![]() ![]() ![]() |
|
Lost again today, getting a bit old now
dfnt:
Lost again today, getting a bit old now
Same - lost IPv6 just after 9:30, haven't got it back yet (I'm leaving it for now rather than down/upping it).
Yep seeing the same also.
I've found that renewing via the following solves it for me:
release dhcpv6-pd interface pppoe0
delete dhcpv6-pd duid
renew dhcpv6-pd interface pppoe0
It is in-fact a bug with the Edgerouter. I've found PPPoE re-establishes causing IPv6 to drop. I believe this will be fixed in the next software release.
Michael Murphy | https://murfy.nz
Referral Links: Quic Broadband (use R122101E7CV7Q for free setup)
Are you happy with what you get from Geekzone? Please consider supporting us by subscribing.
Opinions are my own and not the views of my employer.
As a workaround for the meantime I have a 2 part script:
First - runs on a server on my local network - I call this ping.sh - it runs every 5min via Cron:
#!/bin/bash
ping6 -c 1 murfy.nz
if [ $? -eq 0 ]; then
echo "We're alive!"
else
echo "`date -u`: IPv6 dropped. Running resetv6.sh on router" >> ping.log
ssh router 'at now -f /config/scripts/restartv6.sh' >> ping.log
echo "-----" >> ping.log
fi
The second is located in /config/scripts/restartv6.sh on the Edgerouter:
#!/bin/vbash
echo "Resetting IPv6"
/opt/vyatta/bin/vyatta-op-cmd-wrapper release dhcpv6-pd interface pppoe0
/opt/vyatta/bin/vyatta-op-cmd-wrapper delete dhcpv6-pd duid
/opt/vyatta/bin/vyatta-op-cmd-wrapper renew dhcpv6-pd interface pppoe0
It is horrible hack. Also this requires you to have a SSH key on your Edgerouter. You can use the loadkey utility under "configure" in EdgeOS to do this.
Michael Murphy | https://murfy.nz
Referral Links: Quic Broadband (use R122101E7CV7Q for free setup)
Are you happy with what you get from Geekzone? Please consider supporting us by subscribing.
Opinions are my own and not the views of my employer.
It's back up for me this morning. I did nothing. This on a SonicWall, I did try and manual restablish/renew the DHCP lease etc. That didn't work. So I left it.
Definitely something going on upstream and independent of equipment.
So mines not working now. I've got IPv6 addresses being assigned to all devices on my network but I've got no connectivity out via IPv6; all servers and machines fail to connect. Nothing has changed on my config on the USG for months so something is happening at 2degrees end I'm sure.
Mine came back at around 10pm last night (without me doing anything). I've had to disable some remote checks to IPv6 services so I don't wake up to a ton of alerts, but otherwise it isn't impacting me very much as 99% of what I do is over IPv4. Would be nice to know what's happening and when it might get fixed though (still haven't heard anything from 2degrees about the ticket I raised a few weeks ago).
michaelmurfy:
Yep seeing the same also.
I've found that renewing via the following solves it for me:
release dhcpv6-pd interface pppoe0
delete dhcpv6-pd duid
renew dhcpv6-pd interface pppoe0
It is in-fact a bug with the Edgerouter. I've found PPPoE re-establishes causing IPv6 to drop. I believe this will be fixed in the next software release.
This is what I've been doing as well, as opposed to reconnecting the pppoe0 interface, cleaner as nothing goes down doing it
Benoire:
So mines not working now. I've got IPv6 addresses being assigned to all devices on my network but I've got no connectivity out via IPv6; all servers and machines fail to connect. Nothing has changed on my config on the USG for months so something is happening at 2degrees end I'm sure.
Yep seeing same behaviour on a Turris Omnia, OpenWRT based. Worked fine for months but playing up recently.
dfnt: Sent a a pcap to Nick anyway
The same thing happened in 2 beta 1
Yeah, not something we see as an issue on our side, sorry. When is the new code due?
https://nick.mackechnie.co.nz | NZ ISP latency monitoring - https://smokeping.thenet.gen.nz
I wonder if it's related to those weird PPPoE outages we see sometimes (on IPv4). Its when the PPPoE session appears to be up but nothing routes over the connection. I've seen it happen across a number of different brands of products (Fortinet, Sonicwall, Zyxel, etc). On the Sonicwall we've overcome it using a logical probe (ICMP or TCP) under WAN Load Balancing (except we only have 1 WAN connection), when the probe fails it forces a complete PPPoE disconnect reconnect. Unfortunately on the Sonicwall because we are not getting assigned an IP for WAN (or gateway) and only using LL I cannot enable this feature for v6.
Is anyone still using static v6?
Finally clicked at what I'd done wrong in my static configuration - bit of a derp moment. Anyway now I have static working again I'm able monitor my IPv6 WAN interface for logical failures.
|
![]() ![]() ![]() |