From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753632AbbE1OmB (ORCPT ); Thu, 28 May 2015 10:42:01 -0400 Received: from know-smtprelay-omc-7.server.virginmedia.net ([80.0.253.71]:54541 "EHLO know-smtprelay-omc-7.server.virginmedia.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751207AbbE1Olw (ORCPT ); Thu, 28 May 2015 10:41:52 -0400 X-Originating-IP: [81.106.150.188] X-Spam: 0 X-Authority: v=2.1 cv=JuUM15MC c=1 sm=1 tr=0 a=DGj713NdaxKrsjjgQne7PA==:117 a=DGj713NdaxKrsjjgQne7PA==:17 a=J0QyKEt1u0cA:10 a=IkcTkHD0fZMA:10 a=NLZqzBF-AAAA:8 a=h1PgugrvaO0A:10 a=IVdhC3tJdGl8ci1b5tkA:9 a=njDjB1uALaY54g9L:21 a=bfhStzq-rBJOr_KZ:21 a=QEXdDO2ut3YA:10 Date: Thu, 28 May 2015 15:41:49 +0100 From: Ken Moffat To: Cong Wang Cc: LKML , Linux Kernel Network Developers Subject: Re: Lost network connectivity in 4.0.x Message-ID: <20150528144149.GA29350@milliways> References: <20150524024352.GA15747@milliways> <20150524032938.GA16664@milliways> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline X-Clacks-Overhead: GNU Terry Pratchett Content-Transfer-Encoding: 8bit In-Reply-To: User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, May 27, 2015 at 10:53:00PM -0700, Cong Wang wrote: > (Please always Cc netdev for networking bugs.) > > On Sat, May 23, 2015 at 8:29 PM, Ken Moffat wrote: > > On Sun, May 24, 2015 at 03:43:52AM +0100, Ken Moffat wrote: > >> Anybody else suffering frm lost network connectivity in 4.0.x > >> kernels ? A couple of times this week, vim on an nfs-3 mount hung > >> and I had to reboot. Both of those occasions were on an AMD desktop > >> with the r8169 driver, running 4.0.3. I thought it might be > >> specific to that machine. For the last two or three days I've been > >> using an intel, and about 10 minutes ago it suffered the same problem > >> while running 4.0.4. Using ping from another term showed that it > >> had no connectivity to the server on my local network. > >> > >> This is a bit hard to diagnose - nothing in the logs. > >> > > I forgot to add that this is with the released gcc-5.1 : I keep > > forgetting that some people use old compilers ;-) > > > > Is there any way you can help to narrow down the problem? > Thanks for the reply. The problem is continuing to show up, but irregularly and often only after the machine has been booted for a long time (with s2ram, but I don't think I've used s2ram on every occasion). > For example: > > 1) What is your network setup? iptables? routes? etc. > I'm using iptables. Ah, yes - it started dropping packets around the time I last had a problem: May 27 00:48:26 ac4tv dhclient: DHCPREQUEST on eth0 to 192.168.7.254 port 67 May 27 00:48:27 ac4tv dhclient: DHCPACK from 192.168.7.254 May 27 00:48:27 ac4tv dhclient: bound to 192.168.7.152 -- renewal in 1787 seconds. That address came from my router, and I had been getting the same address for an hour, tbut then the dropped packet messages start appearing - they are for a different address, one that would have been offered by my server: May 27 00:53:16 ac4tv kernel: [31922.316798] IPTABLES Packet Dropped: IN=eth0 OUT= MAC=c8:60:00:97:07:35:bc:ae:c5:57:70:c5:08:00 SRC=192.168.7.11 DST=192.168.7.121 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=0 DF PROTO=TCP SPT=2049 DPT=1005 WINDOW=28960 RES=0x00 ACK SYN URGP=0 May 27 00:53:17 ac4tv kernel: [31923.316612] IPTABLES Packet Dropped: IN=eth0 OUT= MAC=c8:60:00:97:07:35:bc:ae:c5:57:70:c5:08:00 SRC=192.168.7.11 DST=192.168.7.121 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=0 DF PROTO=TCP SPT=2049 DPT=1005 WINDOW=28960 RES=0x00 ACK SYN URGP=0 and those continued until I forced a reboot. > 2) Can you check the stats to see if there is any error? > `ip -s -s li show`, `ethtool -S ` > I don't have ethtool installed, and that ip command appears ok at the moment: 1: lo: mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 RX: bytes packets errors dropped overrun mcast 3964 66 0 0 0 0 RX errors: length crc frame fifo missed 0 0 0 0 0 TX: bytes packets errors dropped carrier collsns 3964 66 0 0 0 0 TX errors: aborted fifo window heartbeat transns 0 0 0 0 0 2: eth0: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 link/ether c8:60:00:97:07:35 brd ff:ff:ff:ff:ff:ff RX: bytes packets errors dropped overrun mcast 224661061 277642 0 0 0 0 RX errors: length crc frame fifo missed 0 0 0 0 0 TX: bytes packets errors dropped carrier collsns 278152429 370438 0 0 0 0 TX errors: aborted fifo window heartbeat transns 0 0 0 0 6 > 3) Do a bisect? > > Thanks! That doesn't seem very practical when the machine is ok for a couple of days at a time. ĸen -- Nanny Ogg usually went to bed early. After all, she was an old lady. Sometimes she went to bed as early as 6 a.m.