Mvneta ... ethernet eth1: bad rx status

Software SW bugs discussion
1

Hi!

I get a lot of the following kernel messages (1000/hour):

2017-10-24T21:25:31+02:00 err kernel[]: [496506.899704] mvneta f1034000.ethernet eth1: bad rx status 0fa10000 (crc error), size=1437
2017-10-24T21:25:37+02:00 err kernel[]: [496510.438878] mvneta f1034000.ethernet eth1: bad rx status 0fa10000 (crc error), size=1398
2017-10-24T21:25:38+02:00 err kernel[]: [496485.750253] mvneta f1034000.ethernet eth1: bad rx status 0f810000 (crc error), size=938
2017-10-24T21:25:41+02:00 err kernel[]: [494054.516096] mvneta f1034000.ethernet eth1: bad rx status 4f810000 (crc error), size=219
2017-10-24T21:25:43+02:00 err kernel[]: [496338.853626] mvneta f1034000.ethernet eth1: bad rx status 0f810000 (crc error), size=1520

As you can see the size varies and is smaller than MTU (=1500), but sometimes larger than MTU.
I tried different MTU settings without success.
I changed cables => no change

Below I add a few more diagnostics (ifconfig, ethtool). Ifconfig shows only rx errors, but a lot of them.
Please I really I am out of ideas to what can cause this error. Any help/ideas are welcome!

Kind regards

Ulrich

# ifconfig eth1
eth1      Link encap:Ethernet  HWaddr D8:58:D7:00:34:1B  
          inet addr:217.211.243.192  Bcast:217.211.243.255  Mask:255.255.255.0
          inet6 addr: fe80::da58:d7ff:fe00:341b/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:134943244 errors:69846 dropped:0 overruns:0 frame:0
          TX packets:69896899 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:532 
          RX bytes:107824132224 (100.4 GiB)  TX bytes:141854998143 (132.1 GiB)
          Interrupt:38 

# ethtool eth1
Settings for eth1:
        Supported ports: [ TP MII ]
        Supported link modes:   10baseT/Half 10baseT/Full 
                                100baseT/Half 100baseT/Full 
                                1000baseT/Half 1000baseT/Full 
        Supported pause frame use: No
        Supports auto-negotiation: Yes
        Advertised link modes:  10baseT/Half 10baseT/Full 
                                100baseT/Half 100baseT/Full 
                                1000baseT/Half 1000baseT/Full 
        Advertised pause frame use: No
        Advertised auto-negotiation: Yes
        Link partner advertised link modes:  10baseT/Half 10baseT/Full 
                                             100baseT/Half 100baseT/Full 
        Link partner advertised pause frame use: No
        Link partner advertised auto-negotiation: Yes
        Speed: 100Mb/s
        Duplex: Full
        Port: MII
        PHYAD: 1
        Transceiver: external
        Auto-negotiation: on
        Link detected: yes
2

I know topic is old but I started getting this same messages as well on my 2nd Omnia configured AP only.

3

check with ethtool -S eth2 | grep crc , substitute the interface with the one that exhibits the error.

4

root@ap:~# ethtool -S eth2 | grep crc
bad_crc_event: 51028
Could it be cable problem?

I also get mac_receive_error: 6583

5

Not sure, just reckon rather not but you try and change the cable and see if it makes a difference.

Else it could be a hardware problem with the TO’s ethernet port or some bad frames coming down the line from the upstream device.

This looks like some serious trouble.

6

Just as general note to check more comprehensively for other bad | error counts - ethtool -S eth2 | grep 'bad\|error'

7

As I expected! I’ve replaced patch-cord cables with new ones and errors are definitely gone.

1 Like
8

I started having flood of messages about crc errors again. What do I do? Is it physically broken?

Some other user is having the same problems on Omnia running vanilla OpenWRT:

9

Hello,
can you please send the output of ip -s link show <name of problematic interface> and more details like the version of TOS?

10

It was something OSI physical level.

I disconnected all the RJ-45 plug and connected them a few times to scratch a bit the connection. Moved cables around rebooted Omnia and voila! No longer any errors. It was weird because it negotiated Full 1Gbps when there were errors.

I think the cause was just a bad connection somewhere:
root@ap:~# ip -s link show eth2
4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master br-lan state UP
mode DEFAULT group default qlen 532
link/ether d8:58:d7:00:3d:9f brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped overrun mcast
394805449 540555 0 0 0 0
TX: bytes packets errors dropped carrier collsns
1151369657 150644 0 0 0 0

root@ap:~# ethtool -S eth2 | grep ‘bad|error’
bad_octets_received: 0
bad_frames_received: 0
bad_fc_received: 0
mac_receive_error: 0
bad_crc_event: 0
eee_wakeup_errors: 0

I was getting mac_receive_error: over9k and bad_crc_event: also_alot. For now its fixed.

1 Like
11

No I am still getting errors. Its definetely better then before but still nothing more I can do with cables in the walls.

root@ap:~# ip -s link show eth2
4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master br-lan state UP
mode DEFAULT group default qlen 532
link/ether d8:58:d7:00:3d:9f brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped overrun mcast
521637175 670072 2 0 0 0
TX: bytes packets errors dropped carrier collsns
1160679343 195601 0 0 0 0

And two times in dmesg since reboot.

12

Hello,
I suggest to check the remote side, too. Maybe the remote device is dying?

13

That would be not good. Remote device is second Omnia:

root@router:~# ethtool -S lan4
NIC statistics:
in_rx_error: 8
in_fcs_error: 27
out_pause: 0
excessive: 0
collisions: 0
deferred: 0
out_fcs_error: 0

root@router:~# ip -s link show lan4
9: lan4@eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br-la
n state UP mode DEFAULT group default qlen 1000
link/ether d8:58:d7:00:20:ae brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped overrun mcast
1195600209 992798 0 0 0 0
TX: bytes packets errors dropped carrier collsns
671512493 773288 0 0 0 0

I am quite confident its cabling problem. Its a bit better then before but still I get packet loss between devices.

14

Okay, now we are getting somewhere. You can try to manually set both ports to link on 100Mbit/s speed and see the difference.

15

[66068.453007] mvneta f1034000.ethernet eth2: Link is Up - 100Mbps/Full - flow cont

I saved number of errors and with 100Mb they dont appear. But how come the link was gigabit capable and suddenly its not. It was running error free since I connected everything up.