Do ipv6 announce messages loop back?

I am getting firewall block messages every minute in the kernel log of a Kubuntu machine Ox1

[UFW BLOCK] IN=eth0 OUT= MAC= SRC=fe80:0000:0000:0000:82fa:5bff:fe31:fa27 DST=ff12:0000:0000:0000:0000:0000:0000:8384 LEN=118 TC=0 HOPLIMIT=1 FLOWLBL=464139 PROTO=UDP SPT=37113 DPT=21027 LEN=78

which seems to be local announce messages being sent from Ox1 ethernet adaptor. Do local announce messages loop back?

Also I am getting messages:

[UFW BLOCK] IN=wlan0 OUT= MAC= SRC=fe80:0000:0000:0000:4685:00ff:fe44:3f72 DST=ff12:0000:0000:0000:0000:0000:0000:8384 LEN=118 TC=0 HOPLIMIT=1 FLOWLBL=237879 PROTO=UDP SPT=37113 DPT=21027 LEN=78

which seems to be local announce messages being sent from Ox1 wlan adaptor however the wlan adaptor is not currently connected. Why is syncthing generating messages from an unconnected adaptor?

It’s sending stuff to all adapters regardless it seems.

Yes but the mystery to me here is why a message from this same computer

> SRC=fe80:0000:0000:0000:82fa:5bff:fe31:fa27

is coming in on eth0

> IN=eth0

to itself (ie from Ox1 to Ox1)

And also why messages appear to be coming in on a disconnected interface:


again to itself from Ox1 to Ox1.

The only way I can think of this happening is if Ox1 sent messages on the loopback interface,but then it should show:

IN = lo

(fe80:0000:0000:0000:82fa:5bff:fe31:fa27 is the ipv6 address of this same machine, Ox1)

Any ideas?

It depends on a bunch of things, but it’s not unusual for IPv6 multicasts to be received again on the interface they were sent, yes.

Thanks. Surprising. Maybe the router is relaying them.

I guess I need to take a year off to understand ipv6. :confused:

It may, but it’s probably internal to the computer. One rationale is that multicast is often used for various kinds of discovery, and one program on the computer (like a word processor) may be interested on what another (a printer driver) has to announce. I think you can run two Syncthing instances and have local discovery between them work, as well.

You must be correct that it is internal to the computer - the 2nd case I mentioned where a packet was coming in on a disconnected interface (wlan0) shows that.

Thanks for you insight.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.