Home of the Squeezebox™ & Transporter® network music players.
Page 6 of 6 FirstFirst ... 456
Results 51 to 58 of 58

Thread: Wi-Fi 6

  1. #51
    Junior Member
    Join Date
    Apr 2020
    Posts
    6

    Verizon G3100 router

    Anyone having experience with Verizon G3100 router?

    My 2 squeezebox radios have exactly the same issues since i upgraded to Verizon's most recent router a couple of months ago. They lose connection after a few minutes. if connected to ethernet , they work just fine.

  2. #52

    wlanpoke to the rescue!

    I'm not sure if this has been mentioned in this thread or not, but there's a solution/workaround to this issue mentioned in another thread. It's a little script called 'wlanpoke' that you install on your radio and it basically monitors the WiFi connection and automatically resets the connection whenever it goes out. I experienced WiFi connection issues for months with my four SB radios and I tried all sorts of different router settings and Vonets Ethernet bridges, but nothing I tried was successful for me. Finally I stumbled across this wlanpoke script and since installing it on my radios they have been working great again. No more lost WiFi connections!!

    It's fairly easy to install if you're comfortable using a command line. Just download the zip file and follow the installation instructions in the manual.txt file.

  3. #53
    Senior Member
    Join Date
    Nov 2007
    Posts
    122
    I have found this thread of great interest, as I may be the exception. And that may offer some hope to other Wi-Fi 6 Asus users.

    We have been running an Asus AX-enabled mesh system for over a year now with 3 wireless SB legacy clients (2 Touch & 1 Radio) - with no problems whatsoever. An RT-AX88U is the router, with two AC86Us as nodes. Wi-Fi 6 is FULLY enabled on the AX88. And the Radio in particular prefers connecting to the 88 over the closer AC86 (probably because the 88s signal is just that good).

    I just spent four hours today running a Touch in an outbuilding 70 feet away from its nearest 86 node, and it played flawlessly.

    What we have encountered on one or two occasions where a wireless client refused to communicate with LMS (red signal meter) was a DHCP conflict, which was remedied by clearing the DHCP table on the router. It also helps to assign the LMS host a static address, so the clients always know where it is.

    However, as our AX client roster is growing, next week I will be swapping out both AC86s for a pair of AX XT8 nodes. One of our node locations (reaching that outbuilding) requires a wireless backhaul that the three band XT8 can better accommodate in a series configuration. The 88 will remain the router. I will be closely watching whether the legacy wireless SBs can tolerate these newer AX nodes.

    I suspect from the issues others are describing that the issue is router based. And at least in this configuration, the AX88 has gotten along with the legacy SBs, and I do not expect that to change.

    We have a total of 10 SB clients, and over 50 network clients total, but the majority are ethernet-connected.
    Last edited by sgmlaw; 2021-04-30 at 23:35.

  4. #54
    Senior Member
    Join Date
    May 2017
    Posts
    784
    Interesting to hear how this goes. Nice to know would be what your environment is. Is the building outside a city, are there other wifi sources in the neighborhood etc.
    SqueezeBoxes: 1x Transporter (Living room) 1x SB2 (shed), 1x Radio (Kitchen), 1x Boom (Dining room), 1x piCorePlayer (jacuzzi), 1x piCorePlayer (Garden) 1x OSMC + Squeezelite (Movie room), 1x Touch (Study 2), few spare unit's (SB2, SB3, Boom, Touch)
    Server: LMS on Pi3B+ 8.1.2 on PcP 7.0.1
    Network: Draytek, Netgear Smart Switch 24p, Ubiquiti PoE, 3x Ubiquity

  5. #55
    Senior Member
    Join Date
    Nov 2007
    Posts
    122
    Quote Originally Posted by edwin2006 View Post
    Interesting to hear how this goes. Nice to know would be what your environment is. Is the building outside a city, are there other wifi sources in the neighborhood etc.
    Some environmental particulars: The main structure is four levels totaling 3,500-4,000 sq. ft., with the maximum interior run dimension being approximately 125 ft. It is not uncommon for a client to be 2-3 floors and 5-6 walls away from the server. No masonry or steel construction, which is a plus in this context (but I will add that one level is partly below-grade, with a cinderblock envelope). The extended exterior roaming zone covers about 2 acres of mixed wooded and open space, with an outbuilding @ 70 ft. from the main structure. The general outside vicinity is moderately congested, but primarily in the 2.4 band with lower-tech traffic. I can detect good signal from five surrounding outside wireless networks besides our own from the property.

    This physical environment is why the property was previously wired for gigabit ethernet when 10/100 fast ethernet and 802.11b were still the standards (the primary distribution trunks to the server have since been upgraded to be 10G-capable). The primary switching equipment is Cisco 350 level quality, and those elements have retained a 10-14 year effective service life. But we have stuck with Asus prosumer-grade routers and wireless components due to the faster obsolescence curves of those elements (and they tend to be upgraded every 3-4 years as newer wireless standards emerge).

    It was a difficult single-point wireless install in years past, and earlier AP and repeater solutions were frankly excessively complicated (including from a security perspective). And the ethernet infrastructure decision has proven a boon for many years, as there is still nothing better for fixed clients (such as SB players). Our TPs and Touches never have dropouts thanks to ethernet.

    Mobile clients were the historic problem, but there were few of them. That has obviously changed with the rise of IoT. The introduction of mesh technologies has eliminated that last obstacle, and we now get excellent signal across the entire property, especially with ethernet backhauls (yet another benefit of that earlier investment). The future is AX, but the wireless tools have dramatically improved in the last four years.

    One tip for others, to reduce primary wireless loads and routing congestion is to embrace independent automation mesh protocols, such as Zigbee and Redlink (Honeywell). Loading down your primary wireless channels with 50 smart light bulbs and ten smart thermostats and sensors is asking for router and bandwidth headaches. These separate proprietary networks remain separate, with the only link to the primary network being a manageable ethernet hub/gateway link.

    As I mentioned, in our instance, we only need to deal with a relative handful of older legacy wireless clients. And I understand that our network infrastructure exceeds the usual residential standards. But IT WORKS in our Asus AX router-driven environment, and that is the takeaway for this conversation.

    If the Asus AX nodes prove to present an obstacle, we will probably still keep them and keep working the problem. The one Touch could probably be run to ethernet. And iPeng playing on an iPad Pro is not a bad substitution for a balky Radio.
    Last edited by sgmlaw; 2021-05-01 at 08:03.

  6. #56
    Senior Member
    Join Date
    Nov 2007
    Posts
    122
    Having spent several days with a pair of Asus XT8 as nodes to an RT-AX88U router, in a fully-enabled 802.11ax wireless network with older SB legacy clients, I can now give some preliminary observations. Of note is that in the middle of configuration and shake down, Asus released a firmware update of the XT8, which complicated matters for an additional day or two.

    Generally speaking, the integration of the XT8s as new nodes went off uneventfully. I prefer to connect new nodes by direct ethernet to the router, and then enable and configure the new nodes through the HTTP router controls. The first XT8 was configured with an ethernet backhaul. The second was configured as a wireless backhaul bound to the first XT8, which utilized the second 5ghz channel as a dedicated backhaul, as designed. Of note is that the primary XT8's ethernet backhaul traverses an intermediate switch layer. So the subordinate XT8's backhaul path to the router consists of a 5 ghz leg, and an ethernet leg through a managed Cisco 350 series switch back to the AX88.

    Following a full network reboot, all clients successfully logged into the new mesh network, and the network then re-assigned the wireless clients amongst the various nodes over a 24 hour period.

    As earlier stated, we had three (of ten total) Squeezeboxes operating in wireless networking: two Touches, and one little-used Radio that is mainly confined to one location. These did exhibit connection issues as the Asus mesh started juggling around its wireless clients.

    First, it is clear that the XT8 currently struggles with older 2.4ghz radios, particularly operating under n and earlier 802.11 standards. What we observed was that the nodes would simply drop them out when transferring clients (all clients) amongst the nodes. This was not uniform, and newer dual band clients have remained more stable with the XT8s at 2.4 ghz; what we noticed with these is that the XT8 would initially pick them up at 5 ghz, and then they would reassign to the router at 2.4 ghz. However, the 802.11g radios in the Touches struggled to maintain connection with the XT8s over time, and the Radio's transmitter was even a worse match for the XT8. Binding the SB clients to the nodes did not alleviate the drop outs that inevitably occurred as other roaming clients were being reassigned. We also noticed some conflicting DHCP reassignments with the Radio, which has happened previously with this model in Asus router environments. When that occurs, the tell-tale sign is that the client connects to the network, but is unable to communicate with LMS.

    Of further note is that most of our 2.4 ghz clients are IoT or third-party issued devices, and are confined to an Intranet-restricted guest network for security reasons. In an Asus mesh system, all guest network traffic goes only through the main router; the nodes are not used. What we observed was, predictably, that all 2.4 ghz guest clients were completely unaffected by the mesh changes. This confirmed that the AX88 at its present firmware revision has a very stable 2.4 ghz section.

    Because we had some very early issues with the Radio when the AX88 was first introduced, and now the AX88 has a very stable 2.4 ghz section, I believe that Asus will eventually do the same with its XT8 series. But for now, the implementation in the XT8 is problematic to the elderly legacy radios in these SB clients.

    The good news is we were able to get all three Squeezebox clients working, using various workarounds, but primarily by binding the clients to the AX88 router. For one Touch, because ethernet was available in the room, we simply ran a cable to it, and eliminated the wireless issue altogether. Yes, this is cheating, as that SB is no longer wireless. But it was quick and easy. The device that sacrificed its ethernet to the Touch has a much newer AC radio, and has maintained a stable and excellent connection with the nearest XT8 node (and a faster one than its fast ethernet allows).

    For the Radio, what we found to work was binding its assignment to the AX88 router, and clearing all earlier DHCP table assignments for it (this can be done via the Asus app). While the Radio and Touch's transmitters are antiquated, their 2.4ghz range offers a big advantage over newer 5 ghz radios. So even bonded to the router at some distance, the Radio receives a strong (100% LMS/ -59 dBm) signal, without resort to node-roaming.

    For the second Touch, in an outbuilding, the solution proved more problematic. That outbuilding relies on a node for good signal, and the router is quite a distance away. Binding the Touch to the distant router resulted in a relatively weak signal that was unacceptable IMO (60% LMS/ -79 dBm). So for that client, we used a more powerful 2.4 ghz Netgear 802.11n network extender operating in media bridge mode (and with a static IP assignment), with the Touch using its ethernet port. We allowed the extender to operate normally in the mesh network, and it initially connected with the nearest XT8 node, but then dropped out. However, the extender was later picked up by the router on its 2.4 ghz band, where it has remained locked. The extender maintains a good -65 dBm signal back to the router, better than the Touch's internal radio. However, we will probably at some point replace that extender with a 5 ghz AC model to allow a roaming connection to the much nearer XT8 node.

    So the summary is that the Asus XT8 does present some issues with older legacy 2.4 ghz clients. If your entire wireless network is going to consist of only XT8s, then you are going to have a problem running a wireless SB client, at least for now. But if you can maintain at least one other node (or in our case the router) as a non-XT8 device, then the solution is to bind the SB clients to that more mature device. The SB line is not incompatible with 802.11ax wireless technology in my opinion, only with some less mature and/or SOTA equipment (and that should improve). Were our mesh network entirely made up of AX88s, there would likely be no issues whatsoever. The problems are the continuing 2.4 ghz band teething issues with the XT8s.

    Again, our entire wireless network is now 802.11ax-enabled. And at least in the case of the Asus RT-AX88U, the old SB wireless clients function still well with it. The only limitation is the range of some of the SB radios over longer distances, in which case a more powerful range extender in media bridge mode presents a good solution. For those balancing their AX network clients with these older legacy ones, bear in mind that inserting an older AC node into these Asus mesh networks to accommodate legacy equipment will degrade AX client performance. Finding a more mature AX device to handle legacy traffic (such as the AX88) eliminates that compromise.

    I hope this provides some encouragement to those still struggling with 802.11ax wireless networks.
    Last edited by sgmlaw; 2021-05-09 at 00:27.

  7. #57
    Senior Member
    Join Date
    Nov 2007
    Posts
    122
    I will just add that we are now 4 days out, including several DHCP lease renewals, repeated client roams and shuffles, and including yet another set of firmware updates yesterday. And our Radio in particular (considering the forum) is maintaining a rock-solid connection to LMS. So it can be done in an AX network.

    Of note is that our one outbuilding Touch, wired into a stronger 802.11n extender, jumped over to one of the XT8 nodes for a while, and didnĺt miss a beat.

    I did notice when adjusting our one Receiver that the Controller did initially struggle with a connection to the XT8 node, but then locked on to the 88U, and all operated well.

    So again, if you are running WiFi 6, find an early release AX router or AP for your wireless Squeezeboxen. These will have the most mature firmware, and should be most likely to handle the old legacy radios.

  8. #58
    Senior Member
    Join Date
    Nov 2007
    Posts
    122
    Turns out the hiccup with the Controller may have been the Receiver. The Receiver just dropped dead in the past 24 hours. A fresh power supply, and it is back up as good as new.

    Otherwise, the one Touch/Extender combo in the distant outbuilding eventually was automatically moved to the nearest XT8 node, and is enjoying a stable -51 dBm/130 Mbps 802.11n connection.

    And it, along with the Radio, are performing perfectly in a full WiFi 6/802.11ax mesh network. A couple more firmware updates, and we may release the Radio to roam the nodes.

    So that is our story. Again, I hope it brings encouragement to those struggling that there is still a lot of good wireless life left in these old Squeezeboxes in 2021.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •