Juniper MX80 + Cisco ASR1002 LACP Ether-channel

Good day! The task was set to make an Ether-channel of 4 gigabit ports between 2 pieces of iron from different manufacturers, namely Juniper MX80 and Cisco ASR1002. If it worked without problems between Cisco Catalyst and ASR, then you had to tinker with the Juniper.


It all starts simple. On a tsisk we create the Port-channel interface. The required ip addresses and parameters are set. Next, we enter on the necessary interfaces On this router there are no balancing settings, but the outgoing ASR traffic seems to be balancing normally. on katalist included load-balance on src-dst-mac. With Tsiska finished, let's move on to the juniper. set device-count by the number of aggregated interfaces, in this case there will be one interface. on the necessary interfaces we enter create an aggregated interface Link rises pings go. I could not find any docks for balancing on the juniper, and when I submitted traffic to the channel, I found that traffic was balanced from the tsiska to the jumper, and everything went from one jumper to the tsiska in one channel.
interface Port-channel 1




channel-group 1 mode active





chassis {
aggregated-devices {
ethernet {
device-count 1;
}
}
network-services all-ethernet;
}





ge-1/0/0 {
gigether-options {
802.3ad ae0;
}
}


ae0 {
aggregated-ether-options {
minimum-links 1;
link-speed 1g;
lacp {
active;
periodic fast;
}
}
unit 0 {
family inet {
address 10.100.0.1/30;
}
}
}





Digging documentation did not lead to anything. I was familiar with Juniper for the first time and by studying the configuration of the piece of iron by clicking "?" I came across balancing in the polishing section (for some reason nobody wrote about it in the manuals anywhere) Thus, traffic from the juniper to the tsiska began to be shared on the interfaces. I hope someone will be useful, because for me it was not the most trivial. Sorry for my Russian first experience writing articles.

policy-options {
policy-statement balance {
then {
load-balance per-packet;
}
}
}