Hi @Christophe7983 

Thanks for posting in our business forum.

Config of your other three devices? Screenshots.

If you set it as passive, it is pending to receive the LACPDU handshake from the peer. That's important to form the LACP.

And remove the VLAN before you config LACP. As for now, I gotta need to make sure that LACP works with your devices. For the rest of the L2 setup, that'll be yours.

Hi @Christophe7983 

Thanks for posting in our business forum.

Christophe7983 wrote

  @Christophe7983 

 

update: got the LACP working on three out of four ports on the Synology. Forgot to remove LAG from the default LAN. Once it was exclusive to VLAN10 I worked.

 

No succes on the Truenas Scale device alas.

 

Could it have something to do with the hash algorithms not matching up?

On TP-link side:

Hash Algorithm

Select the Hash Algorithm, based on which the switch can choose the port to forward the received packets. In this way, different data flows are forwarded on different physical links to implement load balancing. There are six options: SRC MAC: The computation is based on the source MAC addresses of the packets. DST MAC: The computation is based on the destination MAC addresses of the packets. SRC MAC+DST MAC: The computation is based on the source and destination MAC addresses of the packets. SRC IP: The computation is based on the source IP addresses of the packets. DST IP: The computation is based on the destination IP addresses of the packets. SRC IP+DST IP: The computation is based on the source and destination IP addresses of the packets.

 

 

On TrueNas Scale side:

LACP

Select LACP to use the most common protocol for LAGG interfaces based on IEEE specification 802.3ad

In LACP mode, the interfaces negotiate with the network switch to form a group of ports that are all active once. The network switch must support LACP for this option to function.

Select the Transmit Hash Policy option from the dropdown list. LAYER2+3 is the default selection. Others: LAYER2, LAYER3+4.

Select the LACPDU Rate to use.

• Select SLOW (default) to set the heartbeat request to every second and the timeout to a three-consecutive heartbeat loss that is three seconds.
• Select FAST to set the timeout rate at one per second even after synchronization. Using FAST allows for rapid detection of faults.

Select the interfaces to use in the aggregation.

 

I'm kinda stumbling in the dark. Started thinking about all the possible combinations 6 (on tp-link) x 3 LAYER combos x 2 LACPDU rates = 36 attemps if I don't lose count.

VLAN is not considered then. OK.

I don't think TrueNAS is making this clear. Or it is simply saying that L2 matches with another TrueNAS while LAG-ing? That's not clear and I cannot give you the answer.

LACPDU rate does not matter. Fast or Slow is just the rate of transmitting the packet. It is in Active mode so you have this option. Passive mode just sit and wait for the active to send the packet. We just need to focus on the hash policy because it determines the value and eventually can affect the link.

My suggestion is to ask TrueNAS about its Layer 2. Layer 2 is the MAC address. That for sure. But how does it generate the hash? That'll be the question for you to ask the TrueNAS team. Is it using SRC and DST MAC to generate the value? Or SRC? Or DST? Layer 3 does not matter unless you are going to use Layer 3 to generate the hash. Now we are talking about the L2. So, it has 3 combos * 2(switches). Not 36 attempts.

Hi @Christophe7983

Christophe7983 wrote

  @Clive_A 

 

That seems logical. Ok I'll ask TrueNas. Surely others must have encountered this before with a TP-link jetstream switch. But no posts to be found. I'l start a new thread.

 

Thank you for you help, Clive. Happy Easter

 

gts, Christophe

Happy Easter.

How does it go? Let me know if you get a reply from them. Let's see if we can resolve this problem.