| 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
| ||
|
Message-ID: <5B356C76.3050809@mutluit.com>
Date: Fri, 29 Jun 2018 01:17:10 +0200
From: "U.Mutlu" <um@...luit.com>
To: netdev@...r.kernel.org
Subject: Diagnosing network module for missing link establishment (cxgb3,
Chelsio T320)
Hi,
I've got a pair of used old dual port 10GbE NICs (Chelsio T320
10GBASE-R RNIC (rev 3) PCI Express x4 MSI-X) with 2 modular
transceivers on board the 2 NICs (ie. these can be taken off
of the card for replacement etc.).
The problem is that the cards don't establish a link;
the green LEDs go off after a few seconds after loading
the driver named cxgb3, and there is no indication in the syslog
about any error.
The behavior is the same whether the transceivers are present
on the card or not; the drivers always load successfully.
In the kernel sources the driver is located under
drivers/net/ethernet/chelsio/cxgb3
(I haven't recompiled it; just using the stock kernel from Debian repo).
Below, the interfaces are eth4 and eth5.
There was just once a link, but it never happens again, and
such log entries about link are not happening since then:
Jun 27 13:06:09 c6-local vmunix: [ 504.108102] cxgb3 0000:01:00.0 eth4: link
up, 10Gbps, full-duplex
Jun 27 13:17:30 c6-local vmunix: [ 1185.450256] cxgb3 0000:01:00.0 eth4: link down
So, how can I diagnose and pinpoint what the reason is for not establishing a
link?
The following is from a later reboot:
# dmesg | grep -i "Chelsio\|cxgb3\|eth[0-9]"
[ 0.979800] cxgb3: Chelsio T3 Network Driver - version 1.1.5-ko
[ 0.984162] r8169 0000:02:00.0 eth0: RTL8168evl/8111evl at
0xffffc90000002000, 74:d4:35:92:72:1b, XID 0c900800 IRQ 43
[ 0.984163] r8169 0000:02:00.0 eth0: jumbo features [frames: 9200 bytes, tx
checksumming: ko]
[ 1.319780] cxgb3 0000:01:00.0: irq 55 for MSI/MSI-X
[ 1.319785] cxgb3 0000:01:00.0: irq 56 for MSI/MSI-X
[ 1.319788] cxgb3 0000:01:00.0: irq 57 for MSI/MSI-X
[ 1.319791] cxgb3 0000:01:00.0: irq 58 for MSI/MSI-X
[ 1.319794] cxgb3 0000:01:00.0: irq 59 for MSI/MSI-X
[ 1.319797] cxgb3 0000:01:00.0: irq 60 for MSI/MSI-X
[ 1.319799] cxgb3 0000:01:00.0: irq 61 for MSI/MSI-X
[ 1.319803] cxgb3 0000:01:00.0: irq 62 for MSI/MSI-X
[ 1.319805] cxgb3 0000:01:00.0: irq 63 for MSI/MSI-X
[ 1.319828] cxgb3 0000:01:00.0: Port 0 using 4 queue sets.
[ 1.319868] cxgb3 0000:01:00.0: Port 1 using 4 queue sets.
[ 1.319909] cxgb3 0000:01:00.0 eth1: Chelsio T320 10GBASE-R RNIC (rev 3)
PCI Express x8 MSI-X
[ 1.319949] cxgb3: eth1: 128MB CM, 256MB PMTX, 256MB PMRX, S/N: PT37080022
[ 1.319986] cxgb3 0000:01:00.0 eth2: Chelsio T320 10GBASE-R RNIC (rev 3)
PCI Express x8 MSI-X
[ 5.217529] systemd-udevd[384]: renamed network interface eth0 to eth3
[ 5.342653] systemd-udevd[387]: renamed network interface eth2 to eth4
[ 5.358663] systemd-udevd[383]: renamed network interface eth1 to eth5
[ 11.563555] r8169 0000:02:00.0 eth3: link down
[ 11.563668] r8169 0000:02:00.0 eth3: link down
[ 13.917486] r8169 0000:02:00.0 eth3: link up
# ip link show
1: lo: <LOOPBACK,UP,LOWER_UP> 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
2: eth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP
mode DEFAULT group default qlen 1000
link/ether 74:d4:35:92:72:1b brd ff:ff:ff:ff:ff:ff
3: eth5: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN mode
DEFAULT group default qlen 1000
link/ether 00:07:43:05:8b:16 brd ff:ff:ff:ff:ff:ff
4: eth4: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN mode
DEFAULT group default qlen 1000
link/ether 00:07:43:05:8b:17 brd ff:ff:ff:ff:ff:ff
# ethtool eth4
Settings for eth4:
Supported ports: [ AUI FIBRE ]
Supported link modes: 10000baseT/Full
Supported pause frame use: No
Supports auto-negotiation: No
Advertised link modes: Not reported
Advertised pause frame use: No
Advertised auto-negotiation: No
Speed: Unknown!
Duplex: Unknown! (255)
Port: FIBRE
PHYAD: 1
Transceiver: external
Auto-negotiation: off
Supports Wake-on: d
Wake-on: d
Current message level: 0x000000ff (255)
drv probe link timer ifdown ifup rx_err tx_err
Link detected: no
# ifconfig
...
eth4 Link encap:Ethernet HWaddr 00:07:43:05:8b:17
inet addr:192.168.50.4 Bcast:192.168.50.255 Mask:255.255.255.0
UP BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
Interrupt:18 Memory:fe811000-fe811fff
eth5 Link encap:Ethernet HWaddr 00:07:43:05:8b:16
inet addr:192.168.60.5 Bcast:192.168.60.255 Mask:255.255.255.0
UP BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
Interrupt:18 Memory:fe811000-fe811fff
(I also tried same network 192.168.50 for both, no difference in outcome)
Btw, can a link be established between the 2 transceiver ports on the
same card? I think this should be possible, right?
Kernel: 3.16.0-4-amd64 (Debian v8 stock kernel)
--
Thx
Powered by blists - more mailing lists