0

I currently use an embedded system with 4G (Usb0 interface) and two Ethernet interfaces.

The 4G pcie module works when the eth0 and eth1 interfaces are disabled but not when eth0 and eth1 are enabled (apart from a few times).

Here's what the route-n command gives me

Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
0.0.0.0         192.168.1.1     0.0.0.0         UG    0      0        0 usb0
10.0.1.0        0.0.0.0         255.255.255.0   U     0      0        0 eth0
169.254.0.0     0.0.0.0         255.255.0.0     U     1000   0        0 eth0
192.168.1.0     0.0.0.0         255.255.255.0   U     1      0        0 usb0
192.168.50.0    0.0.0.0         255.255.255.0   U     0      0        0 eth1

This is what the 4g (usb0) interface gives me with the tcpdump command when I do a ping

tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on usb0, link-type EN10MB (Ethernet), capture size 262144 bytes
00:14:28.161098 IP 192.168.1.100.49969 > 8.8.8.8.domain: 44229+ A? 2.ubuntu.pool.ntp.org. (39)
00:14:28.261402 IP 192.168.1.100.49969 > 8.8.8.8.domain: 36904+ AAAA? 2.ubuntu.pool.ntp.org. (39)
00:14:28.261586 IP 192.168.1.100.60466 > 8.8.8.8.domain: 7899+ PTR? 100.1.168.192.in-addr.arpa. (44)
00:14:28.262062 IP 192.168.1.1 > 192.168.1.100: ICMP net 8.8.8.8 unreachable, length 75
00:14:28.264825 IP 192.168.1.1 > 192.168.1.100: ICMP net 8.8.8.8 unreachable, length 75
00:14:28.264835 IP 192.168.1.1 > 192.168.1.100: ICMP net 8.8.8.8 unreachable, length 80
00:14:33.167600 IP 192.168.1.100.34691 > 8.8.4.4.domain: 44229+ A? 2.ubuntu.pool.ntp.org. (39)
00:15:08.227252 IP 192.168.1.100.49053 > 8.8.8.8.domain: 35401+ PTR? 1.1.168.192.in-addr.arpa. (42)
00:15:08.228507 IP 192.168.1.1 > 192.168.1.100: ICMP net 8.8.8.8 unreachable, length 78
00:15:11.409362 IP 192.168.1.100.45109 > 8.8.8.8.domain: 11625+ PTR? 79.1.0.10.in-addr.arpa. (40)
00:15:28.245636 IP 192.168.1.100.51858 > 8.8.8.8.domain: 55400+ PTR? 4.4.8.8.in-addr.arpa. (38)
00:15:28.624809 IP 192.168.1.100 > 8.8.8.8: ICMP echo request, id 4160, seq 15, length 64
00:15:29.625472 IP 192.168.1.100 > 8.8.8.8: ICMP echo request, id 4160, seq 16, length 64
00:15:48.626493 IP 192.168.1.100 > 8.8.8.8: ICMP echo request, id 4160, seq 35, length 64
00:15:48.627266 IP 192.168.1.1 > 192.168.1.100: ICMP net 8.8.8.8 unreachable, length 92
00:15:49.625565 IP 192.168.1.100 > 8.8.8.8: ICMP echo request, id 4160, seq 36, length 64
00:15:49.626481 IP 192.168.1.1 > 192.168.1.100: ICMP net 8.8.8.8 unreachable, length 92

Do you have any idea?

thank you

0

You omitted the ping command, but the dump shows attempts to resolve DNS via 8.8.8.8 (Google nameserver) which are answered by "not reachable". So you probably did a

ping 2.ubuntu.pool.ntp.org

Guess: Internet via the 4G provider works, but the ISP of the 4G network is blocking the google nameserver. But when you enable eth0 or eth1, you are getting DHCP information that contains 8.8.8.8 as nameserver.

Verification: When the "internet is working" via 4G, try a ping to a well-known numeric IP address, see that it works, look at the dump. Remember the address. Try the same numeric ping when the "internet is not working". Compare the dump. If it succeeds, then the problem is indeed the DNS server.

Possible solution: Configure whatever you are using to accept DHCP information (dhclient?) to ignore the nameservers given, but retain the nameserver you get from the 4G ISP.

  • Hello, thank you very much for your answer: But I notice something strange, I've rebooted the system 3 times since more troubles, but I'm connected on eth0 from my pc with putty on the system without router and ping to 8.8.8.8 working on the 4G without problems. – tdav Mar 8 '18 at 9:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.