public inbox for netdev@vger.kernel.org
 help / color / mirror / Atom feed
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Jan Hinnerk Stosch <janhinnerk.stosch@gmail.com>
Cc: netdev@vger.kernel.org
Subject: Re: regression when connecting to ipv6 localhost
Date: Sun, 14 Oct 2012 23:35:42 +0200	[thread overview]
Message-ID: <1350250542.21172.17229.camel@edumazet-glaptop> (raw)
In-Reply-To: <CAO_aot6w9yaWacK9kE66m+Lv0LWXY9hKzLH8b=1dp=23fo_W2A@mail.gmail.com>

On Sun, 2012-10-14 at 23:21 +0200, Jan Hinnerk Stosch wrote:
> 2012/10/14 Eric Dumazet <eric.dumazet@gmail.com>:
> > On Sun, 2012-10-14 at 22:39 +0200, Jan Hinnerk Stosch wrote:
> >
> >> ntpd    368  ntp   16u     IPv4              10453      0t0    UDP *:ntp
> >> ntpd    368  ntp   17u     IPv6              10454      0t0    UDP *:ntp
> >> ntpd    368  ntp   18u     IPv4              10460      0t0    UDP
> >> localhost.localdomain:ntp
> >> ntpd    368  ntp   19u     IPv4              10461      0t0    UDP
> >> hinniepc.fritz.box:ntp
> >> ntpd    368  ntp   20u     IPv6               8775      0t0    UDP
> >> [fe80::922b:34ff:fe33:c292]:ntp
> >> ntpd    368  ntp   21u  netlink                         0t0  10463 ROUTE
> >
> > Hmm, your ntpd doesnt listen normally to ::1 UDP 123
> >
> > grep ntpd /var/log/syslog :
> >
> > (Here I have : )
> >
> > ntpd[4912]: ntpd 4.2.6p3@1.2290-o Mon Aug 20 15:15:21 UTC 2012 (1)
> > ntpd[4913]: proto: precision = 0.980 usec
> > ntpd[4913]: ntp_io: estimated max descriptors: 1024, initial socket
> > boundary: 16
> > ntpd[4913]: Listen and drop on 0 v4wildcard 0.0.0.0 UDP 123
> > ntpd[4913]: Listen and drop on 1 v6wildcard :: UDP 123
> > ntpd[4913]: Listen normally on 2 lo 127.0.0.1 UDP 123
> > ntpd[4913]: Listen normally on 3 eth1 172.30.42.8 UDP 123
> > ntpd[4913]: Listen normally on 4 lo ::1 UDP 123
> > ntpd[4913]: Listen normally on 5 eth1 fe80::222:15ff:fe87:3ff6 UDP 123
> > ntpd[4913]: peers refreshed
> > ntpd[4913]: Listening on routing socket on fd #22 for interface updates
> >
> >
> 
> Ah, ok. Didn't read correctly..
> 
> journalctl /usr/bin/ntpd gives:
> Oct 14 16:51:22 hinniepc ntpd[368]: proto: precision = 0.117 usec
> Oct 14 16:51:22 hinniepc ntpd[368]: ntp_io: estimated max descriptors:
> 1024, initial socket boundary: 16
> Oct 14 16:51:22 hinniepc ntpd[368]: Listen and drop on 0 v4wildcard
> 0.0.0.0 UDP 123
> Oct 14 16:51:22 hinniepc ntpd[368]: Listen and drop on 1 v6wildcard :: UDP 123
> Oct 14 16:51:22 hinniepc ntpd[368]: Listen normally on 2 lo 127.0.0.1 UDP 123
> Oct 14 16:51:22 hinniepc ntpd[368]: Listen normally on 3 eth0
> 192.168.178.27 UDP 123
> Oct 14 16:51:22 hinniepc ntpd[368]: Listen normally on 4 lo ::1 UDP 123
> Oct 14 16:51:22 hinniepc ntpd[368]: peers refreshed
> Oct 14 16:51:22 hinniepc ntpd[368]: Listening on routing socket on fd
> #21 for interface updates

So ntpd claims to "Listen normally on 4 lo ::1 UDP 123" but lsof doesnt
display a socket bound to ::1 port 123

So I would say there is a ntpd bug.

You could try to strace ntpd from the very beginning of its life, so
that we check what it does exactly.

service ntp stop
strace -o /tmp/STRACE /usr/bin/ntpd -g -u ntp:ntp

  reply	other threads:[~2012-10-14 21:35 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-14 15:38 regression when connecting to ipv6 localhost Jan Hinnerk Stosch
2012-10-14 19:29 ` Eric Dumazet
2012-10-14 19:48   ` Jan Hinnerk Stosch
2012-10-14 19:54     ` Eric Dumazet
2012-10-14 19:57       ` Jan Hinnerk Stosch
2012-10-14 20:25         ` Eric Dumazet
2012-10-14 20:26           ` Eric Dumazet
2012-10-14 20:28             ` Eric Dumazet
2012-10-14 20:39               ` Jan Hinnerk Stosch
2012-10-14 20:45                 ` Eric Dumazet
2012-10-14 20:47                   ` Eric Dumazet
2012-10-14 20:56                     ` Jan Hinnerk Stosch
2012-10-14 21:02                       ` Eric Dumazet
2012-10-14 21:07                 ` Eric Dumazet
2012-10-14 21:21                   ` Jan Hinnerk Stosch
2012-10-14 21:35                     ` Eric Dumazet [this message]
2012-10-14 21:54                       ` Jan Hinnerk Stosch
2012-10-14 22:01                         ` Eric Dumazet
2012-10-14 22:11                           ` Jan Hinnerk Stosch
2012-10-15  5:42                             ` Eric Dumazet
2012-10-15  7:03                               ` Jan Hinnerk Stosch
2012-10-15  7:14                               ` Joe Perches
2012-10-15  7:39                                 ` Eric Dumazet
2012-10-15  7:43                                   ` Eric Dumazet
2012-10-15 11:25                                   ` Joe Perches
2012-10-15 12:31                                     ` Eric Dumazet
2012-10-15 13:10                                       ` Jan Hinnerk Stosch
2012-10-15 13:21                                         ` Eric Dumazet
2012-10-15 13:28                                           ` Jan Hinnerk Stosch
2012-10-16 16:07                                             ` Jan Hinnerk Stosch
2012-10-16 16:43                                               ` Eric Dumazet
2012-10-16 16:55                                                 ` Eric Dumazet
2012-10-16 17:37                                                   ` [PATCH] ipv6: addrconf: fix /proc/net/if_inet6 Eric Dumazet
2012-10-16 18:30                                                     ` Jan Hinnerk Stosch
2012-10-16 18:39                                                       ` David Miller
2012-10-16 22:01                                                       ` Joe Perches
2012-10-16 18:55                                                     ` David Miller
2012-10-15  9:25                               ` regression when connecting to ipv6 localhost Eric Dumazet
2012-10-15 10:32                                 ` Jan Hinnerk Stosch
2012-10-15  9:07                         ` Jan Ceuleers

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1350250542.21172.17229.camel@edumazet-glaptop \
    --to=eric.dumazet@gmail.com \
    --cc=janhinnerk.stosch@gmail.com \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox