kernel: br0 / eth2 messages

[pgershon]

For the last three days, my router (GT-AX6000) has been logging the following messages periodically - maybe 9 times in a row, then stopping for a few hours, then posting one or two more times, then stopping and starting seemingly randomly. Nothing is being unplugged.

Jul 17 18:17:22 kernel: eth2 (Int switch port: 1) (Logical Port: 1) (phyId: 2) Link DOWN.
Jul 17 18:17:22 kernel: br0: port 2(eth2) entered disabled state
Jul 17 18:17:27 kernel: eth2 (Int switch port: 1) (Logical Port: 1) (phyId: 2) Link Up at 1000 mbps full duplex
Jul 17 18:17:27 kernel: br0: port 2(eth2) entered blocking state
Jul 17 18:17:27 kernel: br0: port 2(eth2) entered listening state
Jul 17 18:17:29 kernel: br0: port 2(eth2) entered learning state
Jul 17 18:17:31 kernel: br0: port 2(eth2) entered forwarding state
Jul 17 18:17:31 kernel: br0: topology change detected, propagating

This set of messages has followed 2 of the times I got the kernel message - may or may not be related.

Jul 17 20:52:36 hostapd: eth7: STA 5c:3e:1b:d4:c1:1d IEEE 802.11: associated
Jul 17 20:52:36 hostapd: eth7: STA 5c:3e:1b:d4:c1:1d RADIUS: starting accounting session F45C3A536E223063
Jul 17 20:52:36 hostapd: eth7: STA 5c:3e:1b:d4:c1:1d WPA: pairwise key handshake completed (RSN)
Jul 17 20:52:54 hostapd: eth7: STA 5c:3e:1b:d4:c1:1d IEEE 802.11: disassociated
Jul 17 20:52:54 wlceventd: wlceventd_proc_event(494): eth7: Deauth_ind 5C:3E:1B4:C1:1D, status: 0, reason: Disassociated due to inactivity (4), rssi:-91

I do not k now where the RADIUS instruction eight be coming from.

Any ideas what either or both of these message groups might indicate?

 

[ColinTaylor]

They're all normal messages. The first set relate to the ethernet device connected to port 2. What is that device? It could be a PC that's in sleep mode.

The second set are normal Wi-Fi messages. This particular Apple device appears to have a very weak signal.

 

[pgershon]

Second set of messages is my iPhone. I assume the weak signal because it is connecting to another AP that is closer. But why the RADIUS message? I never got that before.

The first set of messages is more interesting to me. Port 2 has a MoCa adaptor that connects over coax to another MoCa adaptor attached to a RT-AX92U serving as an AP. It should not go to sleep or turn off - I believe it should stay active all the time. My read of the message is that the MoCa / AP connection is going down for a split second then resuming and reconnecting. But it happens with frequency about 5-10 times per day at random intervals. Should I be concerned and is there anything to do? This was a run in my walls that was very hard to retrofit with Cat-5e and I assumed I could repurpose the unused coax wires.

FWIW, there is a similar set of entries in the log of the RT-AX92U at matching times, so it looks like the MoCa connections goes down for a split second a few time a day (or at least that is my interpretation):

Jul 19 10:37:30 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link DOWN.
Jul 19 10:37:30 kernel: br0: port 1(eth0) entered disabled state
Jul 19 10:37:34 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link UP at 1000 mbps full duplex
Jul 19 10:37:34 kernel: br0: port 1(eth0) entered forwarding state
Jul 19 10:37:34 kernel: br0: port 1(eth0) entered forwarding state

 

[ColinTaylor]

I couldn't say whether that's normal for MoCa because we don't use them over here. Did you move the adapter to another port, because it's saying port 3 now? Bear in mind that what the router says is "port 2" may not be the same as what's printed on the case.

 

[pgershon]

The adaptor has not been touched. It is at eth2/br2 (port 2) on the router and eth0/br0 on the Access Point. My two entries above were different - one from router and the other from AP. Here it is again:

ROUTER:
Jul 19 10:37:31 kernel: eth2 (Int switch port: 1) (Logical Port: 1) (phyId: 2) Link DOWN.
Jul 19 10:37:31 kernel: br0: port 2(eth2) entered disabled state
Jul 19 10:37:34 kernel: eth2 (Int switch port: 1) (Logical Port: 1) (phyId: 2) Link Up at 1000 mbps full duplex
Jul 19 10:37:34 kernel: br0: port 2(eth2) entered blocking state
Jul 19 10:37:34 kernel: br0: port 2(eth2) entered listening state
Jul 19 10:37:36 kernel: br0: port 2(eth2) entered learning state
Jul 19 10:37:38 kernel: br0: port 2(eth2) entered forwarding state
Jul 19 10:37:38 kernel: br0: topology change detected, propagating

AP:
Jul 19 10:37:30 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link DOWN.
Jul 19 10:37:30 kernel: br0: port 1(eth0) entered disabled state
Jul 19 10:37:34 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link UP at 1000 mbps full duplex
Jul 19 10:37:34 kernel: br0: port 1(eth0) entered forwarding state
Jul 19 10:37:34 kernel: br0: port 1(eth0) entered forwarding state

Link was down on both for 3 seconds from 10:37:31-10:37:34.

I found some others on the internet discussing similar issues. But I dont see these setting on my Hitron HT-EM4s. I sent an email to Hitron asking.

 

[pgershon]

Company got back to me but do not have an answer. From reading other internet threads, sounds like I might try disabling EEE, but I dont know how to do that with ASUS.

 

[ColinTaylor]

Are you using Merlin's firmware? If you are you could install the ethtool package and check the status of EEE. I suspect it's already disabled.

# ethtool --show-eee eth1
EEE Settings for eth1:
        EEE status: disabled
        Tx LPI: disabled
        Supported EEE link modes:  Not reported
        Advertised EEE link modes:  Not reported
        Link partner advertised EEE link modes:  Not reported

 

[drinkingbird]

The adaptor has not been touched. It is at eth2/br2 (port 2) on the router and eth0/br0 on the Access Point. My two entries above were different - one from router and the other from AP. Here it is again:

ROUTER:
Jul 19 10:37:31 kernel: eth2 (Int switch port: 1) (Logical Port: 1) (phyId: 2) Link DOWN.
Jul 19 10:37:31 kernel: br0: port 2(eth2) entered disabled state
Jul 19 10:37:34 kernel: eth2 (Int switch port: 1) (Logical Port: 1) (phyId: 2) Link Up at 1000 mbps full duplex
Jul 19 10:37:34 kernel: br0: port 2(eth2) entered blocking state
Jul 19 10:37:34 kernel: br0: port 2(eth2) entered listening state
Jul 19 10:37:36 kernel: br0: port 2(eth2) entered learning state
Jul 19 10:37:38 kernel: br0: port 2(eth2) entered forwarding state
Jul 19 10:37:38 kernel: br0: topology change detected, propagating

AP:
Jul 19 10:37:30 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link DOWN.
Jul 19 10:37:30 kernel: br0: port 1(eth0) entered disabled state
Jul 19 10:37:34 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link UP at 1000 mbps full duplex
Jul 19 10:37:34 kernel: br0: port 1(eth0) entered forwarding state
Jul 19 10:37:34 kernel: br0: port 1(eth0) entered forwarding state

Link was down on both for 3 seconds from 10:37:31-10:37:34.

I found some others on the internet discussing similar issues. But I dont see these setting on my Hitron HT-EM4s. I sent an email to Hitron asking.

ETH2, BR2, and Port 2 are not all the same thing.

Sounds like your MOCA link is having some trouble. I'd troubleshoot that, many adapters intentionally will take the ethernet side down when the coax side goes down.

But first you need to test and map what physical ports map to what eth ports by plugging and unplugging a cable to a known good device. ETH2 is not necessarily port 2, and they will all fall under BR0.

 

[pgershon]

I strongly suspect that my router has EEE on, as that is a known problem for my MOCA (and known to cause exactly this behavior). I am trying to figure out the easiest way to install ethtool as a means to see the state of EEE and hopefully disable.

 

[pgershon]

ETH2, BR2, and Port 2 are not all the same thing.

I mistyped - eth2/br0 (port 2) on the router. The MOCA plus into port 2. eth0/br0 on the AP - MOCA connects to the WAN port and all other ports are empty.

 

[drinkingbird]

I mistyped - eth2/br0 (port 2) on the router. The MOCA plus into port 2. eth0/br0 on the AP - MOCA connects to the WAN port and all other ports are empty.

If that is the link that keeps flapping then your MOCA link is probably noisy or poor. Focus on that. One of the adapters could be failing too but I'd lean toward bad connection given all the possible variables.

 

[drinkingbird]

I strongly suspect that my router has EEE on, as that is a known problem for my MOCA (and known to cause exactly this behavior). I am trying to figure out the easiest way to install ethtool as a means to see the state of EEE and hopefully disable.

It seems unlikely, I've rarely seen EEE and haven't heard anyone have any issues with it on Asus routers and MOCA. Possible, but doubtful (especially if the link is always active and passing traffic).

 

[pgershon]

What is the best way to determine EEE status? Ethtool was suggested but I have not figured out how to install it.

And what else might you suggest I look at? The MOCA coax connection is a straight run from the router to the basement, a female-female F-type joiner, and a straight coax from basement to AP. No other splits and nothing else on the line. The MOCA light on the adaptor is not a solid blue however - it's very close but it has an ultrafast flicker to it - I wonder if that could be an indicator. Not sure what I could do to fix however if one of the coax cables is bad (or even how to test)?

 

[Ripshod]

Over ssh

opkg install ethtool

EEE is generally off anyway.

 

[drinkingbird]

What is the best way to determine EEE status? Ethtool was suggested but I have not figured out how to install it.

And what else might you suggest I look at? The MOCA coax connection is a straight run from the router to the basement, a female-female F-type joiner, and a straight coax from basement to AP. No other splits and nothing else on the line. The MOCA light on the adaptor is not a solid blue however - it's very close but it has an ultrafast flicker to it - I wonder if that could be an indicator. Not sure what I could do to fix however if one of the coax cables is bad (or even how to test)?

Sounds like the MOCA adapter has a bad power supply or is totally shot. Are both adapters doing it or just one?

Is the coax RG6 or RG58/59?

If it's the same MOCA adapter on both ends try swapping the power supplies. If the flicker moves along with the power supply it is dead, no longer converting AC to DC properly which will cause flickering LEDs and other issues.

 

[bluepoint]

I mistyped - eth2/br0 (port 2) on the router. The MOCA plus into port 2. eth0/br0 on the AP - MOCA connects to the WAN port and all other ports are empty.

Can you elaborate more how you setup your MOCA network? I'm not sure why you have the MOCA bridge connected to the WAN port it's suppose to be in your LAN network.

 

[pgershon]

Can you elaborate more how you setup your MOCA network? I'm not sure why you have the MOCA bridge connected to the WAN port it's suppose to be in your LAN network.

As I understand things, the MOCA bridge acts in lieu of a cat-5e wire connection. Therefore it connects from a LAN port on the router (in my case port 2) to the WAN port of the access point. The access point is connected to the main network via the WAN port, no? I googled it just now to confirm I am not losing it. How else would you get full functionality out of the access point?

 

[drinkingbird]

As I understand things, the MOCA bridge acts in lieu of a cat-5e wire connection. Therefore it connects from a LAN port on the router (in my case port 2) to the WAN port of the access point. The access point is connected to the main network via the WAN port, no? I googled it just now to confirm I am not losing it. How else would you get full functionality out of the access point?

You have it connected correctly. Have you tested if it is a power issue?

 

[pgershon]

I swapped out the second set of MOCA connectors I have for the ones I had in place. I am going to wait 24-36 hours and then look at my logs to see if it made a difference. The blue light flickers the same way with the new MOCA adaptors and power supply as with the old, Not sure if that is relevant. It close to a steady blue - I would have said steady except the blue light next to it is purely on.

 

[drinkingbird]

I swapped out the second set of MOCA connectors I have for the ones I had in place. I am going to wait 24-36 hours and then look at my logs to see if it made a difference. The blue light flickers the same way with the new MOCA adaptors and power supply as with the old, Not sure if that is relevant. It close to a steady blue - I would have said steady except the blue light next to it is purely on.

What kind of coax and how long is the run?