AP not boradcasting SSID

[dustin2]

Just finalized setup of an AP330, it's a standalone unit being used in my home. I have 3 SSID's that I configured in the Network Profile, it is the only profile I have. I have a VLAN for each SSID, so I ensured I made eth0 a trunk port which is connected to my Cisco Switch, connected to the RPi.

When I was running a config hand built in CLI, everything worked. Now that I have updated from 6.1R1 to 6.5R12 and have attempted to apply a config via the HiveManager Connect, there are no SSID's.

I just checked the AP, it was "stuck" on purple, which is either config or bootup mode, based on the led light information from the help files. I performed a reboot after saving the config, this was done via SSH so I know the static IP and basic routing is working fine. I would prefer to use the manager to configure the AP, but if necessary I have a backup of my CLI config to put in place. My CLI config is much smaller than the generated config, so I'm not sure what to look for.

Any help is appreciated!

[spurgelaurels]

NEVER MIND. I got it. Had to set the native VLAN on the Meraki to 1... Not sure why it's working now, but it is.

[spurgelaurels]

Apologies for the necropost, but this is exactly the issue I'm having.

When eth0 on my AP370 is set to trunk (native 1) the SSIDs don't show up at all.

When set to uplink, the SSIDs show up, but I can't get an IP address down to the clients. Upstream from eth0 is a Meraki MX65 with the port configured as trunk (native 20, which is my default home network)

[samantha_lynn]

Thank you for the detailed follow up! To answer your question, all changes need to be manually pushed out as an update to the APs, they won't update anything on their own.

[dustin2]

Mainly posting this in case someone else has similar problems and needs something to troubleshoot against. My adivce: Look at the generated CLI commands and learn them, this will serve you well.

TL;DR: Had my eth0 set to trunk, which was incorrect because there must be a backhaul connection to hook into. The Native-VLAN (in my case) needed to be VLAN 1, since setting the Management VLAN to 110 sets the rest of the ports to that by default as well, which seems like a weird default. *shrug*

That command passes.

I was working on it a little bit and I did get it to take a config and work, but that lasted for maybe a minute before something else that was being applied got in the way, which I just figured out.

I had Eth0 set to Trunk, which was incorrect in this case, or at least it keeps it from working after it pulls a config. Since it can no longer reach the backhaul, it doesn't load the SSID's. Now that I know that, I have set it back to backhaul. Corrects that issue. But now I have a different one, writing below.

Ok, so in the config for the AP, I'm setting the management VLAN to 110, which is what's needed for that to work when it first pulls the config and for each request after that. However, it is also setting the agg0, eth0, eth1, and red0 ports to the same config. Eth0 being set to that is what is causing it to fail now.

The chain of events I just went through were the following: I had it load the config. When it loaded, it did create SSID's now that the Eth0 port is properly configured. In doing so, it connects my computer to the DHCP server, but for some reason with Eth0 native-vlan set to 110, it fails to get a proper outbound connection to the HiveManager and resets the config after the timeout is reached. Repeat every time I try that. This last go around, I took advantage of the window and logged into the AP via SSH, and changed Eth0 native-vlan to 1, which now fixes that issue, at least for the moment for my primary SSID using VLAN 110. The other two SSID's work with this config, so it's literally that VLAN getting in the way somehow. I have "disabled" VLAN 1 in my router for security purposes, however that may not be a best practice, and if so, I may need to correct that when I have the time.

I just looked, if I update the config again, it is going to run int eth0 native-vlan 110 which will break the config. Leads me to a follow-up question: will this attempt to update it on it's own or do I still need to push it out? If I can control the release, then I'm fine with re-running the command when I update it since I'm at the end of configurations anyways.

Ok, just realized I needed to set the default Native-VLANunderneath the Management VLAN option, this should have some kind of (?) icon or something to clarify what they do for those that may not immediately get it.