Arch Linux

cypher_zero

Member

Registered: 2014-10-23

Posts: 50

Networking randomly stops working; only rebooting resolves

For the past few weeks now, my computer will randomly stop working on all network connections.  Everything appears to be configured correctly and nothing has been changed from a configuration standpoint, but I will suddenly be unable to pass network traffic.  I'm kind of at a loss as to what to do at this point.  Any suggestions or help would be greatly appreciated. 

Some more info:
I have two network adapters, a physical ethernet NIC that's built into my motherboard and a USB wireless dongle.  Both are necessary as they connect to separate networks; the ethernet connects to my internal work network, and the wireless connects to an external internet connection.  The ethernet connection has a static IP address that I've configured though netctl (/etc/netctl/enp10s0-static). I've got a script set up to change the routing table on the computer to prefer the wireless for anything not on my internal network as NetworkManager overrides the route settings for some reason.  I've configured my routing like such:

default via 172.16.1.1 dev wlp11s0u1u2  proto static  metric 600 
default via 10.47.112.2 dev enp10s0  proto static  metric 700 
10.0.0.0/8 via 10.47.112.2 dev enp10s0  proto static  metric 500 
10.47.112.0/21 dev enp10s0  proto kernel  scope link  src 10.47.119.52  metric 100 
172.16.1.0/24 dev wlp11s0u1u2  proto kernel  scope link  src 172.16.1.112  metric 600

Here's what my /etc/netctl/enp10s0-static file looks like:

Description='Ethernet static 10.47.119.52'
Interface=enp10s0
Connection=ethernet
IP=static
Address=('10.47.119.52/21')
Routes=('10.0.0.0/8 metric 500 via 10.47.112.2' 'default metric 700 via 10.47.112.2')
#Gateway='192.168.1.1'
DNS=('10.47.4.210' 10.47.4.211')

When the problem occurs, neither connection will be working.  Restarting NetworkManager doesn't resolve, neither does disabling and renabling the adapter (ip link set dev enp10s0 down && ip link set dev enp10s0 up), physically unplugging and replugging in the cable/adapter, etc. 

Again, any help, advise or suggestions would be greatly appreciated.

ewaller

Administrator

From: Pasadena, CA

Registered: 2009-07-13

Posts: 19,785

Re: Networking randomly stops working; only rebooting resolves

Have you ensured that there are no other network related services enabled in systemd besides NetworkManager?   Not netctl, not dhcpcd, not wpa_supplicant,  nothing...

---

Nothing is too wonderful to be true, if it be consistent with the laws of nature -- Michael Faraday
Sometimes it is the people no one can imagine anything of who do the things no one can imagine. -- Alan Turing
---
How to Ask Questions the Smart Way

cypher_zero

Member

Registered: 2014-10-23

Posts: 50

Re: Networking randomly stops working; only rebooting resolves

Looks like I had netctl enabled.  Disabled that... I'll see how it goes after I reboot

cypher_zero

Member

Registered: 2014-10-23

Posts: 50

Re: Networking randomly stops working; only rebooting resolves

No luck.  I even went so far as to completely remove netctl, rebooted, and the same issue persists.  To be clear, when this happens both my wired and wireless go out.  I can only assume it's something to do with the network stack, but I have no idea how to go about troubleshooting that.

cypher_zero

Member

Registered: 2014-10-23

Posts: 50

Re: Networking randomly stops working; only rebooting resolves

Still hoping to find a resolution to this as it's driving me up a wall now.  It happened again yesterday.  Something else I noticed when it happened though was that my USB 3.0 powered up stopped working.  It was still getting power and I could charge my phone off of it, but it wasn't passing any data to the computer.  I'm inclined to think it's related in part because I moved the WiFi dongle off of that hub and onto a different USB port and was able to use WiFi (sorta).

cypher_zero

Member

Registered: 2014-10-23

Posts: 50

Re: Networking randomly stops working; only rebooting resolves

Issue repeated itself again sometime between 16:30 and 17:15 while I was away from my desk. I've captured a log from that time here if anyone's willing to help me sort through it as nothing is jumping out at me (or maybe I just don't know how to read it): https://www.dropbox.com/s/clhwh24t4r4sb … l.txt?dl=0

adamrights

Member

From: New York

Registered: 2011-08-27

Posts: 5

Website

Re: Networking randomly stops working; only rebooting resolves

Well,  The messages from NetworkManager:(wlan0): error getting associated BSSID: No such device littered over the entire log is likely contributing to whatever is causing the wifi dongle drivers to crash -- after that resetting NM won't help. What's your NM config look like? It appears udev renames the interface to wlp11s0u1u2 -- for some reason NM is still trying to use wlan0 for something.

Your wireless drivers kernel crash; that's where everything goes down:

Mar 03 17:15:46 cypherPC kernel: r8712u 5-1.2:1.0 wlp11s0u1u2: r8712u: pipe error: (-71)
Mar 03 17:15:46 cypherPC kernel: xhci_hcd 0000:0b:00.0: WARN Cannot submit Set TR Deq Ptr
Mar 03 17:15:46 cypherPC kernel: xhci_hcd 0000:0b:00.0: A Set TR Deq Ptr command is pending.
Mar 03 17:15:46 cypherPC kernel: r8712u 5-1.2:1.0 wlp11s0u1u2: r8712u: pipe error: (-71)
Mar 03 17:15:46 cypherPC kernel: AMD-Vi: Event logged [IO_PAGE_FAULT device=0b:00.0 domain=0x0011 address=0x000000000671c200 flags=0x0000]
Mar 03 17:15:46 cypherPC kernel: AMD-Vi: Event logged [IO_PAGE_FAULT device=0b:00.0 domain=0x0011 address=0x000000000671c280 flags=0x0000]
Mar 03 17:15:46 cypherPC kernel: AMD-Vi: Event logged [IO_PAGE_FAULT device=0b:00.0 domain=0x0011 address=0x000000000671c240 flags=0x0000]
Mar 03 17:15:46 cypherPC kernel: AMD-Vi: Event logged [IO_PAGE_FAULT device=0b:00.0 domain=0x0011 address=0x000000000671c2c0 flags=0x0000]
Mar 03 17:15:50 cypherPC NetworkManager[9574]: <warn>  (wlan0): error getting associated BSSID: No such device
Mar 03 17:15:51 cypherPC kernel: xhci_hcd 0000:0b:00.0: xHCI host not responding to stop endpoint command.
Mar 03 17:15:51 cypherPC kernel: xhci_hcd 0000:0b:00.0: Assuming host is dying, halting host.
Mar 03 17:15:51 cypherPC kernel: r8712u 5-1.2:1.0 wlp11s0u1u2: r8712u: pipe error: (-108)
Mar 03 17:15:51 cypherPC kernel: r8712u 5-1.2:1.0 wlp11s0u1u2: r8712u: pipe error: (-108)
Mar 03 17:15:51 cypherPC kernel: r8712u 5-1.2:1.0 wlp11s0u1u2: r8712u: pipe error: (-108)
Mar 03 17:15:51 cypherPC kernel: r8712u 5-1.2:1.0 wlp11s0u1u2: r8712u: pipe error: (-108)
Mar 03 17:15:51 cypherPC kernel: xhci_hcd 0000:0b:00.0: HC died; cleaning up
Mar 03 17:15:51 cypherPC kernel: usb 5-1: USB disconnect, device number 2
Mar 03 17:15:51 cypherPC kernel: usb 5-1.1: USB disconnect, device number 4
Mar 03 17:15:51 cypherPC kernel: usb 5-1.1.1: USB disconnect, device number 5
Mar 03 17:15:51 cypherPC kernel: r8712u 5-1.2:1.0 wlp11s0u1u2: xmit_bh => bDriverStopped or bSurpriseRemoved
Mar 03 17:15:51 cypherPC kernel: usb 6-1: USB disconnect, device number 2
Mar 03 17:15:51 cypherPC kernel: usb 6-1.1: USB disconnect, device number 3
Mar 03 17:15:51 cypherPC kernel: usb 5-1.2: USB disconnect, device number 7
Mar 03 17:15:52 cypherPC wpa_supplicant[832]: wlp11s0u1u2: CTRL-EVENT-DISCONNECTED bssid=3c:15:c2:f2:f5:20 reason=3 locally_generated=1
Mar 03 17:15:52 cypherPC vmnet-natd[1864]: RTM_NEWLINK: name:wlp11s0u1u2 index:9 flags:0x00001002
Mar 03 17:15:52 cypherPC vmnetBridge[1799]: RTM_NEWLINK: name:wlp11s0u1u2 index:9 flags:0x00001002
Mar 03 17:15:52 cypherPC vmnetBridge[1799]: Removing interface wlp11s0u1u2 index:9
Mar 03 17:15:52 cypherPC dhclient[28392]: receive_packet failed on wlp11s0u1u2: Network is down
Mar 03 17:15:52 cypherPC NetworkManager[9574]: <info>  NetworkManager state is now CONNECTED_LOCAL
Mar 03 17:15:52 cypherPC NetworkManager[9574]: <info>  NetworkManager state is now CONNECTED_GLOBAL
Mar 03 17:15:52 cypherPC NetworkManager[9574]: <info>  Policy set 'enp10s0' (enp10s0) as default for IPv4 routing and DNS.
Mar 03 17:15:52 cypherPC kernel: bridge-wlp11s0u1u2: disabling the bridge
Mar 03 17:15:52 cypherPC NetworkManager[9574]: <info>  Writing DNS information to /usr/bin/resolvconf

Next time it goes down try removing and readding the drivers -- then bring the interface back up, etc:

 sudo modprobe -r r8712u 

 sudo modprobe r8712u 

NetworkManager config files could help.
The tail of dmesg on crash will probably show the drivers fail you see above.

cypher_zero

Member

Registered: 2014-10-23

Posts: 50

Re: Networking randomly stops working; only rebooting resolves

Thanks!  I'll give that a shot. 

Which NetworkManager config files would you want to see?  I haven't made any changes to /etc/NetworkManager/NetworkManager.conf and the only other config files I'm seeing are for my actual connections. 

Here's /etc/NetworkManager/NetworkManager.conf:

[main]
plugins=keyfile
dhcp=dhclient
dns=default

## Set static hostname
#[keyfile]
#hostname=foobar

## HTTP-based connectivity check
#[connectivity]
#uri=http://nmcheck.gnome.org/check_network_status.txt
#interval=100

cypher_zero

Member

Registered: 2014-10-23

Posts: 50

Re: Networking randomly stops working; only rebooting resolves

No dice.

cypher_zero

Member

Registered: 2014-10-23

Posts: 50

Re: Networking randomly stops working; only rebooting resolves

Well, I'm not sure if it's 100% fixed, but I haven't had the issue since updating to the latest kernel.  Here's hoping that takes care of the problem.