{5} Assigned, Active Tickets by Owner (Full Description) (13 matches)

I FLASH NANO STATION WITH THE NEW NG BUT NO CHECKING ON THE DASHBOARD ALSO STAY ON CHANEL 3 INSTEAD CHANEL 5

I'm encountering something like bug #32, but different enough that I thought it should be it's own ticket. I have 5 OM1P's running r265, three as gateways, two as repeaters, all in the same mesh area. They are set in bridged mode, so that we can hand out IPs from our DHCP server as well as allow domain authentication on the computers using wireless.

This works fine with one gateway hooked up, but a few minutes after I boot up any of the other gateways, access to the LAN/Internet becomes very spotty. (I can still connect to the APs, but on SSID#2 I get no IP address and on SSID#1, I cannot access any LAN resources.) I'm guessing Spanning Tree Protocol is being triggered on our Cisco switches due to a broadcast storm or it detecting the mesh as an Ethernet loop.

Leaving it up and running overnight showed very spotty checkins, it looked like the gateways connection to the LAN was constantly going up and down.

If I remove all but one gateway, everything works fine.

Hi, I know Client Isolation (AP Isolation) is not yet implemented in NG. Clients associated with any node in the mesh are visible to all other clients throughout the mesh.

I can't deploy even a test network with NG due to this. I have a network running at my home with 3 wifi cameras streaming to a local server, and am impressed with NG in general, but I can't deploy it for enduser use until Isolation is implemented.

Looking forward to when you're able to implement Isolation between clients associated with AP1. Hope to provide some more useful feedback at that time.

thanks. -Dale

OM2p system, with 4 nodes, one gateway. The 3 remote nodes have client systems attached. NG mode, SSID2 bridged, wired bridged. They are POS terminals at a bar - all use the same subnet. All POS terminals are on Static IP.

Turns out - that unless at least one device does a dhcp request on a wired port of an OM2P - the static IP devices cannot communicate. So unless there is one DHCP packet after an OM2P resync - the network fails. In the lab we had a dhcp client device so it worked. In a panic, we have added a managed switch on the OM2P remote wired port that will create that DHCP packet - but we have to reboot that switch - so it does dhcp - often.

I could find no mention that wired remote only works if there is a recent DHCP request.

A wireshark capture of the OM2P wired port shows that until one DHCP request arrives, the OM2P device is doing DHCP requests itself on the wired port - and getting no response.

So - if the OM2P gets no response to DHCP on the wired port - why not assume it is a wired remote and start handling client traffic? At this point we need to remove the OM2P and install WDS AP devices to allow static IP devices to route consistently - what a disappointment!

would it be possible to bridge both SSIDs to different LANs? Lets say they have a separated backbone for there access points but employees would like to access local resources.

Could it be done by VLAN tagging or by setting SUBNET for "guest LAN" and "private LAN" on the dashboard and hock one node to the "GUEST LAN" and another to the "PRIVATE LAN". And routing could be setup from the IP the unit receives?

Bridge for SSID#1 is a "MUST" for many of our customers as they alrady have invested in AAA systems. Bridge for SSID#2 is nice

I would like to know the procedure to do a controlled OTA upgrade.

Is there some commands that can force an OTA upgrade so I can upgrade my nodes one at a time?

This would help since I can control the upgrade of these units without affecting users.

Thanks