Hey everyone,
so thus far I’ve always had the pleasure of working with CRS3xx series switches and never had to deal with switch chip configurations. I just did all from the bridge, and hardware offloading just works.
Today I find myself setting up VLANs on a CRS112. I did everything as usual, enabled VLAN filtering on the bridge and the switch crashed due to CPU usage going through the roof. A quick google search landed me here, https://help.mikrotik.com/docs/display/ROS/Basic+VLAN+switching#BasicVLANswitching-CRS1xx/CRS2xxseriesswitches, thanks to which I was able to set it up, no problem.
Now, whenever one adds a tagged port or an ingress translation this does nothing. Until, of course, the VLAN is added to the VLAN table and those SAME PORTS are specified AGAIN.
This is ok for simple setups. But when ports and VLAN IDs start piling up, it sometimes becomes hard to spot the mistake. I think it would be helpful that the VLAN table gets dynamically populated whenever a tagged port or ingress port is added. And that this can be replaced by a manual entry with custom settings, the same way routing tables are handled?
And if that’s not possible, it would be helpful to mark tagged or ingress ports as “invalid” if there’s no associated VLAN table entry.
What do you guys think? Anyone else ran into this?