linux-kernel - [POSSIBLE BUG]: tg3 network is "lost"

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 02 Jul 2007 10:20:22 +0200
From:	serge <lkm@...-linux.com>
To:	linux-kernel@...r.kernel.org
Subject: [POSSIBLE BUG]: tg3 network is "lost"

Hello everybody.

This is a second post, because title was wrong on the first post. Please
excuse my english, i'm french.

I have a big trouble with several server, web server. They are all
debian sarge based.

Program that run on tem are principally Apache2, PHP5, Zend cluster with
session management, Postfix and NFS client

Theses servers work well and, suddendly, one of them lost network.

Let me describe when i say "lost network":
- The server doesn't answer to a ping, all network application on it
doesn't anwer too
- The server can't ping another server, or make another network connection
- The server loose NFS communication with the NFS server
- On the console, if i type ifconfig i see the interface configured with
it's own IP
- On the console if i launch iptraf i only see icmp host unrechable message
- On the console if i type "mii-tool" it say that link is 100FX
negociated and link is UP
- The only way i have found to solve the trouble is to launch
"/etc/init.d/network restart". the network come back.
- When the server "lost" network, i don't have any message even with
debug activated on kernel
- When i restart network, i have these kernel message (seems to be
"normal"):

92554.221511] PM: Writing back config space on device 0000:02:02.0 at
offset b (was 164814e4, writing d00e11)
[492554.221528] PM: Writing back config space on device 0000:02:02.0 at
offset 3 (was 804000, writing 804010)
[492554.221534] PM: Writing back config space on device 0000:02:02.0 at
offset 2 (was 2000000, writing 2000010)
[492554.221540] PM: Writing back config space on device 0000:02:02.0 at
offset 1 (was 2b00000, writing 2b00146)
[492554.525866] PM: Writing back config space on device 0000:02:02.0 at
offset b (was 164814e4, writing d00e11)
[492554.525883] PM: Writing back config space on device 0000:02:02.0 at
offset 3 (was 804000, writing 804010)
[492554.525889] PM: Writing back config space on device 0000:02:02.0 at
offset 2 (was 2000000, writing 2000010)
[492554.525895] PM: Writing back config space on device 0000:02:02.0 at
offset 1 (was 2b00000, writing 2b00146)
[492556.321312] tg3: eth0: Link is up at 100 Mbps, full duplex.
[492556.321334] tg3: eth0: Flow control is on for TX and on for RX.

- Theses server are IP statics (not DHCP client).
- This is not cable or switch trouble (servers are on different switch)
- They are kernel 2.6.19.3, but we have the same trouble with other
kernel like 2.6.17.1

I attach you some information about server, like lspci -vvv, config.gz.

Thanks for reading me and for your help.

Download attachment "config.gz" of type "application/gzip" (7471 bytes)

View attachment "lscpi_vvv.txt" of type "text/plain" (15184 bytes)