Dennis Allison wrote:
Running htmldoc via an External Method that uses popen2() under zope 2.6.2b3 and python 2.1.3 (both by the book...) I get zombie processes that just hang in there... Now zombie processes exist when a parent does not wait() for a child, so the problem would look like it's in popen(). The htmldoc process seems to run to completion and generate the expected output. ... Oh, yes -- the OS is RH9 and the machine is a dual processor Athlon. I suspect some OS related problem since this same code works without creating Zombies on RH7.3.
Very likely. RH9 uses a new threading library which has had some, uh, growing pains. There are workarounds that you can find on the list (a certain environment variable will cause the old library to be used). Or keep on top of the latest versions and hopefully it'll clear up. (A bug report to RH and/or Python would probably help.) Can you reproduce it from regular Python? --jcc -- "Code generators follow the 80/20 rule. They solve most of the problems, but not all of the problems. There are always features and edge cases that will need hand-coding. Even if code generation could build 100 percent of the application, there will still be an endless supply of boring meetings about feature design." (http://www.devx.com/java/editorial/15511)