Hehe, so I have a small problem where some scripts get wedged after they are forked. They're not taking up any CPU time, they're just zombie'ing along...but they're not listed as zombies.

Normally I kill scripts that have wedged themselves CPU wise, but this new type of wedge-ness was a bit difficult to diagnose.

The scripts should finish within at least</strong> an hour. Luckily, the scripts have the time that they were created written into their file name. Something like this

172.16.1.5-2008-04-16.024001-1208331905.out</code></span>

With a bit of crazy python, i extracted out the date and time to get a number. I can then compare that number to the current time, and if that number is less than an hour ago, splat, the script is killed. It will re-run it's self though when it is killed and should finish correctly the next time.

filename = os.path.basename('/root/172.16.1.5-2008-04-16.024001-1208331905.out')
pieces = filename.split('.')
filedate = ''.join(pieces[3].split('-')[1:])+pieces[4].split('-')[0]</code></span>

haha, man that's about as obscure as perl now isn't it? It will print out this

20080416024001</code></span>