Pppoe passthrough is now broken on Open BSD router

[doverosx]

I’m in Barrhaven on Bell Fibe and it’s been great, using pppoe passthrough (OpenBSD router) with no issues for a while now.

This weekend, I’ve been getting disconnects, crazy high pings and more. Anyone else have problems?

I’ll be *trying* a direct connection to the modem in a bit.

 Using just the modem seems to have resolved the issue so Bell broke something nicely.

[alexricher]

I could be wrong, but I feel this is issue is caused by the new firmware, which introduces PPPoE connectivity issues, as I had no problems before the holidays and haven't changed anything in my setup. Ok, I did add a Ubiquiti switch, but that's way further down the line in my own LAN.

Onto Reddit, a Bell tech said 2.13 was coming and there was nothing one could do to avoid it, it was just a matter of time. And I agree with you, there should be a way to reach to advanced tech whenever those issues occur. I'm not networking genius, but I feel I can handle well my own setup... and I feel this issue occurs between the modem and the router, but most likely the modem.

I wouldn't mind reaching out to Dom, but I believe DSLForums are down since a while, no? Is he onto this forum? Trying to tag him in case he might have an idea... (Oh, there is a Bell_Dom on here! 😀)

@Bell_Dom: Hi Dom! Long time, no talk, I hope you're having a great week. I wanted to have your opinion. Are you aware of any new issues that could have been introduced in recent firmware updates relating to PPPoE? I used to do PPPoE Passthrough without any problem by using DMZ/ADMZ, but since about a week, it stopped working. Now, I'm unable to get my own router working with the modem since a few days. I'm trying to fix this by Monday as I do WFH. Is there a new version that fixes it? Is there a way to revert back to the older firmware to see if this resolves the issue? Thanks so much for your time!

[dks]

Thank you for the traceroute. This suggests that the issue is beyond your modem and likely within Bell's network. The second IP is within Bell's network in Ottawa. This suggests your issue may well be network congestion or equipment related. It is likely not your local setup. 

[dks]

It is doubtful customers will connect with Bell techs directly now. The world at Bell has changed. 

[dks]

Thank you for your question. I have answered in the other thread. This is unlikely related to the firmware upgrade to 2.13, which has been in release for several months. However, the other customer appears, at least by traceroute, to have issues upstream of their Giga Hub. 

There has been one concern about connections by someone with 60+ devices connected to their LAN. In that case, so many connected devices overloaded the Giga Hub. Bell is only responsible up to the Giga Hub. Your LAN is your issue. 

[JB1]

I spoke with Bell and they said the problem is on my end and explicitly said they won't help with anything that isn't their equipment.  If I wanted to I could spend my own money to work with a Bell tech expert (or similar sounding name) -- which sounds Bell adjacent.  I would consider if I knew that they would be able to resolve the PPPOE issue though I am concerned that this would be outside their capability.

My issues kicked off on Jan 8 and progressively got worse until Friday morning @ 2:45 am and have not worked since then.  Gigahub is sending out a network but that's not sufficient enough for our network.

[dks]

That reflects Bell’s policy. On the customer side of the modem, it is a  customer issue. 

[doverosx]

That’s the fun part. The issue *is* their equipment.

[doverosx]

It isn’t customer side though. The proof is obvious in a trace route. 

Well. I shall start the paper trail, escalate to the ombudsman and crtc if it gets to that point.

[Vanadiel]

I don't think they have changed anything on the PPPOE side as mine works flawlessly, but it could be an issue with how they handle ADMZ on the new firmware.

You can try looking at that, as I believe there are a few options in the modem config to manually setup related to ADMZ.

It's for sure not the PPPOE connection itself as otherwise mine would be experiencing the same issue.

[doverosx]

I’m not using ADMZ. Btw the issue is intermittent, making it that much more fun. Again, the pppoe connection is operating just fine, it’s on the WAN side that you see things break down which points to bad routes or maybe some network stack issue on the gigahub. 

I’m not sure what PPPOE device you’re using either, but *BSD systems actually adhere to IETF/RFC standards. The GH violates a lot of networking standards.

[JB1]

I know Unifi pushed out a new update early in January.  So that could be causing the issue on my side but it's unclear.  If it was on Bells side - knowing that other people have their PPPOE running alright is some helpful intel.  Thanks!  If I find out anything Ill let you know.    No ADMZ being used on this end -- prefer not to as well.  

[doverosx]

I just had a thought. What MTU do you get after pppoe negotiation. The PADU is supposed to send the appropriate MTU belt handshake completion, but that might have broken order of operations for standard compliant hardware and the GH.

[JB1]

It was set at 1500 not sure what it was after the PPPOE handshake. I did ssh in and change the value to 1492 which didn't resolve anything unfortunately.

[Vanadiel]

Should be 1492 for MTU value.

[JB1]

Thanks!  Didn't resolve unfortunately.