httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jim Jagielski <...@jaguNET.com>
Subject Re: Status bug FIX ??
Date Fri, 04 Dec 1998 21:07:30 GMT
Dean Gaudet wrote:
> 
> On Fri, 4 Dec 1998, Harrie Hazewinkel wrote:
> 
> > Dean Gaudet wrote:
> > > 
> > > On Fri, 4 Dec 1998, Harrie Hazewinkel wrote:
> > > 
> > > > And since we only use the "server_hostname" of the server_rec
> > > > for "mod_status" I now use the "server_hostname" ptr directly
> > > > in the scoreboard.
> > > 
> > > That, too, breaks across restart.
> > 
> > Funny, not with me. My patch replaces the "vhost name" ptr
> > with a "NULL" during the restart/gracefull. this is done
> > directly after the set of "exit-generation" in the
> > scoreboard. Then "mod_status" checks first this if it is
> > NULL. If so it prints a dash.
> 
> If two children are from different generations they MAY have different
> memory maps.  You're just getting lucky.  I'll commit a fix which works.
> 

I mentioned that the fix still had a race condition... To fully fix
requires, I think, some changes to the API which we need to keep track
of the generations of the child processes and the parent. At the
least, the scoreboard itself will need some extra data. Even with this,
I'm not sure if that totally removes the race or not.

-- 
===========================================================================
   Jim Jagielski   |||   jim@jaguNET.com   |||   http://www.jaguNET.com/
            "That's no ordinary rabbit... that's the most foul,
            cruel and bad-tempered rodent you ever laid eyes on"

Mime
View raw message