After full-upgrade Internet not accessable.

[rsteinmetz70112]

I ran a full_upgrade on my 16.04 machine and while the local network works,
I can login from a desktop using Putty, but I can't ssh from the machine.

I cannot access the Internet.


ping 8.8.8.8
fails

Network-manager says there is
"No valid active connections found!


# ifconfig -a
enp1s0 Link encap:Ethernet HWaddr 70:85:c2:68:f6:fe
inet addr:192.168.1.1 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::7285:c2ff:fe68:f6fe/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:37921 errors:0 dropped:10626 overruns:0 frame:0
TX packets:11134 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:19949614 (19.9 MB) TX bytes:2597973 (2.5 MB)
Interrupt:18

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:65536 Metric:1
RX packets:215 errors:0 dropped:0 overruns:0 frame:0
TX packets:215 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1
RX bytes:29058 (29.0 KB) TX bytes:29058 (29.0 KB)

It worked just fine Friday.

[TheFu]

That IP address seems like it should be for the router.

Did you check the router?
Ethernet cable?
Switch port/
routing table?

[rsteinmetz70112]

The router is 192.168.1.254 - Blame ATT


Did you check the router?
Router works for all other computers on the network.

Ethernet cable?
Swapped cables.

Switch port/
Swapped switches. I have a large switch I normally use but the router has a small switch in it.

routing table?
No

I'm beginning to suspect a hardware problem on the Motherboard.
I think I'll try a live session and see if that works.
Unfortunately that will have to wait until after Thanksgiving, I'll be traveling.

I do have remote access via ssh.

[TheFu]

No

I'm beginning to suspect a hardware problem on the Motherboard.
I think I'll try a live session and see if that works.
Unfortunately that will have to wait until after Thanksgiving, I'll be traveling.

I do have remote access via ssh.

I take it remote access is through a different machine on the LAN?

$ route -n?
does it look correct? I hate the ip r output. Too ugly.

Just for completeness - is this network-manager or netplan AND static or DHCP?

I had something similar with 16.04, but only inside a virtual machine and only for 1 virtual machine connected to the same bridge. There are about 8 other VMs on the same bridge. Never figured out the issue. It fixed itself within a few days without rebooting the switches, routers, or host machine. Rebooting the impacted VM didn't help. A few months later, the same issue happened on the same VM and again, a few days later it self-fixed. I moved that VM to a different VM host and haven't see the issue in over a year. There was nothing different about the problem VM. I use ansible to configure networking for all of them, so besides the IP, it was configured the same as the other VMs for networking.

In short - I've felt THIS pain too and I believe you.

[rsteinmetz70112]

I take it remote access is through a different machine on the LAN?[\quote]
Not exactly there is a machine on the lan which does address translation and directs the connection is directly to the machine in question. I can get in I just can't get out.

[QUOTE]$ route -n?
does it look correct? I hate the ip r output. Too ugly.


~# route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
0.0.0.0 192.168.1.254 0.0.0.0 UG 0 0 0 enp1s0
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 enp1s0


Just for completeness - is this network-manager or netplan AND static or DHCP?

It is static through /etc/network/interfaces


I had something similar with 16.04, but only inside a virtual machine and only for 1 virtual machine connected to the same bridge. There are about 8 other VMs on the same bridge. Never figured out the issue. It fixed itself within a few days without rebooting the switches, routers, or host machine. Rebooting the impacted VM didn't help. A few months later, the same issue happened on the same VM and again, a few days later it self-fixed. I moved that VM to a different VM host and haven't see the issue in over a year. There was nothing different about the problem VM. I use ansible to configure networking for all of them, so besides the IP, it was configured the same as the other VMs for networking.

In short - I've felt THIS pain too and I believe you.

I've since had another 19.04 machine do almost the same thing. I believe it was somehow caused by a routine upgrade. I have discovered that the resolver wasn't working quire right and added nameserver lines to /etc/resolv.conf. That fixed the problem. Now I just have to convince the machine not to rewire /etc/resolve.conf or at least put the right stuff in it.

[rsteinmetz70112]

I'm going to make this solved. I still havn't figured out how /etc/resolv.conf got changed. I suspect a routine upgrade.