[Zope] ZServer dying

Mayers, Philip J p.mayers@ic.ac.uk
Wed, 28 Mar 2001 13:34:19 +0100


I'm having trouble with Zope periodically dying:

1474 /usr/bin/python /usr/local/Zope/z2.py -D -u zope -X -w 8080
1479  \_ /usr/bin/python /usr/local/Zope/z2.py -D -u zope -X -w 8080
1484      \_ /usr/bin/python /usr/local/Zope/z2.py -D -u zope -X -w 8080
1485          \_ [python <defunct>]
1486          \_ /usr/bin/python /usr/local/Zope/z2.py -D -u zope -X -w 80
1487          \_ /usr/bin/python /usr/local/Zope/z2.py -D -u zope -X -w 80
1488          \_ /usr/bin/python /usr/local/Zope/z2.py -D -u zope -X -w 80

The top-level process is doing a wait4() on it's child. Normally, it's child
is doing a select() on the 8080 port, any open connections and a FIFO (which
it presumably will use to communicate with the children). When the child's
top-level child dies, strace shows:

 [wait(0x137f) = 1479]
pid 1479 stopped, [SIGSTOP]

I'm presuming that 1484-8 are threads, since I can't strace/ltrace them. I'm
stumped - how can I debug this? Is one of the threads dying - what is it
likely to be? I've turned on full debugging on the SQL side of things, I'm
pretty certain it's not the server-side connection dying

Help! (Zope 2.3.0 in RedHat 7.0 SMP kernel - Booting with the UP kernel
didn't help)

Regards,
Phil

+----------------------------------+
| Phil Mayers, Network Support     |
| Centre for Computing Services    |
| Imperial College                 |
+----------------------------------+