RE: slapd segfaults with bad LDAP URL (ITS#2132)

[hyc@highlandsun.com]

OK. I've reverted that patch. Waiting for Hallvard's issues to settle out.

  -- Howard Chu
  Chief Architect, Symas Corp.       Director, Highland Sun
  http://www.symas.com               http://highlandsun.com/hyc
  Symas: Premier OpenSource Development and Support 

> -----Original Message-----
> From: Kurt D. Zeilenga [mailto:Kurt@OpenLDAP.org]
> Sent: Wednesday, October 09, 2002 12:01 PM
> To: hyc@highlandsun.com
> Cc: openldap-its@OpenLDAP.org
> Subject: RE: slapd segfaults with bad LDAP URL (ITS#2132)
> 
> 
> This is done for a reason...
> 
> Some systems support mapped addresses.  On these, it is NORMAL
> for an "any" IPv4 bind following an "any" IPv6 bind to fail
> because the "any" IPv6 bind accepts both IPv6 and IPv4.
> 
> Kurt
> 
> At 11:46 AM 2002-10-09, hyc@highlandsun.com wrote:
> >> -----Original Message-----
> >> From: Hallvard B Furuseth [mailto:h.b.furuseth@usit.uio.no]
> >
> >> > The code is set to loop through all the addresses that a hostname
> >> > lookup returns. It will only fail when all of the 
> addresses fail. If
> >> > you get EADDRINUSE on one of them, but another succeeds, it will
> >> > consider this a success.
> >>
> >> That sounds like a bug: I wouldn't want slapd to start
> >> happily with only
> >> v6 without and let me think it had v4 (too).  For now I can
> >> give the -4
> >> option to avoid that, but when the time comes that I do want
> >> v6 as well
> >> it could become a problem.  Or if it started as v6-only 
> while I only
> >> wanted v4 and didn't know about v6.
> >
> >I guess I agree with you. I've now changed it to fail 
> instead of continue
> >when bind fails.
> >
> >  -- Howard Chu
> >  Chief Architect, Symas Corp.       Director, Highland Sun
> >  http://www.symas.com               http://highlandsun.com/hyc
> >  Symas: Premier OpenSource Development and Support
>