Monday, August 08, 2022

Re: Making pppoe more resiliant to outages?

On Mon, 8 Aug 2022 11:27:42 -0000 (UTC)
Stuart Henderson <stu.lists@spacehopper.org> wrote:

> > related ISPs) have the same problem with connectivity. *BUT* I need to
> > appease L1 tech support, some of whom think a serial port is where
> > sailors go for breakfast.
>
> Do you have an ISP-supplied router you can use temporarily?
> That's often a better bet when trying to get them to fix something
> which you strongly suspect is their fault..removes anything "weird"
> from the equation.

Sadly not… my service was a "bring your own router" service. I was
using an old Advantech industrial PC (AMD Geode LX500) as my router at
the time, and the service was using ADSL2 then.

https://www.mail-archive.com/misc@openbsd.org/msg161146.html

That got replaced by the APU3 I have now. I still have that industrial
PC though, so I could coax OpenBSD 7.1 onto it and see how it fares.

> > Have there been any significant changes in the PPPoE driver between 7.0
> > and 7.1?
>
> Not really between 7.0 and 7.1. Session time calculation changed from
> UTC time to uptime and there's handling for fetching nameserver addresses.
>
> But there were some locking-related changes that went in before 7.0;
> -current has taken steps back from these.

Yeah, I did have a look at the changelog later and got the same
impression, although I'm not that familiar with how CVS does tagging
and branching, so following the exact history of what happened was less
clear to me.

I do intend to update to OpenBSD 7.1 eventually, but when
troubleshooting, one normally wants to avoid changing more than one or
two variables at a time. :-)
--
Stuart Longland (aka Redhatter, VK4MSL)

I haven't lost my mind...
...it's backed up on a tape somewhere.

No comments:

Post a Comment