On 03/12/2025 10:46, Chaz Kettleson wrote:
> On Tue, Dec 02, 2025 at 06:09:09AM -0500, Chaz Kettleson wrote:
>> On Tue, Dec 02, 2025 at 05:14:58PM +1000, David Gwynne wrote:
>>> aggr(4) is a better option for LACP for several reasons:
>>>
>>> - in my experience aggr is better at following the standard
>>>
>>> it doesn't get into weird states that blackhole traffic at the worst
>>> possible times.
>>>
>>> - aggr is mpsafe
>>>
>>> the last time i looked at trying to make the lacp code mpsafe in
>>> trunk(4) i ended up writing aggr(4). this lack of mpsafety is now
>>> making it hard to unlock more of the network stack.
>>>
>>> - aggr is faster than trunk
>>>
>>> looking at the benchmarking we did when i first introduced aggr,
>>> it looks like using trunk(4) more than halved the pps you can
>>> forward through a box, while aggr was more like a 10 to 15% hit.
>>> switching from trunk to aggr in our environment doubled our firewall
>>> performance.
>>>
>>> apart from these implementation differences, there's a few operational
>>> ones too that sthen@ has pointed out:
>>>
>>> - aggr uses a random MAC by default, trunk uses the MAC of the first-
>>>    added child port.
>>>
>>> - aggr does not bring the interface up automatically, you must use "up".
>>>
>>> anyone still using trunk(4) for LACP links will need to do at least the
>>> following:
>>>
>>> # mv hostname.trunk0 hostname.aggr0
>>> # echo up >> hostname.aggr0
>>>
>>> if you want a predictable MAC address on an aggr interface, you'll need
>>> to specify the lladdr before bringing it up. for example, my configs
>>> looks like this:
>>>
>>> lladdr fe:e1:ba:d0:e8:43
>>> trunkport ixl0
>>> trunkport ixl1
>>> up
>>>
>>> a bunch of vlan interfaces sit on top of this, which is why it doesnt
>>> have address config of its own.
>>>
>>> an example with ips:
>>>
>>> lladdr fe:e1:ba:d0:7c:ff
>>> trunkproto lacp
>>> trunkport ix0
>>> trunkport ix1
>>> inet 192.0.2.216 255.255.255.0
>>> inet6 2001:db8:230f:bc24:d3d8:7b58:f17c:f96e 64
>>> up
>>>
>> Hello,
>>
>> Do you know of any cases where aggr is _not_ a complete replacement for
>> trunk? A few years ago I setup a Dell PowerConnect 5224 in LACP and was
>> unable to get aggr to work, however, the following does work for me:
>>
>> cat /etc/hostname.trunk0
>> trunkproto lacp
>> trunkport bnx0
>> trunkport bnx1
>> trunkport bge0
>> up
>>
>> I can give it a shot again tonight to see if I missed something when I
>> initially tried aggr.
>>
>> -- 
>> Chaz
>>
> Hello,
>
> Good news, everything appears to work.
>
> 1. mv /etc/hostname.trunk0 /etc/hostname.aggr0
> 2. removed trunkproto lacp (no longer needed)
> 3. updated all hostname.vlan* with parent aggr0
>
> Not sure if something changed since 6.9/7.0 to now, or if I messed
> something up when I originally tried -- but everything is working as
> expected now!
>
> Thanks for the motivation to change this.

that's a relief. thanks for revisiting this. enjoy the extra pps.