Bugs with CURRENT.

[William F. Maton]

Oops, just completing the record...


wfms

---------- Forwarded message ----------
Date: Wed, 27 Feb 2002 19:43:33 -0500 (EST)
From: William F. Maton <wmaton at ryouko.dgim.crc.ca>
To: Jeffrey M. Vinocur <jeff at litech.org>
Subject: Re: Bugs with CURRENT.

On Wed, 27 Feb 2002, Jeffrey M. Vinocur wrote:

> On Wed, 27 Feb 2002, William F. Maton wrote:
>
> > 	I'm using INN-CURRENT-20020224, as part of my IPv6 work.  So far,
> > I'm noticing one bug:  innreport takes forever to run.
>
> Hrm.  Mine runs just fine (takes about eight minutes, but I'm a small
> site).  The only change we made to innreport in the IPv6 patch is:

I've got this happenning on a small site as well.

> -    if ($left =~ /(\S+) connect$/o) {
> +    if ($left =~ /(\S+) (\([0-9a-fA-F:.]*\) )?connect$/o) {
>
> and I don't see how that would get you any running time issue.  Does
> somebody who really knows Perl want to comment if that should be a
> problem?

Ah!  A regex!! :-)

Doesn't look like it on first read.

> > syslogd is going nuts after some 10 minutes of initial INN startup.  I
> > find the following entries in /var/log/syslog:
> >
> > Feb 27 18:56:34 crc-news innd: SERVER internal SITEreader
> > Feb 27 18:56:43 crc-news last message repeated 168588 times
>
> Hrrrrrm.  (I don't see this either, of course.)  I haven't peered much at
> the innd internals, but that appears to be a this-should-never-happen
> message from innd/site.c:SITEreader() -- if no one else has an idea I'll
> try to take a better look later.)

Actually, this bug is what is making syslogd go nuts.  Squash this one
will take care of that.  But I don't understand what's up with innreport.

Let me know if you want access to mcast.dgim.crc.ca to have a look around.
That's what it is there for.

wfms