This topic has been marked solved and closed to new posts due to inactivity. We hope you'll join the conversation by posting to an open topic or starting a new one.
My RN314 runs on the latest FW and has a dual Link to a Cisco Meraki Switch MS220 capable of L3, LACP, LAG, etc.! After creating the team on 802.3ad LACP Layer 3+4 in the RN314 it states ok but after the reboot of the RN314 it changes back to 802..3ad LACP layer 2 every time again?!
In the meantime, I'm running on FW level 6.9.3 and the setting is stable but still, there is no data received over one of the two connections within the bond. One is constantly receiving 4Mbps data from 4 separate devices in total so it should be possible to use both ports with setting L3+L4, correct?
The L3+L4 setting only applies when sending. The switch LACP setting is what matters in the NAS receive direction. Most switches are hard-wired to use layer-2 hashing.
The odds of receiving streams from 4 devices on the same link is 1/8 - not that probable, but not shocking either. Using L3+L4 doesn't change those odds - what it does is potentially allow 2 streams going to the same device to use different links.
You might consider using a static LAG on the switch, and set up the NAS to use round-robin.
After creating the team on 802.3ad LACP Layer 3+4 in the RN314 it states ok but after the reboot of the RN314 it changes back to 802..3ad LACP layer 2 every time again?!
This does sound like a bug, but I am wondering how much it actually impacts your performance. Even with a layer 3+4 hash, you still have a 50-50 chance that flows using different layer 4 ports will end up sharing the same NIC.
Do you have control over the hash in the Cisco switch (I don't with my switches).
I have the latest OS running 6.9.0 and I have limited functionality on my Meraki MS220 Switches, mostly is done by web interface. The interface mentions that it had a trunk running on two ports.
People have reported issues with DHCP as well, and random disconnects.
I really wonder how this release ever made it though testing......
// T
[Sun Oct 29 21:54:49 2017] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[Sun Oct 29 21:54:49 2017] bond0: link status definitely up for interface eth0, 1000 Mbps full duplex
[Sun Oct 29 21:54:50 2017] e1000e: eth1 NIC Link is Down
[Sun Oct 29 21:54:50 2017] bond0: link status definitely down for interface eth1, disabling it
[Sun Oct 29 21:54:50 2017] bond0: first active interface up!
[Sun Oct 29 21:54:53 2017] e1000e: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[Sun Oct 29 21:54:53 2017] bond0: link status definitely up for interface eth1, 1000 Mbps full duplex
[Sun Oct 29 21:55:12 2017] e1000e: eth0 NIC Link is Down
[Sun Oct 29 21:55:12 2017] e1000e 0000:01:00.0 eth0: speed changed to 0 for port eth0
[Sun Oct 29 21:55:12 2017] bond0: link status definitely down for interface eth0, disabling it
[Sun Oct 29 21:55:12 2017] bond0: first active interface up!
[Sun Oct 29 21:55:13 2017] e1000e: eth1 NIC Link is Down
[Sun Oct 29 21:55:13 2017] e1000e 0000:04:00.0 eth1: speed changed to 0 for port eth1
[Sun Oct 29 21:55:13 2017] bond0: link status definitely down for interface eth1, disabling it
[Sun Oct 29 21:55:13 2017] bond0: first active interface up!
[Sun Oct 29 21:55:15 2017] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[Sun Oct 29 21:55:15 2017] bond0: link status definitely up for interface eth0, 1000 Mbps full duplex
[Sun Oct 29 21:55:15 2017] bond0: first active interface up!
[Sun Oct 29 21:55:16 2017] e1000e: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[Sun Oct 29 21:55:16 2017] bond0: link status definitely up for interface eth1, 1000 Mbps full duplex
[Sun Oct 29 21:56:07 2017] e1000e: eth0 NIC Link is Down
[Sun Oct 29 21:56:07 2017] e1000e 0000:01:00.0 eth0: speed changed to 0 for port eth0
[Sun Oct 29 21:56:07 2017] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[Sun Oct 29 21:56:20 2017] e1000e: eth0 NIC Link is Down
[Sun Oct 29 21:56:20 2017] e1000e 0000:01:00.0 eth0: speed changed to 0 for port eth0
[Sun Oct 29 21:56:20 2017] bond0: link status definitely down for interface eth0, disabling it
[Sun Oct 29 21:56:20 2017] bond0: first active interface up!
[Sun Oct 29 21:56:23 2017] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[Sun Oct 29 21:56:23 2017] bond0: link status definitely up for interface eth0, 1000 Mbps full duplex
[Sun Oct 29 22:07:03 2017] e1000e: eth0 NIC Link is Down
[Sun Oct 29 22:07:03 2017] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[Sun Oct 29 22:07:54 2017] e1000e: eth0 NIC Link is Down
[Sun Oct 29 22:07:54 2017] e1000e 0000:01:00.0 eth0: speed changed to 0 for port eth0
[Sun Oct 29 22:07:54 2017] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
but after the reboot of the RN314 it changes back to 802..3ad LACP layer 2 every time again?!
This should be fixed in the latest hot-fix. You could try rebooting your NAS. The hot-fix should be installed within 15 minutes or so after that. If you do nothing (i.e. don't reboot your box) it'll be installed within a week.
After creating the team on 802.3ad LACP Layer 3+4 in the RN314 it states ok but after the reboot of the RN314 it changes back to 802..3ad LACP layer 2 every time again?!
For offline issue in bonding mode on RN31X, it must update to 6.9.1-T119.
In the meantime, I'm running on FW level 6.9.3 and the setting is stable but still, there is no data received over one of the two connections within the bond. One is constantly receiving 4Mbps data from 4 separate devices in total so it should be possible to use both ports with setting L3+L4, correct?
In the meantime, I'm running on FW level 6.9.3 and the setting is stable but still, there is no data received over one of the two connections within the bond. One is constantly receiving 4Mbps data from 4 separate devices in total so it should be possible to use both ports with setting L3+L4, correct?
The L3+L4 setting only applies when sending. The switch LACP setting is what matters in the NAS receive direction. Most switches are hard-wired to use layer-2 hashing.
The odds of receiving streams from 4 devices on the same link is 1/8 - not that probable, but not shocking either. Using L3+L4 doesn't change those odds - what it does is potentially allow 2 streams going to the same device to use different links.
You might consider using a static LAG on the switch, and set up the NAS to use round-robin.
@StephenBwrote: The L3+L4 setting only applies when sending. The switch LACP setting is what matters in the NAS receive direction. Most switches are hard-wired to use layer-2 hashing.
Isn't any hashing algorithm valid exactly only when sending (so applied to - and for - traffic leaving the ports' aggregation on device where it is configured)?
@StephenBwrote: The L3+L4 setting only applies when sending. The switch LACP setting is what matters in the NAS receive direction. Most switches are hard-wired to use layer-2 hashing.
Isn't any hashing algorithm valid exactly only when sending (so applied to - and for - traffic leaving the ports' aggregation on device where it is configured)?
The OP said "One is constantly receiving 4Mbps data from 4 separate devices in total so it should be possible to use both ports with setting L3+L4, correct?" I pointed out that was a misunderstanding. And I said what you said - that the setting only applies when sending.
I forced the Switch in LAG for the two Ethernet-ports and it looks it's going to work. The Pioneer Pro with 802ad is working properly but even that one looks to work better than before so I think the automation functionality in case of LAG is not working as good as I hoped together with Netgear.
