11:34:53 IN the PCAP, 1 of the LACP states is out of sync, not sure why 12:18:29 Also.... took a few pcaps from a working aggregate and noticed on a few my servers that the actor key is the same 1001... i didnt pcap the other aggregate so maybe we start at 1000 and increment per server? 12:54:03 Fun.... the Partner Key is different for the 2 interfaces that are making my aggregate 12:54:09 I believe that is a *problem* 13:58:42 @Smithx10 is one partner keys (the "bad" one) that increment-by-one value? Or is it some junk value? 13:58:48 (DEBUG kernels may help you here.) 15:21:04 [illumos-gate] 14735 Users with a large group sidlist cause performance problems in SMB -- Matt Barden 16:35:47 danmcd: they were cross connected of course 16:45:11 Doh! 19:46:54 errr 19:47:32 and you were cross as a result no doubt 19:47:33 :) 19:48:22 Evidently, tho we have this situation https://gist.github.com/Smithx10/459c5bd563229ed019db522157d45386 19:48:58 when I did a dladm remove-aggr for the interface that mac address we used as the lacp system id ... the aggregation changed systemids to the only remaining member 19:50:23 The ToR really didn't like this.... the "vpc consistency-parameters" got angry because the local value didnt match the per value after I added it back in 19:50:49 I'm wondering what behavior is correct 19:51:07 Should the Cisco switch side just accept the new systemid being advertised by the LACP PDUs ? 19:51:49 Strangely it allows the i40e3 interface to continue to function lol 19:53:23 When creating a port channel, LACP does its due diligence and compares a few values. The first is the System ID. This is a unique value on each switch. It is a 64-bit value which can be manually assigned or automatically generated. Auto assignment comes from the system priority and the System MAC address. The second value is an identifier for the LAG, such as the port-channel ID. These four values (two at each end) create a LAGID (Link 19:53:24 Aggregation Group Identifier). All ports in a port-channel must have the same LAGID. If there is a mismatch, the ports are excluded or the LAG stays down, depending on the implementation. 19:54:25 Because our system ID changed out from under the switch, it keeps i40e1 "hibernated" or suspended 20:04:52 lol, aggregates are gross 20:05:15 long live the aggregate 21:07:56 people who like reading about system calls will find a bunch of updates to the section 2 man pages for Solaris in: https://gist.github.com/alanc/b4659167c33c40e159c16b9d048ea2ed 21:08:18 people who like fixing typos in illumos man pages will find a todo list in: https://pastebin.com/U8b9qXpW 21:14:13 oh wow