Only just got got around to re-testing. Had to do some work on my config generator amongst other things
Seems it still doesn't like this which I believe is valid
RFC4360 is not very helpful about this, but from looking at the IANA assignments I believe that origin should be valid for Transitive Two-Octet AS-Specific, Transitive Four-Octet AS-Specific & Transitive IPv4-Address-Specific Extended Community Sub-Types so origin:<2 octet number>:<4 octet number> should be fine
Still throws error: 'H' format requires 0 <= number <= 65535
I think all my other real-world extended communities are getting through otherwise
That should definitively work
Could you please send me the line which is sending this
So I can fix it ..
(a copy and paste of the trace per mail will do)
I am in transit to London ( going to check our BETT stand tonight )
So I may be able to fix the issue in the next 30 minutes .. or it will be .. later ...
I think the bug is now squashed
Please use my thomas-mangin tree not pushed to Exa-Networks yet
nice ! gitter support vi search and replace correction on the last line entered :clap:
Thanks, just had something come so may have to delay testing until tomorrow. Will let you know
nope - still broken ...
hopefully fixit it before the trains arrives at kx
Had my lab setup in a frozen state for most of this week whilst I've been trying to reproduce something. So I've only just got around to looking at this again. I've run my config through the current thomas-mangin tree code and it's now accepting all my real-world-harvested extended-communities, I'm using the 'L' form rather than the origin4/target4 forms. Cheers for sorting this quickly =)
If you need anything I will be available tomorrow PM
And at the London office Monday :smile:
Hiya. Is anyone around to help debug a stacktrace on the latest ExaBGP release?