<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div>Hello Owen,</div><div><br></div><div>While I agree ( and cudos to Job for noticing the issue while the document is still at the draft stage), the current process for allocation is not developer friendly.</div><div><br></div><div><div><span style="background-color: rgba(255, 255, 255, 0);">For ExaBGP, </span><span style="background-color: rgba(255, 255, 255, 0);">I also had to squat a code point to implement </span><a href="http://tools.ietf.org/html/draft-frs-bgp-operational-message-00" style="background-color: rgba(255, 255, 255, 0);">draft-frs-bgp-operational-message</a><span style="background-color: rgba(255, 255, 255, 0);">.</span></div><div><span style="background-color: rgba(255, 255, 255, 0);">I doubt it will eve cause an operational issue, ExaBGP is not used to route packets in anyone's core, but but the current process gave me no choice: it takes implementation to find issues with drafts and/or interrop problems, unexpected behaviours or simply provide a feature to a crying customer even if a draft is never created / never reach RFC status.</span></div></div><div><br></div><div>I remember reading a draft from John which attempted to allocate some code points for experimentation - my memory is fuzzy on the details sorry - but even then this would require re numbering which is not ideal.</div><div><br></div><div>So perhaps in addition to recognising the squatting issue, "we" should look at how the current IETF process works and can be changed to improve experimentation.</div><div><br></div><div>Thomas</div><div><br></div><div><div>Sent from my iPad</div></div><div><br> On 27 Oct 2016, at 16:47, Owen DeLong <<a href="mailto:owen@delong.com">owen@delong.com</a>> wrote:<br><br></div><blockquote type="cite"><div><span>I don’t mind the move to 32, but I hope the vendors are getting appropriately smacked for squatting and that those attributes are not allowed to be misappropriated by the vendors.</span><br><span></span><br><span>We have a standards process for a reason and vendors simply squatting on numbers is a violation of that process which cannot be allowed to stand unless we wish to establish that as precedent and simply allow vendors to claim numbers as they wish.</span><br><span></span><br><span>This already happened with the BSD community in their implementation of a pseudo-VRRP like capability and now two different vendors have abused BGP path attributes.</span><br><span></span><br><span>This is not a good path for us to continue.</span><br><span></span><br><span>Owen</span><br><span></span><br><blockquote type="cite"><span>On Oct 26, 2016, at 11:19 PM, Job Snijders <<a href="mailto:job@ntt.net">job@ntt.net</a>> wrote:</span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>Dear Internet,</span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>Through this beacon it was discovered that a vendor was squatting on BGP</span><br></blockquote><blockquote type="cite"><span>Path Attribute value 30. And another vendor sat on 31.</span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>So, a twisted turn of events, the Large BGP Communities effort has ended up</span><br></blockquote><blockquote type="cite"><span>with BGP Path Attribute value 32 - very befitting if you look at the very</span><br></blockquote><blockquote type="cite"><span>problem we're trying to solve :-)</span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>The beacon has been updated to use the new IANA assigned value, nothing</span><br></blockquote><blockquote type="cite"><span>else was changed. Hopefully we are in the clear this time around!</span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>Please verify if you can see 192.147.168.0/24 and 2001:67c:208c::/48</span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>Kind regards,</span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>Job</span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>On Tue, Oct 11, 2016 at 05:01:56PM +0200, Job Snijders wrote:</span><br></blockquote><blockquote type="cite"><blockquote type="cite"><span>Large BGP Communities are a novel way to signal information between</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>networks. An example of a Large BGP Communities is: 2914:4056024901:80.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>Large BGP Communities are composed of three 4-octet integers, separated</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>by something like a colon. This is easy to remember and accommodates</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>advanced routing policies in relation to 4-Byte ASNs. It is the tool that has</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>been missing since 4-octet ASNs were introduced.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>IANA has made an Early Allocation of the value 30 (LARGE_COMMUNITY) in</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>the "BGP Path Attributes" registry under the "Border Gateway Protocol</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>(BGP) Parameters" group.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>The draft can be read here: <a href="https://tools.ietf.org/html/draft-ietf-idr-large-community">https://tools.ietf.org/html/draft-ietf-idr-large-community</a></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>Additional information about Large BGP Communities can be found here:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span><a href="http://largebgpcommunities.net/">http://largebgpcommunities.net/</a></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>Starting today (2016.10.11), the following two BGP beacons are available</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>to the general public, with AS_PATH 2914_15562$</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   Both these prefixes have a Large BGP Community attached:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   2001:67c:208c::/48</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   192.147.168.0/24</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   Large BGP Community - 15562:1:1</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>The NLNOG RING BGP Looking Glass is running the latest version of BIRD</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>which understands the Large BGP Community Path Attribute.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>IPv4 LG: <a href="http://lg.ring.nlnog.net/prefix_detail/lg01/ipv4?q=192.147.168.0/24">http://lg.ring.nlnog.net/prefix_detail/lg01/ipv4?q=192.147.168.0/24</a></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>IPv6 LG: <a href="http://lg.ring.nlnog.net/prefix_detail/lg01/ipv6?q=2001:67c:208c::/48">http://lg.ring.nlnog.net/prefix_detail/lg01/ipv6?q=2001:67c:208c::/48</a></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>In theory, since this is an optional transitive BGP Path Attribute, all</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>the Looking Glass' peers should boomerang the Large Community back to</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>the LG.  However we currently observe that 50 out of 75 peers propagate</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>the Large BGP Community to the LG.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>Relevant Router commands to see if you receive the attribute, or whether</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>one of intermediate networks has stripped the attribute from the route:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   IOS: show ip bgp path-attribute unknown </span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       shows all prefixes with unknown path attributes.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>    IOS #2 - like on route views:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>        route-views>sh ip bgp 192.147.168.0</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>         BGP routing table entry for 192.147.168.0/24, version 98399100</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>         Paths: (39 available, best #30, table default)</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>           Not advertised to any peer</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>           Refresh Epoch 1</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>           701 2914 15562</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>             137.39.3.55 from 137.39.3.55 (137.39.3.55)</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>               Origin IGP, localpref 100, valid, external</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>               unknown transitive attribute: flag 0xE0 type 0x1E length 0xC</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>                 value 0000 3CCA 0000 0001 0000 0001</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>               rx pathid: 0, tx pathid: 0</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>         </span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   IOS-XR: (you must look at specific prefixes)</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       RP/0/RSP0/CPU0:Router#show bgp  ipv6 unicast 2001:67c:208c::/48 unknown-attributes </span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       BGP routing table entry for 2001:67c:208c::/48</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       Community: 2914:370 2914:1206 2914:2203 2914:3200</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       Unknown attributes have size 15</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       Raw value:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       e0 1e 0c 00 00 3c ca 00 00 00 01 00 00 00 01 </span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   JunOS:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       user@JunOS-re6> show route 2001:67c:208c::/48 detail </span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       2001:67c:208c::/48 (1 entry, 1 announced)</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>           AS path: 15562 I</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>           Unrecognized Attributes: 15 bytes</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>           Attr flags e0 code 1e: 00 00 3c ca 00 00 00 01 00 00 00 01</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>A note about router Configurations:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>Ensure you are not fitlering the path attributes, eg:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>JunOS:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   [edit protocols bgp]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   user@junos# delete drop-path-attributes 30</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>XR:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   configure</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   router bgp YourASN</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       attribute-filter group ReallyBadIdea ! avoid creating bogons</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>       no attribute 30 </span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>     !</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>   !</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>Contact persons: myself or Jared Mauch or NTT NOC. BGP Session</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>identifier 83.231.213.230 / 2001:728:0:5000::a92 AS 15562.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>Kind regards,</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span>Job</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><span></span><br><span></span><br></div></blockquote></body></html>