On Saturday 01 November 2003 06:40 pm, Matt Edwards wrote:
> ---- Original Message -----
> From: "Dirk Meyer" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>; "Matt Edwards"
> Sent: Saturday, November 01, 2003 7:20 AM
> Subject: Re: apache mod_ssl build error
> > Matt Edwards schrieb:,
> >
> > > Yep.  Me again.  Sorry. :)  Did you ever find a solution to this (see
> below my
> > >  name) besides moving to 5.1?  I am currently having the same issue,
> (though n
> > > ot with portupgrade),  with the building of mod_ssl itself.  Trouble
> > > is,
> I'm r
> > > unning 5.1.
> >
> > I have no problem on FreeBSD 5.1
> >
> > > ap_fnmatch.c:143: `FNM_CASE_BLIND' undeclared (first use in this
> function)
> > you might habe a "fnmatch.h" ins /usr/local/include/
> > (maybe from the heimdal port)
> > $ pkg_info -W /usr/local/include/fnmatch.h
> >
> > If you deinstall this file, mod_ssl will build without problems.
> >
> > kind regards Dirk
> >
> > - Dirk Meyer, Im Grund 4, 34317 Habichtswald, Germany
> Thank you Dirk.  That was exactly what I needed to get it to build.  I did
> in fact have heimdal installed.  I appreciate you getting back to me on
> this.  I un-installed heimdal, built and installed mod_ssl and then
> reinstalled heimdal again.  The only question that remains now is will the
> re-install of heimdal cause any problems?  So far kerberos is working fine,
> but I have yet to configure apache for the ssl stuff to see if it is
> working ok (not that I can't, I just haven't gotten around to doing it yet.
> :).
> thanks,
> matt

And a big Thanks from me, too.  On just one of my servers, apache stopped 
portupgrading (within the past several weeks); it hadn't really grabbed my 
attention until a power failure in Newark on Wednesday made my operating 
version of apache+mod-ssl unrunnable (or so I thought) and I actually had to 
pkg_delete apache from the machine and then do a pkg_add -r apache+mod_ssl to 
get a working version back.

The upshot of the apache blasting power failure was that it wasn't apache at 
all that was the actual culprit.  When the machine rebooted and came back 
online, for some odd reason, portsentry grabbed port 80 and wouldn't let go.  
While apachectl startssl didn't return a command line error, the /var/log/
httpd-error logfile recorded that the port was already in use.  ps -ax |grep 
http only returned grep http.  I found the problem with netstat or sockstat 
(I forget now, it was a Real late night, that night), killed portsentry, 
restarted apache, and restarted portsentry.  All has been well since.  Now, 
though, I know what to do to properly portupgrade apache next time.


Tim Kellers

[EMAIL PROTECTED] mailing list
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Reply via email to