Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 9 Dec 2017 19:58:28 +0200
From:      Konstantin Belousov <kostikbel@gmail.com>
To:        Michael Tuexen <tuexen@FreeBSD.org>
Cc:        src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-stable@freebsd.org, svn-src-stable-11@freebsd.org
Subject:   Re: svn commit: r324686 - stable/11/sys/netinet
Message-ID:  <20171209175828.GT2272@kib.kiev.ua>
In-Reply-To: <20171203133946.GJ2272@kib.kiev.ua>
References:  <201710171242.v9HCgHG4030707@repo.freebsd.org> <20171203133946.GJ2272@kib.kiev.ua>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Dec 03, 2017 at 03:39:46PM +0200, Konstantin Belousov wrote:
> On Tue, Oct 17, 2017 at 12:42:17PM +0000, Michael Tuexen wrote:
> > Author: tuexen
> > Date: Tue Oct 17 12:42:17 2017
> > New Revision: 324686
> > URL: https://svnweb.freebsd.org/changeset/base/324686
> > 
> > Log:
> >   MFC r322648:
> >   
> >   Ensure inp_vflag is consistently set for TCP endpoints.
> >   
> >   Make sure that the flags INP_IPV4 and INP_IPV6 are consistently set
> >   for inpcbs used for TCP sockets, no matter if the setting is derived
> >   from the net.inet6.ip6.v6only sysctl or the IPV6_V6ONLY socket option.
> >   For UDP this was already done right.
> >   
> >   PR:		221385
> > 
> > Modified:
> >   stable/11/sys/netinet/tcp_usrreq.c
> > Directory Properties:
> >   stable/11/   (props changed)
> > 
> > Modified: stable/11/sys/netinet/tcp_usrreq.c
> > ==============================================================================
> > --- stable/11/sys/netinet/tcp_usrreq.c	Tue Oct 17 11:20:32 2017	(r324685)
> > +++ stable/11/sys/netinet/tcp_usrreq.c	Tue Oct 17 12:42:17 2017	(r324686)
> > @@ -1899,6 +1899,8 @@ tcp_attach(struct socket *so)
> >  #ifdef INET6
> >  	if (inp->inp_vflag & INP_IPV6PROTO) {
> >  		inp->inp_vflag |= INP_IPV6;
> > +		if ((inp->inp_flags & IN6P_IPV6_V6ONLY) == 0)
> > +			inp->inp_vflag |= INP_IPV4;
> >  		inp->in6p_hops = -1;	/* use kernel default */
> >  	}
> >  	else
> 
> I highly suspect that this commit causes the breakage.  Example is taken
> from the 'nestat -f inet6 -n' output, but in fact the problem is much
> more serious, e.g. squid acls are broken as well:
> tcp6       0      0 2001:470:1f14:13.23047 2a03:2880:f003:c.443   ESTABLISHED
> tcp46      0      0 0.0.0.1.22             69.223.183.85.58766    ESTABLISHED
> 
> The tcp46 and 0.0.0.1 values in the second line are clearly nonsensical.

Ping.  This looks like quite serious regression in stable/11.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20171209175828.GT2272>