Scheduled Maintenance: We are aware of an issue with Google, AOL, and Yahoo services as email providers which are blocking new registrations. We are trying to fix the issue and we have several internal and external support tickets in process to resolve the issue. Please see: viewtopic.php?t=158230

 

 

 

[SOLVED] Ethernet interface seems "deaf" (NVIDIA MCP61 a2)

Linux Kernel, Network, and Services configuration.
Post Reply
Message
Author
mdutra
Posts: 7
Joined: 2021-07-26 18:54

[SOLVED] Ethernet interface seems "deaf" (NVIDIA MCP61 a2)

#1 Post by mdutra »

Hello,

I am relatively new to Debian (I've been using it for about 2 years) but I am an experienced IT professional and I am having a really tough time replacing my WiFi link for a cable connection.
I cannot get an IP using DHCP ("No DHCPOFFERS received") nor can I ping the router assigning an static IP to my interface ("Destination Host Unreachable").
I have all the required firmware installed and the networking itself is working fine, as I can get an IP and access the Internet from other computers here.
I thought my motherboard could be damaged and I replaced it today by another one, but the result was the same.
I also tried a Debian Live USB stick, to check if the problem was in my Debian setup, but the result was the same.
Can anyone provide any pointers? I'm really out of ideas now.
Thanks,

M.
Last edited by mdutra on 2021-08-02 18:53, edited 2 times in total.

User avatar
stevepusser
Posts: 12930
Joined: 2009-10-06 05:53
Has thanked: 41 times
Been thanked: 71 times

Re: Ethernet interface seems "deaf"

#2 Post by stevepusser »

A good idea is to try another distro's Live USB, such as MX's or Mint's, just to see if they can get it to work.

It never hurts to tell us as much as you can about the troublesome hardware, too. One good way is to install inxi and tell us the results of "inxi -Nxxz"
MX Linux packager and developer

mdutra
Posts: 7
Joined: 2021-07-26 18:54

Re: Ethernet interface seems "deaf"

#3 Post by mdutra »

Hey Steve!

Thank you for your reply. It was really appreciated.

I tried the latest MX version and got "The network connection has been disconnected" displayed four times, then it gave up. The same happened with Debian Live.

Here's inxi's output for my system:

Code: Select all

Network:   Device-1: NVIDIA MCP61 Ethernet vendor: ASRock 939NF6G-VSTA Board type: network bridge driver: forcedeth v: kernel 
           port: d080 bus ID: 00:07.0 chip ID: 10de:03ef 
           Device-2: Realtek RTL8192EE PCIe Wireless Network Adapter driver: rtl8192ee v: kernel port: e800 bus ID: 03:00.0 
           chip ID: 10ec:818b 
Do you think it could be a driver problem? How could I diagnose it?

Thanks!

mdutra
Posts: 7
Joined: 2021-07-26 18:54

Re: Ethernet interface seems "deaf"

#4 Post by mdutra »

Let me illustrate what is going on. Maybe it will help someone help me:

If I configure the interface to work with DHCP on /etc/network/interfaces:

Code: Select all

auto enp0s7
iface enp0s7 inet dhcp
... and then I try to bring it up, this is what happens:

Code: Select all

mdutra@fx8:~$ sudo ifup enp0s7
Internet Systems Consortium DHCP Client 4.4.1
Copyright 2004-2018 Internet Systems Consortium.
All rights reserved.
For info, please visit https://www.isc.org/software/dhcp/

Listening on LPF/enp0s7/bc:5f:f4:d8:c0:ab
Sending on   LPF/enp0s7/bc:5f:f4:d8:c0:ab
Sending on   Socket/fallback
DHCPDISCOVER on enp0s7 to 255.255.255.255 port 67 interval 7
DHCPDISCOVER on enp0s7 to 255.255.255.255 port 67 interval 9
DHCPDISCOVER on enp0s7 to 255.255.255.255 port 67 interval 19
DHCPDISCOVER on enp0s7 to 255.255.255.255 port 67 interval 17
DHCPDISCOVER on enp0s7 to 255.255.255.255 port 67 interval 9
No DHCPOFFERS received.
No working leases in persistent database - sleeping.
On the other hand, if I set an static IP putting the following on /etc/network/interface:

Code: Select all

iface enp0s7 inet static
    address 192.168.0.10/24
    broadcast 192.168.0.255
    network 192.168.0.0
    gateway 192.168.0.1 
... I can bring the interface up:

Code: Select all

mdutra@fx8:~$ ip addr
(...)
2: enp0s7: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether bc:5f:f4:d8:c0:ab brd ff:ff:ff:ff:ff:ff
    inet 192.168.0.10/24 brd 192.168.0.255 scope global enp0s7
       valid_lft forever preferred_lft forever
... but I cannot ping the router (which is at 192.168.0.1):

Code: Select all

mdutra@fx8:~$ ping -c 4 192.168.0.1
PING 192.168.0.1 (192.168.0.1) 56(84) bytes of data.
From 192.168.0.10 icmp_seq=1 Destination Host Unreachable
From 192.168.0.10 icmp_seq=2 Destination Host Unreachable
From 192.168.0.10 icmp_seq=3 Destination Host Unreachable
From 192.168.0.10 icmp_seq=4 Destination Host Unreachable

--- 192.168.0.1 ping statistics ---
4 packets transmitted, 0 received, +4 errors, 100% packet loss, time 30ms
pipe 3
I also don't see any errors in the logs (I keep monitoring with "tail -f /var/log/...").

Any suggestions would be highly appreciated.

Miguel.

User avatar
bw123
Posts: 4015
Joined: 2011-05-09 06:02
Has thanked: 1 time
Been thanked: 28 times

Re: Ethernet interface seems "deaf"

#5 Post by bw123 »

https://askubuntu.com/questions/611781/ ... -14-04-lts

I had a minute to search for "debian MCP61 Ethernet" on the interwebz. Found the above link I ddn't follow to the bug but it's a known issue apparently...

Have you tried bring iface down, then
# modprobe -r forcedeth
# modprobe forcedeth msi=0 msix=0

if it works, then create
/etc/modprobe.d/forcedeth.conf
with contents
options forcedeth msi=0 msix=0

good luck,
bw
resigned by AI ChatGPT

mdutra
Posts: 7
Joined: 2021-07-26 18:54

Re: Ethernet interface seems "deaf"

#6 Post by mdutra »

Hey bw, thank you for your suggestion!

For some reason, my kernel crashed every time I tried to remove the module, but I went ahead and created the forcedeth.conf file to give it a try. No observable effect after rebooting the system.

Curiously, I had searched for "MCP61" and what I found is that the forcedeth driver should work fine with my hardware.

As an additional piece of information, I was tinkering with ethtool and some things caught my attention:

Code: Select all

mdutra@fx8:~$ sudo ethtool enp0s7
Settings for enp0s7:
	Supported ports: [ MII ]
	Supported link modes:   10baseT/Half 10baseT/Full 
	                        100baseT/Half 100baseT/Full 
	Supported pause frame use: No
	Supports auto-negotiation: Yes
	Supported FEC modes: Not reported
	Advertised link modes:  10baseT/Half 10baseT/Full 
	                        100baseT/Half 100baseT/Full 
	Advertised pause frame use: No
	Advertised auto-negotiation: Yes
	Advertised FEC modes: Not reported
	Speed: 100Mb/s
	Duplex: Full
	Port: MII
	PHYAD: 1
	Transceiver: internal
	Auto-negotiation: on
	Supports Wake-on: g
	Wake-on: d
	Link detected: yes
mdutra@fx8:~$ sudo ethtool -t enp0s7
The test result is FAIL
The test extra info:
link      (online/offline)	 0
register  (offline)       	 0
interrupt (offline)       	 1
loopback  (offline)       	 0
1. Why the "Supported ports" only shows "MII"? Shouldn't it include TP?
2. I know some interfaces don't support tests, but is it possible some interrupt conflict is indicated here? How could I further investigate it?

Also, I tried monitoring packet statics and interface traffic using ethtool and tcpdump:

Code: Select all

mdutra@fx8:~$ sudo ethtool -S enp0s7
NIC statistics:
     tx_bytes: 35573
     tx_zero_rexmt: 260
     tx_one_rexmt: 0
     tx_many_rexmt: 0
     tx_late_collision: 0
     tx_fifo_errors: 0
     tx_carrier_errors: 0
     tx_excess_deferral: 0
     tx_retry_error: 0
     rx_frame_error: 0
     rx_extra_byte: 0
     rx_late_collision: 0
     rx_runt: 0
     rx_frame_too_long: 0
     rx_over_errors: 0
     rx_crc_errors: 0
     rx_frame_align_error: 0
     rx_length_error: 0
     rx_unicast: 78
     rx_multicast: 376
     rx_broadcast: 75
     rx_packets: 529
     rx_errors_total: 0
     tx_errors_total: 0
     tx_deferral: 0
     tx_packets: 260
     rx_bytes: 136145
     tx_pause: 0
     rx_pause: 0
     rx_drop_frame: 0
Apparently the interface is sending and receiving bytes but... Where to? Take a look at the ARP requests from tcpdump... They seem to get never responded:

Code: Select all

mdutra@fx8:~$ sudo tcpdump -i enp0s7
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on enp0s7, link-type EN10MB (Ethernet), capture size 262144 bytes
00:56:34.216163 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 158
00:56:34.216183 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 149
00:56:34.216193 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 201
00:56:34.216202 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 213
00:56:34.216212 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 215
00:56:34.216222 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 229
00:56:34.216231 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 158
00:56:34.217726 ARP, Request who-has 192.168.0.10 tell 192.168.0.10, length 28
00:56:34.217805 ARP, Request who-has 192.168.0.1 tell 192.168.0.10, length 28
00:56:34.223888 IP6 :: > ff02::16: HBH ICMP6, multicast listener report v2, 1 group record(s), length 28
00:56:34.227895 IP 192.168.0.10 > 224.0.0.22: igmp v3 report, 1 group record(s)
00:56:34.316331 IP 192.168.0.10.mdns > 224.0.0.251.mdns: 0 [2q] PTR (QM)? _ipps._tcp.local. PTR (QM)? _ipp._tcp.local. (45)
00:56:34.395965 IP6 :: > ff02::1:ffd8:c0ab: ICMP6, neighbor solicitation, who has fe80::be5f:f4ff:fed8:c0ab, length 32
00:56:34.404892 IP 192.168.0.10.mdns > 224.0.0.251.mdns: 0 [4q] [6n] ANY (QM)? FX8._device-info._tcp.local. ANY (QM)? fx8.local. ANY (QM)? 10.0.168.192.in-addr.arpa. ANY (QM)? FX8._smb._tcp.local. (211)
00:56:34.415931 IP6 :: > ff02::16: HBH ICMP6, multicast listener report v2, 1 group record(s), length 28
00:56:34.416441 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 286
00:56:34.416467 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 277
00:56:34.416482 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 329
00:56:34.416502 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 341
00:56:34.416519 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 343
00:56:34.416535 IP 192.168.0.10.38948 > 239.255.255.250.1900: UDP, length 357
00:56:34.430430 IP 192.168.0.10.mdns > 224.0.0.251.mdns: 0*- [0q] 4/0/0 PTR _device-info._tcp.local., PTR FX8._smb._tcp.local., PTR _smb._tcp.local., PTR FX8._device-info._tcp.local. (127)
00:56:34.655417 IP 192.168.0.10.mdns > 224.0.0.251.mdns: 0 [4q] [6n] ANY (QM)? FX8._device-info._tcp.local. ANY (QM)? fx8.local. ANY (QM)? 10.0.168.192.in-addr.arpa. ANY (QM)? FX8._smb._tcp.local. (211)
00:56:34.906319 IP 192.168.0.10.mdns > 224.0.0.251.mdns: 0 [4q] [6n] ANY (QM)? FX8._device-info._tcp.local. ANY (QM)? fx8.local. ANY (QM)? 10.0.168.192.in-addr.arpa. ANY (QM)? FX8._smb._tcp.local. (211)
00:56:34.995972 IP 192.168.0.10 > 224.0.0.22: igmp v3 report, 1 group record(s)
00:56:35.107277 IP 192.168.0.10.mdns > 224.0.0.251.mdns: 0*- [0q] 6/0/0 (Cache flush) TXT "model=MacSamba", (Cache flush) A 192.168.0.10, (Cache flush) PTR fx8.local., (Cache flush) SRV fx8.local.:445 0 0, (Cache flush) TXT "", (Cache flush) SRV fx8.local.:0 0 0 (187)
00:56:35.219894 ARP, Request who-has 192.168.0.1 tell 192.168.0.10, length 28
00:56:35.316905 IP 192.168.0.10.mdns > 224.0.0.251.mdns: 0 [2q] PTR (QM)? _ipps._tcp.local. PTR (QM)? _ipp._tcp.local. (45)
I'm really lost. I'm also reviewing the Debian Administrator's Handbook, but found no tips there to help me with this.

User avatar
bw123
Posts: 4015
Joined: 2011-05-09 06:02
Has thanked: 1 time
Been thanked: 28 times

Re: Ethernet interface seems "deaf"

#7 Post by bw123 »

https://ubuntuforums.org/showthread.php ... st12566248

Another old link. You didn't say if you've already tried these old methods, that's where I'd start.

Have you booted to single user mode and used rmmod instead? Seems like ASrock claimed an issue with CPU. modprobe wants to "intelligently" remove and add modules, so that might be why the hang. You can't remove something from a chain of modules sometimes.

Try rebuild initramfs after editing around files in modprobe.d

Also, have you checked out and experimented with the other parameters for focedeth kernel driver?

# modinfo forcedeth

If it was me, I'd slap a working nic in there and experiment in spare time... some hardware takes years to figure out.

What about blacklisting forcedeth, then loading it after boot?
resigned by AI ChatGPT

mdutra
Posts: 7
Joined: 2021-07-26 18:54

Re: Ethernet interface seems "deaf"

#8 Post by mdutra »

I have finally found the solution!

The problem is described in details here: https://bugs.mageia.org/show_bug.cgi?id=26154

I arrived at that blog because my error log contained several lines with:

Code: Select all

forcedeth 0000:00:07.0 enp0s7: Got tx_timeout. irq status: 00000032
To solve the problem I edited the file /etc/default/grub, appending pci=nomsi to the line beginning with GRUB_CMDLINE_LINUX_DEFAULT=

Code: Select all

GRUB_CMDLINE_LINUX_DEFAULT="quiet pci=nomsi"
Then, I ran:

Code: Select all

sudo update-grub
And now ethtool reports no test errors:

Code: Select all

mdutra@fx8:~$ sudo ethtool -t enp0s7
The test result is PASS
The test extra info:
link      (online/offline)	 0
register  (offline)       	 0
interrupt (offline)       	 0
loopback  (offline)       	 0
That's it! I'm so happy!

Post Reply