Firmware Road Map

[sirhc]

There is the ability to assign non routable IPs to each VLAN under the VLAN tab.

[tonym]

LLDP-MIB and CISCO-CDP-MIB support for discovery information would be great. (LLDP preferred over CDP).

Would be great to have this information available for our LibreNMS link mapping :)

[sirhc]

We do support LLDB and CDP but I am guessing "-MIB" is a way to automatically discover then download the needed MIB files from the device?

[speed101]

Not quite sure were request go but is the switch core capable of supporting 802.1ah Mac-in-Mac? I'm currently looking into redoing our network and trying to see what devices support what current technologies.

https://en.wikipedia.org/wiki/IEEE_802.1ah-2008

[Eric Stern]

Not quite sure were request go but is the switch core capable of supporting 802.1ah Mac-in-Mac? I'm currently looking into redoing our network and trying to see what devices support what current technologies.

https://en.wikipedia.org/wiki/IEEE_802.1ah-2008

Unfortunately our switch core does not support this.

[highlands]

Sirhc, it's been a few months and I hopes I am not being premature, however I am in hopes that your processor manufacturer might have updated you with when the "next generation" switch cores might be available. the feature sets of MPLS, SFP+, and are some strong features that I believe would befit many of us. It's been a quarter and was hoping for an update if one is available.

It would be a real break through and huge asset if the new DC 24 port switch had that designed in to it! I would pay more for it and would bet others would too. I understand (I think) Netonix was trying to make a product that would be better than the 'tough crap' line but you have made a 'Swiss army knife' that just stomps them it to yester-year! Please do not stop moving forward on pushing this great product line forward.

John

[sirhc]

We will have a new line of switches later this year that have SFP+ and a lot of other things.

[highlands]

Chris, you and your company and the employee's there rock! We can't wait (but I guess we have to) for the new product line.

John

[sakita]

Any update on MACFF and DHCP Relay? Seems to me this was left with MACFF being added to the todo list but DHCP not being added.

MACFF may be a useful feature (e.g. along side DHCP snooping) but it is not a substitute for DHCP relay.

For now our workaround is that a neighboring switch that supports DHCP relay ends up relaying the request. This works but we lose one of the valuable features of DHCP relay which is that we can know what switch the DHCP request came from.

Someone may ask why not have a DHCP server on each VLAN? DHCP relay allows us to use one DHCP server for the whole network - clean central management of addresses and greatly simplified configuration effort.

To compare the two, here are the basics on MACFF and DHCP relay...

Wikipedia...

In small networks, where only one IP subnet is being managed, DHCP clients communicate directly with DHCP servers. However, DHCP servers can also provide IP addresses for multiple subnets. In this case, a DHCP client that has not yet acquired an IP address cannot communicate directly with the DHCP server using IP routing, because it does not have a routable IP address, nor does it know the IP address of a router.

MAC-Forced Forwarding (MACFF) is used to control unwanted broadcast traffic and host-to-host communication. This is achieved by directing network traffic from hosts located on the same subnet but at different locations to an upstream gateway device. This provides security at Layer 2 since no traffic is able to pass directly between the hosts.

<snip>

Allied Telesis switches implement MACFF[1] using DHCP snooping to maintain a database of the hosts that appear on each switch port. When a host tries to access the network through a switch port, DHCP snooping checks the host’s IP address against the database to ensure that the host is valid.

MACFF then uses DHCP snooping to check whether the host has a gateway Access Router. If it does, MACFF uses a form of Proxy ARP to reply to any ARP requests, giving the router's MAC address. This forces the host to send all traffic to the router, even traffic destined to a host in the same subnet as the source. The router receives the traffic and makes forwarding decisions based on a set of forwarding rules, typically a QoS policy or a set of filters.

Also, is there (or will there be) a public list of the upcoming features that are in the roadmap?

[Travis_WVN]

Gonna drop my request in again for switch-fabric type features! I'd absolutely love to see SPB (802.1aq) or TRILL on these switches. It would really help us stay competitive to be able to tie all our towers together with multiple 1Gb backhauls and really pump some bandwidth across our network.

And thanks a ton for working on that DC-isolated switch! Makes it a lot easier when we colo on cell sites.