Rob,
I just took a look at the tcpdump of packets leaving AMPRGW and they appear to be the usual full RIP table. I am not aware of any rate limiting at this end; of course, there could be some further downstream.
In a recent 5 minute cycle, the rip sender logged Apr 2 04:35:01 <local0.info> amprgw ripsend[64800]: 706 routes, 29 + 11/25 packets sent to 524 distinct gateways
and indeed, 30 packets were sent to one of the NL gateways, 82.94.240.92; 29 of them were length 532 containing 25 routes, and the last of length 252, containing 11.
Pings to 82.94.240.92 were a consistent 163 ms and there was 0% packet loss in 100 consecutive pings.
I am not aware of any changes to the rip sender nor to the OS, and all appears to be operating normally.
If there's anything else I can test from this end, let me know. - Brian
On Tue, Apr 02, 2019 at 01:19:27PM +0200, Rob Janssen wrote:
Since yesterday, or maybe earlier, we do not receive the full RIP table anymore and our route table for IPIP routes has almost emptied.
A trace of UDP port 520 shows only 2 RIP packets per cycle.
Has something changed? Or does this have to be some rate-limiting on the path to here?
Rob _________________________________________ 44Net mailing list 44Net@mailman.ampr.org https://mailman.ampr.org/mailman/listinfo/44net