The following reply was made to PR mod_jserv/5755; it has been noted by GNATS.
From: Ed Korthof <[EMAIL PROTECTED]>
To: Vladislav Malyshkin <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Subject: Re: mod_jserv/5755: mod_jserv and httpd -HUP restarting problem
Date: Tue, 15 Feb 2000 07:17:53 -0800 (PST)
Hi --
First, I have to ask -- do you see this error occuring when you send
SIGHUP to only the parent process? I won't be surprised if that's the
case, but I want to know either way.
The location which you've shown isn't the best place for this, fyi -- look
around line 201 to see the place where this ought to go (you'll see one
call like this, for SIGTERM, but none for SIGHUP). The difference isn't
huge, but there's no point in registering the signal handler every time we
go through the loop (except for SIGALRM, since that handler may need to
run repeatedly). The fact that we don't set a signal handler for SIGHUP
may or may not be a real problem -- that's why I want to know if you still
see the problem when you only send SIGHUP to the parent.
Apache sets up its own signal handler for SIGHUP -- and I'm a little
hesitant to mess with that. When that handler recieved a SIGHUP, it
starts the standard Apache cleanup -- I don't think this cleanup is needed
or even appropriate w/in the wrapper process (which isn't like a normal
httpd child) but I'd want to verify that this isn't going to cause
problems before changing it.
mod_jserv registers a clean function, so when the standard Apache cleanup
is run, that function should shut down the JVM (or at least you should see
a message to the extent that it's trying to do so). OTOH, it's possible
that there is a race condition or some other error, when SIGHUP is used.
(I think it's been tested much more extensively with SIGTERM.)
Ed
--
See you at ApacheCon 2000 in Orlanda, Florida, March 8-10, 2000.
http://www.Apachecon.com/
