back to article OK, Google: Why does Chromecast clobber Wi-Fi connections?

Wi-Fi router vendors have started issuing patches to defend their products against Google Chromecast devices. TP-Link and Linksys were first out of the blocks with firmware fixes, and TP-Link has posted this explanation of the issue. The bug is not in the routers, but in Google's "Cast" feature, used in Chromecast, Google …

  1. This post has been deleted by its author

  2. Anonymous Coward
    Anonymous Coward

    Its more than just *cast devices doing this, its also all Google Home devices, and also may still be google's Nexus Player

    This is a screw-up of ginormous proportions, worthy of a company with a name like "googol"

    I hope Netgear release firmware for other older routers too, not just their latest Orbi Mesh one.

    1. Mark 65 Silver badge

      I think they should just issue a statement calling out Google's shitty coding.

      1. Anonymous Coward
        Anonymous Coward

        Google employ coders? I thought they just sat around all day drinking poncy coffees like Q whilst rifling through your NHS Heath Data records.

        1. wolfetone Silver badge

          "Google employ coders? I thought they just sat around all day drinking poncy coffees like Q whilst rifling through your NHS Heath Data records."

          That's Virgin Health you're on about, isn't it?

      2. mathew42
        Unhappy

        Router code is just as crap

        Router coding doesn't seem significantly better when a flood of packets will fill up the router's memory and require a physical report.

        In summary, any (malicious) code on the network with access to generate UDP packets can cause most routers to lock-up.

        1. Robert Heffernan

          Re: Router code is just as crap

          "Router coding doesn't seem significantly better when a flood of packets will fill up the router's memory and require a physical report."

          It's not hard...

          If(packetBuffer.IsFull() == true)

          {

          packet.Drop();

          return;

          }

          1. dajames Silver badge

            Re: Router code is just as crap

            If(packetBuffer.IsFull() == true)

            Or even just:

            If( packetBuffer.IsFull() )

            ... because, you know ... Boolean logic.

            The compiler will probably emit the same thing in each case, but the shorter way is easier to read/maintain.

  3. Mark 85 Silver badge

    Send packets every 20 seconds and then goes into some version of "panic mode" if the Mother Ship can't make contact. Right? WTF...is this even necessary? It truly would appear that Google wants to be Big Brother. Add to that, if you're on a metered connection, it could get expensive real fast.

    1. Sampler

      One would assume that's on the lan side, for devices the chromecast will be connecting to for casting from, judging by the wording of the article, the phoning home will be a different flood of packets as it sends back everything it's recorded whilst "asleep" (in terms of Google Home devices).

    2. Lee D Silver badge

      Chromecast announces its presence on the LAN at regular intervals so that devices that want to Cast know there's a Cast-compatible device.

      No different to Airplay, DLNA, or any similar technology.

      The stupidity is in sending a packet for every announcement "missed" because it was asleep, all lumped together the second it wakes up. That's just dumb.

      1. sanmigueelbeer Silver badge
        Thumb Up

        That's just dumb.

        Nope. It's just Google.

        1. Jeffrey Nonken Silver badge

          "Nope. It's just Google."

          Oh, this type of mistake is certainly not uniquely Google.

          1. Anonymous Coward
            Anonymous Coward

            This type of mistake is certainly not uniquely Google

            Maybe not, but few other companies publicly pursue the mistakes of their competitors as aggressively as Google does. I think Google earns just about every iota of scorn that is flung back at it.

      2. Dyson Lu

        The Google coding team responsible for their Cast product has become pretty bad. Worse is that they force-upgrade so you have no choice but swallow the buggy code.

      3. Bandikoto

        It's typical "I've been taught to code, but I don't actually understand computers and the culture/environment in which they're used". People who have no idea of the limitations of where their code will be executed — i.e. why it's unreasonable to bang out N packets all at once, rather than feed them to a rate-moderated emitter, or in this case more than N (for small values of N) is unreasonable to stuff into your for(register x=0;x<N;x++) resend_MDNS_announcement(last_index_sent+x); /* spot the errors */ loop.

        What is 2,000,000 packets blasted out on your typical 802.11g network in a typical crowded urban environment, anyway?

        Why, indeed, is it unreasonable for any one subsystem to be able to allocate all of the available RAM for its own use and then busy-wait while it's waiting for those 2,000,000 packets to be stuffed out the radio?

      4. IceC0ld Bronze badge

        The stupidity is in sending a packet for every announcement "missed" because it was asleep, all lumped together the second it wakes up. That's just dumb.

        short n concise precis of what the fail actually IS - cheers

        1. Roland6 Silver badge

          >short n concise precis of what the fail actually IS - cheers

          Well there are two parts to this, the dumb client being just one, the other is from my very brief analysis of a few routers I've recently had access to from: Netgear, Huawei, EE & Draytek, is that the 'cheap' domestic grade routers have very little real firewall capabilities. Only the Draytek (of my routers) supports DoS and UDP flood protection.

    3. Mike Dimmick

      I presume this is some form of timer that the OS stops when the Chromecast is sleeping - but then runs the timer handler for every 'missed' timer event when it wakes up. Therefore if it's been asleep for a minute, you'll get three attempts - but if it's been asleep for a week, you get 30,240 (= 3 x 60 x 24 x 7).

      1. Anonymous Coward
        Anonymous Coward

        Sounds like one of the internal monitoring tools we used to have on our corporate laptops.

        Sleep, or hibernate the machines, and on boot up, the HD was thrashed for a while, and how long seemed to be related to the length of the sleep. Could be thrashing for 30 mins or more after a weekend for example, and pretty much stopped all other usage of the device till it was done, due to badly set permissions (was low priority, but not low IO).

        Was traced to some monitoring tool, that would periodically scan logs, and create a summary that was pushed to a central server. You guessed it, it didn't take into account sleep or hibernate, and so just ran all it's 'missed' jobs in one go, and in parallel!

      2. NullReference Exception

        Sounds like the behavior of Java's Timer.scheduleAtFixedRate() and/or ScheduledExecutorService.scheduleAtFixedRate(). It fails horribly in this exact manner any time the system clock jumps (due to the system going to sleep, the time being set, etc.). Been burned by that too many times to count.

        Knowing that Google uses Java heavily, I'd wager this is the culprit.

        1. Destroy All Monsters Silver badge
          Headmaster

          Knowing that Google uses Java heavily, I'd wager this is the culprit.

          Nope, they are just using the API.

          "it fails horribly" == "I didn't read the JavaDoc. 21st century programming style?

          In fixed-rate execution, each execution is scheduled relative to the scheduled execution time of the initial execution. If an execution is delayed for any reason (such as garbage collection or other background activity), two or more executions will occur in rapid succession to "catch up." In the long run, the frequency of execution will be exactly the reciprocal of the specified period (assuming the system clock underlying Object.wait(long) is accurate). As a consequence of the above, if the scheduled first time is in the past, then any "missed" executions will be scheduled for immediate "catch up" execution.

          1. sabroni Silver badge

            @Destroy All Monsters

            Any idea why the Java engineers decided running all the missed instances was necessary? Most scheduled tasks I'm familiar with process whatever data is available when they run, so a single instance running would pick up all the missed data. I'm struggling to think of a situation where this "run multiple instances at the same time "behaviour would be of value.

        2. Anonymous Coward
          Facepalm

          So Timer.scheduleAtFixedRate() in fact doesn't schedule at a fixed rate! Nice API...

    4. Anonymous Coward
      Anonymous Coward

      You are just embarrassing, give it a rest.

  4. Phil Kingston Silver badge

    Linksys, Asus, TP-Link, Netgear, Synology all affected.

    Good to see this getting some coverage now (issue started early December).

    Would be good to get at least an acknowledgement of the issue from Google.

    1. Dan 55 Silver badge

      It's a rather sad indictment that the way to get router manufacturers to put out updates to their firmware is Google screwing up.

      1. Anonymous Coward
        Anonymous Coward

        The router manufacturers needed to do something: DoS protection for starters.

    2. Coen Dijkgraaf

      Aknowledgement of the issue from Google.

      @Phil Kingston

      UPDATE: The Register understands that Google has admitted to the flaw, said it impacts a "small number" of users, and committed to providing a fix in short order.

      1. Adam 52 Silver badge

        Re: Aknowledgement of the issue from Google.

        This, in itself, is newsworthy. I can't think of any other time when Google has admitted to making a mistake.

  5. Kevin McMurtrie Silver badge

    Roku F-u

    This isn't new. Some Roku devices have a bug where its WiFi Direct channel is negotiated on the most active channel rather than the least, essentially killing your home WiFi. The difference is that Roku can't be bothered to fix it.

    1. Phil W

      Re: Roku F-u

      I believe the problem is actually that the WiFi direct channel is set the same as the WiFi channel the connection to your router is using, so that the Internet/LAN connection and WiFi Direct connection for the remote are on the same channel. I imagine this is probably due to latency introduced, particular for the remote, if it's having to operate on multiple WiFi channels at the same time.

    2. toejam13

      Re: Roku F-u

      Roku won't fix the WiFi Direct issue because they see it as a feature, not a bug. Luckily, there is an option to disable it buried deep in the settings. Unfortunately, my Roku ignores the option. So I had to fiddle with the power and interference settings to reduce the WiFi Direct transmit power instead.

      1. Mike 'H'

        Re: Roku F-u

        I'm not so sure the issue is "fixable".

        This would imply the Roku has _2_ radios, and not just the single radio I've been able to find so far.

        The single radio it has is a dual-band (but not simultaneously both bands at the same time) 2.4 and 5GHz. You'd need an entirely separate radio to make your setting even possible in a wirelessly-connected Roku scenario.

        Therefore, if you don't have the device plugged into a wired LAN, you are forcibly limited to being on the same channel that the Roku is already tuned to, to access your WLAN.

        When plugged into wired Ethernet, you can rightfully expect it to honor your settings; when connected wirelessly it cannot.

        If your Roku Streaming Stick for example doesn't have a Wired connection to begin with, you're SOL - or you could always leave it as an island, with no Internet connectivity, and expect it to perform admirably on your preset channel Wifi-Direct-only.

        1. Anonymous Coward
          Anonymous Coward

          Re: Roku F-u

          "When plugged into wired Ethernet, you can rightfully expect it to honor your settings; when connected wirelessly it cannot."

          You wouldn't be able to turn of Wifi direct even then because that's how the remote connects.

  6. wobbly1

    when in tandem...

    ....with the "Minusnet" supplied router failing to have an AP isolation mode , (they told me categorically my existing router would be incompatible, and having changed from ADSL2 to FTTC i took them at their word ) making the chromecast invisible to the any devices if it connects to the 5.4 Ghz network, i hadn't noticed this further hit. i have a tortuous work round for the no AP isolation problem, Minusnet have no interest in fixing this as searches for chromecast in their support forums shows.

    1. Anonymous Coward
      Anonymous Coward

      Re: when in tandem...

      Google about, you can more than likely use a BT OpenReach fibre modem (~£15 on eBay) and a router of your choosing. I'm on BT broadband, ask them and only the Home Hub works with it. I have an OpenReach modem and Netgear router.

      The difficulty lies in knowing how to get your router to login to the ISP account (rather than anything overly technical with the line), for BT it's fairly simple as they authenticate the line, so providing bthomehub@btinternet.com without a password works.

      If you can't find this for your ISP and you're fairly technical, look into hacking your router, even if it doesn't display it in the web UI, it'll likely be in a text file somewhere on it if you can gain telnet / file system access.

      1. Chloe Cresswell

        Re: when in tandem...

        Or if you need a new device: vigor 130.

        The only isp I know of with issues on getting usernames and passwords is sky.

        1. Anonymous Coward
          Anonymous Coward

          Re: when in tandem...

          The Vigor 130 is an excellent box, but do note that is an FTTC *modem* not *router*. So you also need something which routes to terminate the PPPoE session on.

          I use Plusnet + Vigor 130 + Mikrotik hEX PoE + Unifi AP AC Lite (x2), and it's a very reliable combination. The Vigor 130 has a web interface so I can see the actual line rate the FTTC has trained up to. Both devices support "baby jumbo" frames of 1508 bytes, so I can get 1500-byte MTU IP even with PPPoE.

          Instead of a physical router, you could use a software router/firewall like pfSense to terminate the PPPoE session. This means that your public IP will be directly routed to that box, which is useful for port forwarding etc.

          1. LDS Silver badge

            Re: when in tandem...

            It looks only that some UK models (maybe others?) are, because on the main site it is sold as a modem/router

            https://www.draytek.com/en/products/products-a-z/router.all/vigor130/

            1. Lee D Silver badge

              Re: when in tandem...

              Draytek Vigor routers are fabulous. I have the 2860 - firmware updates all the time (with new features as well as bug fixes), certified compatible with BT fibre offerings (and ADSL2, and Ethernet, and 4G). Failover, IPv6, all kinds of internal options including web filters and DNS filters and LDAP authentication and AP isolation and dual-frequency radios (including handover between frequencies for compatible hardware), proper QoS, SIP handling (including analog ports that run over SIP on board), VLANs, and every option under the Sun.

              I run my house and a work site off them, they are just solid, fast and so featureful you'll spend your life reading the manual and going "I didn't realise I could do that!"

            2. John Crisp

              Re: when in tandem...

              As far as I remember the UK version of the V120/130 is just a dumb modem, but they market the 'same' model elsewhere in the globe which is actually a full blown router & firewall with an interface and functionality similar to its grown up brethren, albeit with a single port. Same box, totally different guts.

              So the UK gets the cheap dumb version for about the same price as everyone else gets a full on router (which can also do dumb pass through if required).

              I did ask a few times on their forums as to why the UK seemed to be being ripped off, but they deleted all my posts..... took it up with their tech dept too and just got fobbed off with marketing bullshit.

              Long time Draytek user, but after a number of issues I'll be shopping elsewhere when I upgrade.

              1. Phil Kingston Silver badge

                Re: when in tandem...

                I was very pleased with the Vigour 130 I tried, the only issue being a ~10% sync speed drop compared to Broadcom-equipped modems on the same (VDSL) line. YMMV of course.

              2. defiler Silver badge

                Re: when in tandem...

                Long time Draytek user, but after a number of issues I'll be shopping elsewhere when I upgrade.

                I've had a few problems with various Drayteks over the past six months too. It's got to the point where I flung VMware on an old(ish) PC and slapped Mikrotik CHR into a VM. Use the BT Openreach VDSL modem and it's cracking now.

                Draytek still does the wireless, so it was no great surprise (although a little disappointing) that Squeezer on my Android handset kept choking whilst trying to connect to Squeezebox Server last week. I guess I'll be in for a new access point soon too...

                1. Roland6 Silver badge

                  Re: when in tandem...

                  >Long time Draytek user

                  Suspect you may already be aware that Draytek have released updated firmware for many of their products, including legacy products, this month.

                  1. defiler Silver badge

                    Re: when in tandem...

                    Draytek have released updated firmware

                    That's great, but I think the router is just broken. Not often a Draytek breaks, but it's not the first time either. I don't have Chromecast or similar in the house, and it was dropping out multiple times a day when I was the only person in the house and I was sitting running a Citrix session. Nothing taxing.

                    Anyway, Mikrotik is doing nicely for now, although the learning curve to set it up was steeped than Draytek :)

                    1. Roland6 Silver badge

                      Re: when in tandem...

                      >That's great, but I think the router is just broken.

                      Last week I resurrected a 'dead' Draytek (2710Vn); by replacing the power adaptor. It seems whilst the power adaptor was still working (and thus leading you to think it was the router that had failed) it would only supply power to a device only needing a max of 0.5A, the Draytek and it's up to 2.0A draw was too much for it.

        2. Stu J

          Re: when in tandem...

          Even better - go for the VigorBX 2000n...

          > in-built VDSL modem

          > 4G dongle backup option (if you want/need it)

          > VOIP PBX capability (it even supports your existing analogue line and phone, but the sound quality's a bit crappy - VOIP calls using a VOIP provider is perfect though)

          1. wobbly1

            Re: when in tandem...

            thanks all for the recommendations , had a Vigor single port router/ modem about 15 years ago , will be looking to use them again cheers ! :)

  7. DougS Silver badge

    Sounds like it is caching all the requests while it is sleeping

    Then sending them all out when it awakens. Must be fun for the people who rarely use it!

    1. David Roberts Silver badge

      Re: Sounds like it is caching all the requests while it is sleeping

      I did wonder if it just had a loop which ran until the internal date/time counter matched system date/time.

      There will be a limit to cached messages but no limit to brainless loops.

      Whatever, I hope someone reveals the code fix.

    2. Keith_Rhodes

      Re: Sounds like it is caching all the requests while it is sleeping

      If you rarely use it, you probably pull the USB power supply from the socket. That's what I do.

  8. werdsmith Silver badge

    I use the Chromecast thing only rarely, but I yank its power adapter out when not in use (I do this will all powered devices) so there is never this problem.

    1. Lee D Silver badge

      Mine is powered off the projector that it displays on (literally the USB plugs into the projector).

      I have a remote for the projector on my smartphone, so even if I desperately need it in a hurry, press button to turn on projector, by the time I get into a Cast-capable app, it's up and showing on the screen.

    2. TRT Silver badge

      Indeed, I have a mains bar that powers up when the TV is turned on, and the TV sleeps after 3 hours with no remote activity. I wondered what all the shitty UDP spikes were about - I thought it was just the backdrop picture changing. Still, I'll be changing it to the ultra with wired ethernet soon. Does that also suffer from the same?

    3. mathew42
      Facepalm

      You are using the device incorrectly. All devices should be on at all times to accumulate information and help the cloud become smarter.

  9. Dave 126 Silver badge

    Most Chromecasts I've seen in the wild are powered off the television's USB socket. If this is how Google tested them, it might explain how this issue escaped their detection.

    1. Sandtitz Silver badge

      Perhaps. Then again my Sony has the USB ports powered even in standby mode. I had to replace the external HDD for an SSD because of the constant whirring.

      1. Korev Silver badge

        Then again my Sony has the USB ports powered even in standby mode.

        That's not the case on my Samsung. In my mind having the device unpowered when the TV is in standby makes sense.

    2. lybad

      Google (and Amazon for that matter) both recommend their TV dongles are mains powered, as they won't guarantee TV's will power the device reliably.

      1. Dave 126 Silver badge

        > Google (and Amazon for that matter) both recommend their TV dongles are mains powered, as they won't guarantee TV's will power the device reliably.

        There's no harm in the end user trying the Chromecast off TV USB power - it'll soon be clear whether it supplies enough power or not... that disclaimer sounds like a arse-covering exercise because it would be impractical for them to test every TV set available. But yeah, that would suggest my theory of Google testing in that manner night be off the mark.

        1. lybad

          I agree - just pointed out what they recommend.

          In terms of the Fire TV Stick I had, if you powered the device from the TV, it wouldn't apply firmware updates.

  10. Anonymous Coward
    Anonymous Coward

    TP-LINK Chromecast Fix

    The fix issued is for (US) routers, will not load on Routers with firmware (EU). No additional fix as yet.

  11. Beardedtit

    Looks like it is using a timer API that tries to 'catch up' when the callout is delayed (even by sleep), so on wakeup all the callbacks happen back to back.

    Not only does this break the network, the device itself is probably unresponsive until it has finished.

    The real fubar is sending 'I am here' messages out at all, never mind at 20 second intervals.

  12. Gypsy Major

    Freeview interference

    I've had no problem with WiFi but my Chromecast does interfere with freeview reception. Hence I can't leave it on standby.

    1. Dave 126 Silver badge

      Re: Freeview interference

      If it's an issue I guess you place the dongle further away from the TV via an HDMI cable.

      Or it might be an issue with your TV - perhaps supplying USB is upsetting its freeview reception. Try running the Chromecast off a dedicated wall charger.

  13. delph

    I had a fun christmas chasing this bug with wireshark. (yes I should get out more)

    As I found similar problem with android phones waking up and blatting my network with mdns, that seriously irks devices like denon and marantz dlna devices.

    Solution so far has been to run a separate network for google and roku devices and living without controlling my audio from an android device.

    Its almost as if no company actually designs products properly *shock*

    1. Dave Robinson

      I had noticed the same issue with Android phones waking up and buggering the networking on my Marantz DLNA player. My solution involved a lot of swearing over the past couple of months, and then buying another Sonos speaker.

  14. beedee404

    Nice evasion router vendors...

    So... let me get this right...

    There's a device on your (notoriously secure & unhackable) WiFi network, that can disable your router by sending a specific large, rapid sequence of packets.

    And apparently that's the fault of the device manufacturer, not a DoS vulnerability in the router...?

    While the chromecast clearly shouldn't be doing this, surely the routers should also not be behaving in this manner?

    And blaming someone else for showing up a vulnerability in one of your products seems a bit childish...

    1. Wensleydale Cheese Silver badge

      Re: Nice evasion router vendors...

      "And apparently that's the fault of the device manufacturer, not a DoS vulnerability in the router...?"

      It's more like a neighbour's kid repeatedly kicking his football against someone's living room window.

      You can argue all you like that the victim should invest in stronger glass, but the real problem is a lack of control by the kid's parents.

      1. DaLo

        Re: Nice evasion router vendors...

        Or is it more like the neighbour's kid repeatedly kicking a ball against your front door and when he does your door lock pops open.

        The door manufacturer says "well the kid is at fault for kicking the ball against the door several times", you might reply "yeah, he shouldn't have been kicking the ball but right now I'm more concerned about the door's locks popping open"

    2. Anonymous Coward
      Anonymous Coward

      Re: Nice evasion router vendors...

      It's a device on the inside of your firewall on your own network that you've welcomed with open arms and authorised it with a key to access your wireless ness, so how is that a DoS vuln?

      1. DaLo

        Re: Nice evasion router vendors...

        Because "TP-Link warned, the burst will fill up the router's memory and leave a reboot the only option to restore connectivity."

        Hence a DoS, hence it is vulnerable to a DoS.

        1. Warm Braw Silver badge

          Re: Nice evasion router vendors...

          While the router shouldn't totally lock up under these conditions and ought eventually to recover, receiving a large number of short packets back to back will inevitably seriously impact the throughput for other devices on the network and could result in network devices being dropped temporarily, simply because the radio space is being hogged by the amount of traffic from one source.

          Domestic access points are typically not equipped with particularly fast processors and so there is a depdency on the underlying network hardware to be able to discard or ignore data that's arriving too quickly - it's possible in this case that because the packets are multicast they're being passed up the stack to the CPU but there's no limit on the number of buffers that can be allocated for that purpose, the CPU is getting behind and hence memory is being exhausted. A better thing to do might be to stop receiving altogether, but the AP would then effectively go deaf until it had caught up with the multicast traffic. However, unless you have a device that can process all potential traffic at the speed of the physical medium you're going to have problems of some sort - and that device wouldn't be competitive in a domestic market.

          Even if it were, shared-media systems (like wired and wireless ethernet) depend to some extent on the connected devices behaving: if you have a device that jabbers constantly, there's not much you can do.

          In short, it's OK to expect better, but you're not going to get perfect.

          1. Anonymous Coward
            Anonymous Coward

            Re: Nice evasion router vendors...

            Well Warm Brew, your post would be useful if it wasn't already stated that they had provided a firmware fix to the problem so it was relatively easy to fix and therefore could possibly been seen as a design flaw or even a bug in the first place?

            It isn't even difficult to solve, I can think of a number of ways to solve it without much thought. In fact most router manufacturers and Layer 3 Switch manufacturers include DOS protection. It's DDOS that is more difficult.

  15. Patrician

    Have two Chromecasts and an Nvidia Shield TV (Chromecast built in) with VM SuperHub3 and have never, in the three years I had both Chromcasts (first year was with SuperHub2), experienced this issue.

    All three devices are powered up 24/7 too, so it looks to me more like a router issue than a Chromecast issue.

    1. dnicholas Bronze badge

      Well duh, if they're powered 24/7 there won't be any problems

    2. Anonymous Coward
      Anonymous Coward

      Have two Chromecasts and an Nvidia Shield TV (Chromecast built in) with VM SuperHub3 and have never, in the three years I had both Chromcasts (first year was with SuperHub2), experienced this issue.

      With due respect sir, how would you know, given that the Hub 3 is a pile of total and utter shite based around the disgusting, poorly engineered Intel Puma 6 chipset? I can only conclude that the Puma 6 flaws negate the Chromecast flaws. In my experience, the connection to the Chromecast is dodgy, but mainly mid-stream. And then there's the shitty unsuitability for gaming, VOIP, Skype, and some VPN use...I'd like to break my Hub 3 into big jagged fragments and shove it up Virginmedia's boss's arse.

  16. Craigie

    Firmware updates?

    So 99.99% of home users will just have to replace their hardware then? Or junk their Google devices.

  17. handleoclast Silver badge

    A wild guess

    See title: this is a wild guess...

    My guess is that this is some fuck-OOPery. There's a timer somewhere bunging "send DNS multicast" events into a queue. When the device is asleep, because this is a typical fuck-OOP, instead of the timer ceasing to add events to the queue, some handler merely suspends processing the queue so the queue lengthens and lengthens and lengthens. And then the device wakes up...

    Yes, you can achieve the same stupidity in non-OOP languages. It's just that people are less likely to make mistakes like that in non-OOP languages because you have to have some sort of understanding of what's happening instead of putting Lego™ blocks together according to some "pattern."

    If I were to have an even wilder guess, I'd say Java.

    OK, I'm being cynical and ranting non-justifiably against all things OOP merely because I hate OOP. It's Monday. Maybe I'll be in a better mood tomorrow.

  18. whatsyourShtoile

    Google really at fault?

    Pardon me sir but I thought a router was for routing packets. Large numbers of packets, small numbers of packets, it doesn't really matter right?

    1. Anonymous Coward
      Anonymous Coward

      Re: Google really at fault?

      whatsyourShtoile, (Love the name :) )

      It is the usual problem the 'Wrong type of Snow/Leaves/Packets etc !!!" :)

      Lots of other types are OK but ....... <Sigh> <Grin>

      I would be very interested to know what the 'Test Suite' for these routers actual tests.

      I know on wired networks anything and everything is possible due to the miriad of 'Crap' devices that spit out malformed or just plain wrong packets. (Not to mention the deliberately hacked and fudged stuff that tries to break thing as well.)

      Yet another example of not thinking through how devices will be 'Really' used rather than the 'Marketing based Daydream' that is written in all the 'glossies'.

      1. whatsyourShtoile

        Re: Google really at fault?

        If you can DoS a router with 1 cheap device sending UDP packets then something would seem to be wrong. Can't the router just drop the packets? It's UDP after all. Why would the router need to fall over and die?

        1. Roland6 Silver badge

          Re: Google really at fault?

          > Can't the router just drop the packets? It's UDP after all.

          Suspect that is the workaround the router vendors have implemented, in fact probably just a variation on a firewall rule.

  19. SquidEmperor

    My Kids also have Google Bug

    As the parent of four teenagers I'd like someone to provide a patch for their behaviour which closely resembles this report. If I dare to turn off the WiFi router I am bombarded with sudden and escalating requests which ultimately overwhelm me and can only be resolved by multiple boot-ups.

    1. GloomyTrousers

      Re: My Kids also have Google Bug

      I've looked into replicating your bug; alas I was unable to reproduce.

      (https://www.xkcd.com/583/)

    2. Anonymous Coward
      Anonymous Coward

      Re: My Kids also have Google Bug

      Squid Emperor,

      That is not a Bug ..... it is a 'Hidden' feature. :)

      Should have read the Operating Instructions/Lifecycle Diagram and the 'Teenagers' addendum ..........(very small print, on the bottom of the inside of the last page.)

      :) :)

  20. JDX Gold badge

    BT HomeHub?

    I just concluded my HH router's 2.4GHz chip had died because I kept losing that SSID (5GHz worked fine). The only thing that changed is I added a CC Audio on the 2.4GHz network.

    I concluded it was (another) crappy HH and invested £100 in a nice shiny router. Is it possible I shouldn't have? I should note the CC is hardly ever used whereas the SSID started disappearing about once a day. However I don't know if CC re-awakes periodically if not used?

    1. Lord Kipper III

      Re: BT HomeHub?

      I too had concluded that our BT Home Hub 5 was living up to its (perhaps unfair?) reputation of being terrible due to frequent Wifi problems at home preventing our teenagers from using their gadgets and me watching Star Trek on Netflix via Chromecast and I was on the cusp of binning it and buying someone else's router. First step on that BTexit journey was the £18 Openreach VDSL modem off ebay. Next step was finding a router that did not resemble a randy porcupine that Mrs Sinclair would be happy to see on the telephone table in our hall (in fairness, the Home Hub is relatively anonymous looking except when being hit by angy teenagers who can't find the reset button). As an interim, I plugged in a Home Hub 4 that I had serving up wifi in our garage and that appears to be working fine so far.

      So it would be Google and not BT I have to blame for this personalised DDOS system that out four Chromecasts are providing (three permanently on and one plugged into the TV USB port). Thanks a bundle Google, at least I didn't splash out a sizable sum for a replacement router (although that's only because I couldn't get my wife's permission to buy something that had the performance plus bells and whistles).

      Over to you Google.

    2. Roland6 Silver badge

      Re: BT HomeHub?

      >I just concluded my HH router's 2.4GHz chip had died because I kept losing that SSID (5GHz worked fine).

      Agree with your reasoning, however, it doesn't quite work on my setup. I have an EE Brightbox2 with two SSID's on each of the 2.4 and 5 frequency bands, recently one (and only one) of the 2.4 SSIDs has started to disappear (also about once a day) and then after a couple of minutes reappear... Not knowingly added any new devices...

  21. TheRealRoland
    Thumb Down

    Bought one a while back, returned it almost immediately

    i hated the fact it was constantly phoning home, and using a lot of bandwidth. Not sure if it's still the case, but it was downloading the background images it's showing while it's not casting. that could only be changed using the Android app, and not by the integrated webpages, when accessing the Chromecast by webpage.

    So then i blocked the outgoing traffic of the chromecast - and then it 'glitched' or 'hiccuped' every 20 seconds. Now i know why.

    Never again.

    1. BongoJoe

      Re: Bought one a while back, returned it almost immediately

      That was very reminiscent of our Amazon Fire Stick. It would continually download screensaver pictures and block our broadband.

  22. Andrew Jones 2

    I'm sure it sounds like a reasonable explanation from the router vendors, but there is a small problem that they are overlooking.. Chromecast (all generations), Chromecast Audio and anything running Google Assistant - DO NOT enter any sort of sleep mode. They stay awake 24/7 when powered and they broadcast mDNS messages a lot more frequently than every 20 seconds! That's how devices on the network know that they exist.

  23. Nifty

    Apple iPhoto sync - looks like same thing

    With my previous Netgear router, as soon as the iMac woke up and tried to auto sync My Photo Stream with the iDevices, a general WiFi lock-up would happen. After upgrading to an Asus router the issue seemed to disappear and possibly (though never admitting there was ever a problem of course) Apple have fixed it on their side too.

  24. sharkyblunt

    My wife has been on at me for weeks to phone up Virgin and complain because our SuperHub keeps borking, dropping the 2.4GHz and 5GHz SSIDs about once a day and only works again after a reboot. I was too busy / lazy to do so. Could it be that my expert decision not to blame Virgin was in fact an inspired one and it's our two Chromecasts that are at fault?

    1. Juan Inamillion

      I don't use Chromecast but after a couple of years of fighting with a two Virgin 'Superhubs', both of which would also bork at least once or twice a day - often more - I switched the last one to 'Modem Only', bought a Synology RT1900ac router and have had completely trouble-free wifi PLUS increased coverage for the past eighteen months.

      I've also tested (and fitted to friends networks) the TP-Link VR400 AC1200 router with equally good results.

      YMMV.

  25. Eugene Crosser
    Boffin

    MDNS just a _trigger_ for a router problem

    > It should be noted that the router vendors' fixes are mitigations while the world waits for Google to patch Cast

    This does not seem to be the case. A hundred thousand packet burst is an "unfriendly" behavior, good neighbours should not do that. But it may happen for a number of reasons, and network devices ought to be able to deal with it gracefully. Drop packets - OK. "Shut down some of router's primary functions" - not OK.

    1. Baldrickk Silver badge

      Re: MDNS just a _trigger_ for a router problem

      A hundred thousand packet burst is an "unfriendly" behavior, good neighbours should not do that.

      Well there are definitely reasons for that. what if I am transferring bulk data over the network?

  26. Alistair Silver badge
    Windows

    someone at google

    misunderstood the purpose of anacron perhaps?

  27. LaLaLand

    Bugger me! I was blaming my phone as it always appeared when I sat down to watch TV - which was off course when the Chromecase was switched on - I even posted to XDA about it https://forum.xda-developers.com/android/help/los-14-phone-started-to-kill-router-ip-t3728044

    Curse you Google - I went and bought a new phone (OK, not full reason, but this played a part)

    1. Chemist

      "Bugger me! I was blaming my phone as it always appeared "

      Well for several months after an update my wife's Android phone clobbered the Netgear router (not the access point.) After much experimentation and searching it was banned. Other Android devices like a Nexus 7 did the same. Even walking up to the house was sufficient as other people had found. It suddenly stopped ( presumably after an update ).

  28. DPepper

    Great! So Linksys engineers have time to create a patch for a Chromecast flaw, but they still haven't provided a patch for the KRACK vulnerability in my EA3500 Wi-Fi router.

  29. TrumpSlurp the Troll Silver badge
    Windows

    Back in the day

    There was this concept of "high water mark" where you stopped adding to a queue or buffer when it was nearly full.

    Kids, eh?

    1. Roland6 Silver badge

      Re: Back in the day

      There was also the concept that for certain types of messages, only the most recent were relevant, so you simply overwrote old messages. The art was selecting a queue/buffer size that meant when full there was a good chance of it holding a meaningful sequence of messages.

  30. old486whizz

    Idiocy from the routers and el reg

    So, the routers suffer from too many UDP packets.. you know, the ones which have no guarantee to be delivered.. the ones which should be dropped under high congestion or load etc...

    So the routers, instead of dropping these packets, suffer from a DoS attack and can require a reboot, and instead of blaming the routers you blame Google devices for sending out these packets "in a short amount of time".

    So mdns packets of 512-bytes, X 30,000 = 15000Kb, = 15Mb... So 15Mb of transmission data can take down a network router because it "cannot handle it" ?

    I transfer far more TCP data over the network and that is guaranteed!

    Sure, Google shouldn't be sending this data, but let's not pretend that they are sending digital nukes, they are sending digital peanuts through the mail system, which is collapsing because someone has a nut allergy!

    1. D.U.B
      Megaphone

      Re: Idiocy from the routers and el reg

      On a well planned and executed Network I would totally agree, I can run VM's stored on NFS over the wire at lag free gigabit speeds on my Lan.

      You sound savvy, so I would assume you have taken some time to ensure your network works well too.

      Here's the thing though, most folks do not.

      Most home networks I deal with on a daily basis look more like this:

      Hp printer with WiFiDirect enabled, and connect to the wireless router.

      Chromecast putting out WiFi, while connected to wireless router.

      All the Apples searching for Airplay, Airprint, etc ect

      AppleTv doing the bonjour sleep_proxy arp dance, amongst others.

      Everybody is running Dropbox Lan client, two or three print status monitors, searching for that Damn Chromecast to cast to, MdnsResponder because "If you wanna manage those iPhones with windows you gets Apple Networking with it.....the list is endless here, much of it multicast and poorly coded it seems.

      Billy's Xbox is looking for Friends and Media neighbors...and demands UPnP be found.

      That Roku is cramming more extra wifi down my throat.

      Crappy DirectTv installer misconfigured the MoCa bridge, again.

      Why is this WesternDigitalNas putting out Spanning Tree Packets.

      God Damn it, Avast is "helpfully" scanning my network devices, three times, because its on all three of the Kids laptops!!

      The Sony tv has a wifi AP built in too?...damn it!

      Three NetGear Wireless extenders and the RokuTV's still won't connect?...

      Every app on every device is open in the background, searching for local comrades and slurping all that juicy telemetry to phone home.

      What's that you say?

      Security Cams and Smart thermostats? Smart Dishwasher?

      Ports MUST be opened Immediately!!

      Pings? Pings?

      What are these Wan Pings?....

      And Fuck!! where are all these SYN packets coming from?????

      Add in the traffic you really want to push.

      Add another "100,000 packets in a short time" on top of that.

      All of a sudden that $100.00, unpatched, admin/admin, Linksys with the open WiFi bought at WalMart seems to develop a nut allergy....

      And don't even get me started on IPv6.....WiFi interference or plug' n play mesh.

      Lets face it, the Lan has been overrun by greedy Apps, Services, and Devices.

      Average users are less than clueless and lazy to boot, even adamant in their "God Given Right to not have to learn this, it says its compatible!!".

      Idiocy Indeed....

  31. gnarlymarley
    WTF?

    For some reason the 1990's network security has been lost with this comment.

    Strangely, if code I wrote is allowing someone else to abuse my product, then I am at fault for allowing such abuse. How hard is it for anyone to understand?

  32. Anonymous Coward
    Anonymous Coward

    Apple TV

    I have an older Apple TV box that I only used for about a week.

    It too would bring my network to a crawl unless I unplugged it after watching a show.

    Same with Roku that others have mentioned.

  33. Wolfclaw Silver badge

    :EEK: Google have admitted to the flaw, you hear that Intel, Apple, thats how you do it, don't bury you head in the sand until you get a class action lawsuit and worldwide PR disaster !

  34. Anonymous Coward
    Anonymous Coward

    Google devices in the home

    Take off and nuke it from orbit....

  35. robste

    My 2c, most routers are linux based and are probably based off the same open-source source code, this is why multiple vendors are affected.

    This code/library may have been patched already in later versions and router manufacturers are probably running older kernel releases because of compatibility issues with their SOC's provided by the vendor (possibly only even providing blobs instead of source code) requiring manual patching of the affected older libs to merge in later changes.

  36. rodymary

    6 google home and 2 chromecast... no more wifi at home

    I too have a lot of trouble since a few weeks

    I first suspected either one of my neighbours router overlaping my wifi channel, or a failure of my own router (I was near to buy a dedicated pro netgear router to replace the one from my internet provider (Sagem branded))

    Then I saw all your comments....

    Symptoms are the same as yours : times to times, my main 2,4Ghz SSID just disapears or stay there but is just unconnectable

    So for now I'm only using one google home, the others are just picking up the dust, waiting for an update (Google or Sagem, who will be the first ?)

    Rod

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019