Maybe I got cocky w/ the last attempt and thought it was > obsolete, 'cos I don't see it in the code at all. Shame. Maybe that could be > a quick and dirty way to get the job done. Hmm. I'd really love a "proper" way as well. I'm very stubborn. I believe this must be possible. Just need to learn enough about unix process management :-D Cheers, Leo --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
