Re: [Bloat] Xfinity Flex streaming box starves on cake?
Kenneth Porter writes: > I suspect that my Xfinity Flex box has too small an internal buffer and is > starving when fed by my cake-enabled OpenWrt router. From your config you have around 10x the downstream bandwidth you need for streaming, so unless you are really hitting your connection hard with other things I would not expect packets to buffer in the modem/router at all while you're streaming. So if this really is caused by sqm-scripts (which I think you should do a few more tests to definitely confirm), I would think it more likely it was due to some other weird interaction. Streaming box expecting a certain DSCP marking? Server choking on ACK stream due to filtering? Hard to say... A packet dump of the stream going to the box while the black screen happens may be illuminating, but could be hard to pull off if it's not reliably reproducible... -Toke ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat
[Bloat] Good Wi-Fi test programs?
A local company is rolling out fiber-to-the-home. Subscribers are switching from crummy Comtrend or ZyXEL DSL modems (whose PPPoE interface runs at 3 to 20mbps) to Calix GigaCenter equipment with symmetric ISP link speeds starting at 25mbps on up. The complaints I've started to hear are that the new Wi-Fi signal "is weak" - not reaching as far as the DSL equipment did, and that speeds are abysmal (one report is 2.4mbps down, 0.25mbps upload). Before I insert myself into the "unhappy customer" loop, I would like to ask this august group for thoughts about things to check/test programs to run/phenomena to consider. I would first do the following "easy tests": - Check for conflicting/overlapping Wi-Fi channels. I am fond of the free app, WiFi Analyzer from farproc (http://a.farproc.com/wifi-analyzer) for this test, but there are several similar Android apps. - Compare the signal strength for the DSL modem and the Calix modem, as shown by WiFi Analyzer - Be sure that all computer(s) are using the Calix modem. - Use a variety of speed tests: DSLReports, Fast.com, other favorites? - Compare speedtest results when the test computer is close to, or far from the router. - (If possible) compare the performance for both Wi-Fi and Ethernet - Shut off the DSL modem on my way out the door to be sure it's not causing interference or confusing the situation. Anything else you'd recommend? Second, are there other more fine-grained/analytic tests that I could run? I'll have a Linux or macOS laptop that I could use to look at other measures: retransmits, CRC errors on the Wi-Fi link, etc. And, finally, what do I recommend based on the information gained from those tools? Many thanks. Rich ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat
Re: [Bloat] Good Wi-Fi test programs?
> On 7 Dec, 2020, at 1:00 am, Rich Brown wrote: > > I would first do the following "easy tests": > > - Check for conflicting/overlapping Wi-Fi channels. I am fond of the free > app, WiFi Analyzer from farproc (http://a.farproc.com/wifi-analyzer) for this > test, but there are several similar Android apps. > - Compare the signal strength for the DSL modem and the Calix modem, as shown > by WiFi Analyzer > - Be sure that all computer(s) are using the Calix modem. > - Use a variety of speed tests: DSLReports, Fast.com, other favorites? > - Compare speedtest results when the test computer is close to, or far from > the router. > - (If possible) compare the performance for both Wi-Fi and Ethernet > - Shut off the DSL modem on my way out the door to be sure it's not causing > interference or confusing the situation. > > Anything else you'd recommend? Make sure the customer's devices are using 5GHz rather than 2.4GHz band, where possible. The Calix devices apparently support both and try to perform "band steering", but it's worth double checking. https://www.calix.com/content/calix/en/site-prod/library-html/systems-products/prem/op/p-gw-op/eth-gw/800e-gc-spg/index.htm?toc.htm?76518.htm I also read while briefly scanning the accessible documentation that Calix operates at maximum permitted wifi transmit power and with up to 80MHz RF bandwidth. While this does maximise the range and throughput of an individual AP, many such APs in close proximity will see the RF channel as "occupied" by each others' transmissions more often than if a lower transmit power were used. The result is that they all shout so much that they can't hear themselves think, and clients can't get a word in edgewise to send acks (with generally lower transmit power themselves). You should look for evidence of this while analysing channel occupancy, especially in multi-occupancy buildings. It's probably less of a concern in detached or semi-detached housing. I didn't see any mention of Airtime Fairness technology, which is now a highlighted feature on some other manufacturers' products (specifically TP-Link). Ask whether that is present or can be implemented. You may be able to test for it, if you have established a case where wifi is clearly the bottleneck, by passing a saturating ECN Capable flow through it and looking for CE marks (and/or ECE feedback), since Airtime Fairness comes with built-in fq_codel. - Jonathan Morton ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat
