Hi All

I hope this hasn't been covered before.

I have a client with a 2860 acting as just a router (and switch) and 4x AP903's setup as a wireless mesh (wired).

They all connect back to the 2860, that's the network in its entirety, nothing else is connected (other than the wireless clients).

Problem is, every month or two the whole mesh network breaks and each unit seems to just behave like a standard AP (although still in "mesh" mode).

All are still pingable, I can log into them and everything looks fine. Rebooting doesn't help.

If I try and reset the mesh config on each unit and start again, the root unit wont find any of the other units unless the admin password is set back to default (admin).

I have just updated them all to the latest firmware version and set the mesh network up again. I then changed the admin password back to something secure and all seems fine for the moment.

Has anyone else had the same experience or have any suggestions for stability?

Its all very basic in configuration, single SSID, WPA2 key etc

Thanks in advance.

Just to clarify, are you saying that if, for example, you login to the mesh root and look at the mesh status that it doesn't exist? (I assume that you're using one of the AP903s as the root?)

I have a 2926ac with 4 AP903s in a mesh (I only use the wifi on the 2926 to connect to a phone hotspot, it does not have any of the mesh SSIDs active) and have not experienced anything like that.

Just a thought - under "Mesh Setup" on the nodes you have ticked "Wired Uplink" - as you say they are all wired?

truckshark wrote:
Problem is, every month or two the whole mesh network breaks and each unit seems to just behave like a standard AP (although still in "mesh" mode).

All are still pingable, I can log into them and everything looks fine. Rebooting doesn't help.

My AP903 / AP802 combo exhibit similar symptoms, but in a much shorter timeframe (a few days). However, in my case, rebooting 'solves' the problem - so they are both set to auto-reboot on a daily basis.

You could try setting up a Syslog server - they are both very chatty... and the output is often full of "oops" messages, "Traces", "Stack Dumps" and "Invalid command" (the latter at 5 second intervals!). In other words, they appear riddled with bugs ... but that's just the Draytek way; by the time we get to V3.0.0, they'll be stable

Thanks for your replies, sorry for the late response.

If i log into the mesh root device the status page exists but no devices are listed within, as if it hadn't been setup yet.

I would expect to see the traffic light type system (green for online, red for offline etc) but the nodes are not even listed.

What gets me is that its really easy to reset them back up again but only if i set the root device to have its default admin password!

Yes all nodes have "Wired Uplink" selected.

I might try the syslog and see what it has to say for itself.

I have been monitoring it for the past 10 days and it had so far been stable. Obviously i'm now running on the latest firmware so maybe that has made a difference.

Thanks both of you for your input.

Just a quick update...

Devices have been running solidly since my last post other than for a user error (physical disconnection of the master AP!).

I have just had a call from the customer to say the wireless went down for 2-3 minutes, Mesh Status page indicates no issue.

Syslog just logs numerous "internal error - invalid system command" errors.

I am starting to think this is an issue with the kit and not a configuration/environment issue.