Team, Here is the final installment of the dependent mode AP5131 Mesh testing and I am pleased to say that everything worked as expected! The AP5131 dependent mode client bridge was disconnected from the RF Switch for 3 days with a client connected. Aftert the 72 hour time period expired the client radio was disabled and the client connection was dropped. Without resrtting the AP5131 dependent mode client bridge, I restablished  the connection to the RF Switch. The dependent mode AP established a connection with the RF Switch and once the configuration was pushed the client radio was re-enabled and the client conneciton re-established. To conclude, from my testing there are absolutely no issues using a dependent mode AP5131 as a client bridge once it has recieved its initial configuration the the RF Switch: 1) The dependent mode AP is able to be adopted over the Mesh backhaul link. 2) The dependent mode AP will maintain a client connection (independent WLAN) in the absense of an RF Switch for 72 hours. 3) After 72 hours the client radio will be disabled. 4) Once a connection with the RF Switch is re-established, the dependent mode AP will be automatically re-adopted over the Mesh backhaul link and the client radios re-enabled. Please feel free to ring me directly if you have any questions! Regards, Kevin

Gustavo, this is not what I understood from Kevin´s previous message. Once he defauled the Client AP´s and they worked, it has not became clear to me that we really need to first clabe it in order to get in touch with the switch. By the way he describes the test done, any dependent AP can be installed and it will be adopted over the mesh link and that is it, right?

All, Our initial response that you could not use a dependent mode AP as a client bridge was wrong as we recieved incorrect information from engineering. We appologize for any confusion this has caused! From my own personal testing as well as testing and validation from TNV you can use a depdendent mode AP as a client bridge. Please note the following: 1) The dependent mode AP has to be initially connected using Ethernet to recieve its AAP Mesh configuration. In other words the AP has to be statged somehow! The AP cannot get its configuration without being initially adopted by RF Switch which must be done over Ethernet! 2) Once the dependent mode AP has its configuration it can be turned off and deployed as a client bridge (i.e. completely wireless). The dependent mode AP will maintain its Mesh configuration.  Once powered back on, the dependent mode AP will come on-line, establish a mesh connection with the base-bridge and be adopted by the RF Switch. Again NO Ethernet is required! 3) The client radios will not come on-line until the dependent mode client bridge has been adopted by the RF Switch. When you power on the dependent mode client bridge it will connect to the base bridge using Mesh with no problem. However if the dependent mode client bridge cannot establish a connection with an RF Switch it will NOT enable the client radio. Once the AP has been adopted the client radio will be enabled. 4) If during operation the client bridge looses communication with the RF Switch, the dependent mode client bridge will continue to function for 3 days. Both the mesh link and client radios will remain active! The outstanding question which is being validated as I type is what will happen to the dependent mode AP after 3 days of no communication with the RF Switch. Both myself and TNV are testing this scenerio now and I will provide the list with an update of the results. Regards, Kevin

Gustavo, As with a non-dependent mode AP, pre-staging must be performed using the Ethernet connection so that the AAP configuration is pushed to the AP. Likewise the configuration can be pushed to the AP using TFTP if desired. In other words there is no difference in the pre-staging process between the dependent and non-dependent mode AP as the Mesh configuration has to be applied somehow! However once the initial configuration has been pushed to the AAP, the RF Switch discovery can then be performed over a Mesh link. The AAP will NOT have to be connected to the Ethernet port once the configuration has been applied. This was the questionable aspect which I verified this morning. The only outstanding question is what will the dependent mode AP do after it has lost connection to the RF Switch for 72 hours. I believe the AP will reboot and attempt re-discovery. If this is correct the AAP will re-discover the RF Switch using the Mesh link. We will however need clarification from engineering on this.  I will be raveling on business later this week and will be presenting at the summit next week. As such I will not have any time to test this scenerio until my return in two weeks. Regards, Kevin

Why would the customer have to connect the AAP to wired network? I have allready prooven that the AAP can be adopted over the Mesh link without having any LAN connection. Worst case the AAP would have to be power cycled for re-adoption to occur. KM

This is interesting. From what I have heard so far the 5131d will not activate its WLAN before it connects to the controller. And 72 hours will only work if AAP does not reset. Imagine Connection to RFS fails and 5131d resets. Afaik 5131d will NOT connect to RFS if not connected via LAN. Can someone write down the facts here? Cheers, Oliver

Thanks Kevin for the contribution, So Dependant AP5131 can be used in any part of the mesh but deployment of client bridge AP's must be done by an initial wired connection to the WS. Then, as far as the connection with the WS persists or does not get down for more than 72hs, all the mesh will work fine. If the connection resumes automatically then we have a product to sell. If a new connection to the wire is required again we don't. Can you make this test Kevin, I don't have all needed gears with me. Regards,

Kevin, this is good news, thanks for taking this on. I have one more thing to suggest.  Can you leave your test set-up running for 3 days to check that the Mesh AAP remains connected and functional after this time? Thanks, Chris

Just a comment. From a customers operations standpoint,   if the client-bridge dependent AP5131 goes down after 3 days,  the customer  will need to uninstall the AP5131 from a high ceiling, pole or remote location, bring it down,  connect it again to the network for adoption and hang it again on its location?   I guess if the customer is okay with that scenario, there should be no problem

If this is the case, it would force customers to purchase two different SKUs of AP7131. Logistically, this could make deployment more complicated if installers have to know which AP goes at the edge and which can be wireless, especially if they physically look identical. It also diminishes our "going totally-wireless" concept.

Interesting.....I would like to get to the bottom of this, so may well try and run this up in the lab.  Totally agree with Afshin's comments it does make deployment more complex and of course pushes the price up.  This discussion is very timely as it directly impacts an opportunity I'm working on right now. Chris

Team, Just to conclude this thread, I managed to snag a ADP5131 dependent mode AP running 2.3.1.0-001R and tested it as a Client Bridge and it worked fine! Here is how I tested it: 1) I connected the ADP5131 LAN1 port to a PoE Switch and adopted it to an RFS7000 Switch. 2) I waited for the AAP configuration to be pushed to the AAP and the Mesh backhaul link to be established. 3) I disconnected the ADP5131 from the PoE Switch which powered down the AP. 4) I connected the ADP5131 to a PoE injector which provided power to the AP but no Ethernet connectivity. 5) I monitored the console on the AP and RF Switch and observed the ADP5131 being adopted by the RFS7000 over the Mesh backhaul link. I repeated this test twice defaulting the configuration each time!   To conclude the ADP5131 dependent mode AP can indeed be used as a Client Bridge and RF Switch discovery will be performed over a Mesh link. There is no reason that I can think of  why you cannot use a dependent mode AP as a Client Bridge! The only limitation that I know of is that a ADP5131 dependent mode AP can only survive for three days in the absense of the RF Switch. If you need the AP to survive for longer periods of time you should purchase the non-dependent mode SKU. Regards, Kevin

Hi Kevin, Could you provide clarification on this point... You COULD use a dependant mode AP as a client bridge IF you first stage the whole network by connecting all APs to the RF switch using cable. I understand the downside to this, but just wanted to be clear that it is possible, just not recommended.  Please correct me if I am wrong

From what I have been told it cannot. From what I gather the dependent mode AP will not support contacting the RF Switch over the WLAN and can only contact it over Ethernet. As such a non-dependent mode AP has to be used as a client-bridge. Regards, Kevin

Gustavo, Dependent mode SKU APs support the same features and functionality as non depdendent SKUs. The only difference between the two is that a dependent mode SKU AP must see an RF Switch to function and can only survive for 3 days in the absense of an RF Switch. As Chris mentioned you can use a dependent mode AP as a base bridge, however as the dependent mode AP requires an Ethernet connection to initially contact the RF Switch you cannot use it as a client bridge. Regards, Kevin

Dependent mode APs can be used as a Base Bridge with no problem.  However, a Dependent mode AP must see a switch before RF operation commences.  For this reason we do no recommend using Dependent mode APs as a client bridge.  Any mesh node that may operate as a client bridge (no wired connection to the switch) should be populated by a 'full' AP rather than a D-mode one.    CSH

Thanks guys, now it's clear for me and that way I confirmed what can be offered. Regards,