The Laserjet router



  • I recently posted on this:

    http://forums.thedailywtf.com/forums/p/25688/280590.aspx#280590

    With a some WTF from the Uni I used to work at, there were many more:

     

    University WTF #4.

    One memorable WTF that springs to mind is the case of the disappearing router.

    Every
    now and again we would suffer network outages, connections momentarily
    lost to servers causing untold trouble across the department. Students
    where having trouble logging on because the loading of profiles would be
    interrupted, connections to the exchange server would just come and go.
    Jobs running on one machine would stop because they lost contact with
    data files located on another.


    The network belonged to central IT and I wasn't allowed to fiddle
    with it, besides at this time I had my own work to do. My boss was on
    the case so I left him to it.

    It became apparent that the problem
    was worst when we had lots of students in the building, logging on in
    the labs. My boss thought whatever the problem was, it was caused by
    traffic. He started by abandoning roaming profiles in an attempt to
    reduce traffic at peak times. It didn't work.


    Next he decided, having done everything he could do, that it
    obviously wasn't his fault and it must be down to central IT and their
    shoddy network. So he reported that we had a fault with some of our
    switches.


    Central IT dutifully changed the blades out, the problem didn't go away.

    He kicked up a stink, they changed the whole chassis. To no avail.

    Eventually
    he even got them to swap out a very expensive router, which of course,
    there was nothing wrong with in the first place. The problem persisted.


    In the end he had a team of people from around the department and
    from central IT sitting in the server room monitoring packets, poking
    the router and switches and for the most part scratching heads.

    After
    months of this intermittent network problem I decided I would have to
    step in to make my life liveable, I was being bombarded with
    reports of network problems and being asked to fix it by the end users.
    It was getting embarrassing to keep telling them "Uuhhh, we dunno what's
    wrong, but the boss is trying to fix it."


    So I sat down and did what they should have been doing months ago, work out exactly what the problem is.

    It
    was obvious that network connections were getting interrupted, was this
    a big problem? Localised to certain machines or a switch? What was the
    nature of the beast?


    I did a very simple test, I fired up some computers plugged into
    various subnets, connected to a network share and played an mp3 in media
    player on a loop.

    I went back to doing my work and waited for
    the music to stop, or at least skip. My first task was to witness these
    outages for myself and try and work out where the point of failure was
    before even thinking about why.


    It soon became apparent that the fault only occurred when
    communicating across subnets, I confirmed this by writing a quick app to
    fire off pings and plot a graph of the pings returned. Sure enough,
    machine to machine on the same subnet in all cases was flawless. The
    network never lagged, this was not a traffic issue and it never dropped a
    packet. The issue was only across subnets, so, we have a problem
    involving the router.


    Every now and again it was like packets destined for the router just
    disappeared, the router itself seemed fine, we had already changed the
    hardware once and it couldn't have been a software issue as we had many
    similar routers with the same firmware running across campus and never
    had a problem.


    In the end it was my network monitoring that revealed the problem,
    you remember that system I designed to pick up rouge kit on our network?

    The
    router, x.y.z.1 obviously had to receive packets from machines on each
    subnet and forward them accordingly, as anyone in IT knows this is done
    via ARP. Machines ask "Excuse me, who is x.y.z.1?" and other machines
    reply "I sir am x.y.z.1 and my MAC address is aa-bb-cc-dd-ee-ff, please
    send your packets to this address."


    Most of the time, the router would claim to be x.y.z.1 but every now
    and again a strange foreign MAC address would claim to be x.y.z.1. The
    cause of the problem had been identified, all I had to do now was track
    down why this was happening.


    Fortunately the MAC address was familiar so I didn't have to wrangle
    with central IT to query the switches as to its physical location, it
    belonged to a JetDirect card in a printer in one of the student labs.


    I visited the lab and lots of students where complaining that the printer wasn't working properly, so I started to look at it. I requested the usual test/diagnostics page from the printer and it revealed something very interesting...


    Address assigned by DHCP, address x.y.z.1, assigned by server..... oh dear.

    This
    was not the campus DHCP server, someone else had a DHCP server
    somewhere and it was giving out the IP address of the router to this
    printer, how odd.


    I tracked the rouge server down to our very own server room, on a
    machine run by my boss. For months he had been blaming everyone else and
    making stupid stab in the dark fixes and all the time he was causing
    the problem himself by misconfiguring a DHCP he setup to network boot
    some kit.


    Thousands of pounds worth of equipment had been changed, hundreds of
    man hours lost, shit flung back and forth from fan to fan. That made it
    all the more satisfying to grab a highlighter pen, highlight the
    address of the rouge DHCP and march into his office.


    The reason the network dropped out more when we had students in the
    labs was simple, they couldn't print so they kept turning the printer
    off and on. Every time they did, for a fraction of a second the printer
    would claim to be the router and packets would get sent to the printer,
    until the router responded itself and the printer registered a conflict
    and backed off.

     



  •  TRWTF being that any organization with a sensible IT department would change out an entire server chassis and router setup without him having even attempted to diagnose where the traffic was being lost.



  • Similar situation at a former workplace: a printer in an office would fail to function at some times, so people just used another printer as a workaround and all was fine.

    We eventually spotted a pattern revolving around a certain field-sales droid, that it occured when he was in the building. Eventually we tracked it down to him hard-coding an IP address which conflicted with the IP reserved for the jet-direct card; flicked it back to DHCP and it all worked.

    So, why that IP? Apparently he came in one morning and found he couldn't print, so spotted that IP on the JD card and popped it into his network settings. Then, once he was on the "192.168.20.5" network, he could access the printer (nearest him, but nobody could use the high-volume printer over the other side of the office).

    Yeah, TRWTF was security settings that permitted him to make those changes. But policy dictated that any restriction on sales droids meant a potential lost sale, so we were NOT to touch those holiest of holies on their pedestal.


  • Discourse touched me in a no-no place

    @EncoreSpod said:

    In the end it was my network monitoring that revealed the problem,
    you remember that system I designed to pick up rouge kit on our network?

    @EncoreSpod said:

    I tracked the rouge server down to our very own server room, on a
    machine run by my boss. For months he had been blaming everyone else and
    making stupid stab in the dark fixes and all the time he was causing
    the problem himself by misconfiguring a DHCP he setup to network boot
    some kit.


    Thousands of pounds worth of equipment had been changed, hundreds of
    man hours lost, shit flung back and forth from fan to fan. That made it
    all the more satisfying to grab a highlighter pen, highlight the
    address of the rouge DHCP and march into his office.

    Minor nitpick - that word doesn't mean what you think it does. This is a rouge printer:






    Rogue printers on the other hand may, or may not, be that colour.



  • @PJH said:

    Minor nitpick - that word doesn't mean what you think it does.
     

     

    Hey I said I could ping stuff I never claimed I could spell :D

     



  • We had a similar routing problem at our office last year. After booting your computer, it was a crap shoot if you'd end up on our 10.10.10/24 subnet or our 192.168.1/24 subnet. The 10.10.10/24 subnet wouldn't connect to any of our servers or let you out onto the internet which was problematic for our users. These networks were supposed to be physically separate (one VoIP network, one computer network using completely separate routers and switching hardware) but somehow had gotten bridged. After 6 months of this problem continuing (I initially found a workaround through "arp -d *" from the cli), I gave up asking IT to solve the problem which affected 1/3 - 1/2 of our users daily and interrogated the switches myself. In about an hour I found the bridge; somebody in our graphic arts department had decided the network cable that was only sticking out of one port in the wall should be plugged into both ports in the wall ...



  • $ ping rooter

    ping: unknown host rooter



  • Hmm, never had a laserjet router before... I did recently encounter (through printer upgrades, an expired DHCP lease, and the resulting IP address collision) that one of my colleagues had a laserjet cell phone.

    Apparently Android can do more than I thought. :)



  • When I was in high school/college, one of the guys in my floor plugged his router in wrong, causing this exact thing... the router started acting as a DHCP server to the entire campus.  They had to cut off our building until they could notify him.



  • @Master Chief said:

     TRWTF being that any organization with a sensible IT department would change out an entire server chassis and router setup without him having even attempted to diagnose where the traffic was being lost.

     

    This.

    My first thought was to check links for errors.  When I hear about network slowness, I immediately go to our interface monitoring system and check to see if there are errors on the uplinks (well, after checking the device's network port). That's the most common problem for me.  You've got to start at layer 1 and work up.

    Cisco equipment (probably others -- my strongest experience is with Cisco) has DHCP snooping that will prevent unauthorized DHCP servers from affecting things on the VLAN.  We've have some rogue DHCP servers before.  Usually they're pretty easy to find because they're not issuing addresses on our subnet . . . they issue them with the 192.168.0 or 192.168.1 address that is their default subnet.  If this scenario had x = 192 and y = 168 in the anonymized addresses, then I can see where it would have been more difficult to find.

    But hauling off and replacing routers and switches without doing real engineering work is TRWTF.

     



  • @rad131304 said:

    somebody in our graphic arts department had decided the network cable that was only sticking out of one port in the wall should be plugged into both ports in the wall ...
     

    I had to beat someone over the head with a ClueBat(tm) a couple of years ago when this was happening.  Somehow they thought it was a good idea to take the cable that was lying on the floor and plug it into a spare jack so that it wouldn't be hanging.  I told them they needed to remove the cable when not in use or put a hook on the wall.  Fortunately, one of the VLANs that was involved in this bridge was very lightly used -- usually one device at a time -- so the effect was largely cosmetic (boatloads of "wrong native VLAN" from CDP log messages) instead of functional.



  • EncoreSpod you did a good job writing your story despite people nitpicking.  You did an excellent build up and I could see this being a main article.



  • @Anketam said:

    EncoreSpod you did a good job writing your story despite people nitpicking.  You did an excellent build up and I could see this being a main article.

    It couldn't possibly appere on the front paige without those typoes.



  • Boomzilla, your wanking stamina is quite impressive.



  • @Anketam said:

    EncoreSpod you did a good job writing your story despite people nitpicking.  You did an excellent build up and I could see this being a main article.


    +1



  • I'm disappointed. I thought it was going to be some genius who realised that a modern printer has an embedded PC and decided to save money by using it for routing too.



  • @Anketam said:

    EncoreSpod you did a good job writing your story despite people nitpicking.  You did an excellent build up and I could see this being a main article.

    I agree except for the title which pretty much telegraphed the punchline, detracting from an otherwise great read.



  • @LoztInSpace said:

    @Anketam said:

    EncoreSpod you did a good job writing your story despite people nitpicking.  You did an excellent build up and I could see this being a main article.

    I agree except for the title which pretty much telegraphed the punchline[. . .]

    Sounds like normal front-page article to me.


  • I've seen both at work (a secondary school)  The cleaning staff saw a dangling cable and plugged the other end into the wall, too*.  Our entire network went down while I tried to find a serial cable and console in.  The primary ProCurve didn't have spanning tree enabled.  It does now.

    *This was my first or second guess, but I had to find the darn thing.

    We share a network with a Career Center.  Their IT training lab is the Wild West, and we'd prefer to physically and mentally wall them off, but they need access to our servers.  One of their 'servers' decided to pass out IP addresses to anyone who asked.  LaserJet phones were the least of it.

    This is the same teacher that was doing 15Mb/s on our 20MB/s line from one torrent.  He has been throttled, but not the way I wanted to.

    I made a visit to the wood shop, and now have a clue-by-four behind my desk. 



  • I used to live in a share house with cable internet. One day I was playing around with my old Linux laptop and managed to kill wifi, so I trundled out to the router and plugged in with cat5. After getting my wifi to work I accidentally left the cat5 cable there, dangling on the floor. Someone (it could have been the landlady) plugged it into the wall. Unfortuneatly that was the phone socket, not a network socket. It was a 100Mbps router so the middle pins were shorted so it killed the phone line. (I mentioned cable because if it were DSL this action would have killed the net too. Us lowly tennants didn't have access to the phone as it was in her room. That didn't affect me since mobile phones were/are common, but back then mobile internet wasn't)

    Landlady was mad because she called out the phone company, and since it wasn't their fault they charged a callout fee. I refused to pay/contribute because I didn't plug it in. No-one else confessed. I ended up just leaving soon after.

     



  • @BPFH said:

    Apparently Android can do more than I thought.
    I remember reading somewhere that Android is (was?) pretty broken when it came to DHCP (keeps responding on the old IP after it gets a new lease).
    @Sutherlands said:
    When I was in high school/college, one of the guys in my floor plugged his router in wrong, causing this exact thing... the router started acting as a DHCP server to the entire campus.  They had to cut off our building until they could notify him.
    Apparently somebody did this at my old cable ISP at one time, bringing a whole subnet down.


  • Discourse touched me in a no-no place

    @ender said:

    I remember reading somewhere that Android is (was?) pretty broken when it came to DHCP (keeps responding on the old IP after it gets a new lease).
    http://www.net.princeton.edu/android/android-stops-renewing-lease-keeps-using-IP-address-11236.html#issue

    Under some circumstances, a device running one of the affected versions of Android stop renewing its DHCP lease, yet continues (or resumes) using the IP address after the lease expires.

    We've seen this problem on our networks.


  • This happened on our network once:

    One day, several users stood in front of my door, claiming that they could not connect to the internet anymore. My own connection worked fine, however.

    After having confirmed that it was not simply a case of PEBKAC, I noticed that the computers in question indeed got a valid IP adress - but not a valid DNS adress. And after having done a DHCP lease renewal, my own desktop got the wrong DNS server as well.

    Went over to our router and took a look at what was actually sent on our network - and noticed that there seemed to be a rogue router around which managed to answer to DHCP broadcasts a bit faster than our router. At that time, we did not have managed switches everywhere yet, so we had to use the time-honored method of unplugging cables until we found the guilty party.

    Which turned out to be a very blonde person, so I don't think that it was a conscious act on her part.

    I'm still a bit confused, however, how her laptop came to be configured thus that it gave every computer on the network the correct IP adress (they're statically linked to the MAC) but not a correct DNS.

    Which reminds me: Anyone interested in a "packet of doom"? Stumbled across that one when a resident's PC bluescreened and hardlocked into a weird crash state where it perpetually sent out a packet which crashed our Netgear managed switches on the first attempt. That one was quite interesting - because the switches crashed in such a way that they acted as if they were working (port lights alight/blinking), however, when you pulled out all the network cables, the port status lights were still on and blinking...



  • @Rhywden said:

    Which reminds me: Anyone interested in a "packet of doom"? Stumbled across that one when a resident's PC bluescreened and hardlocked into a weird crash state where it perpetually sent out a packet which crashed our Netgear managed switches on the first attempt. That one was quite interesting - because the switches crashed in such a way that they acted as if they were working (port lights alight/blinking), however, when you pulled out all the network cables, the port status lights were still on and blinking...

    Not sure if this counts, but for a while, whenever I tried to turn on wifi on my Droid, my wifi router would reboot itself! Finally solved the problem by switching from WPA2 with mixed mode encryption to WPA with TKIP encryption... no idea why that fixed it though!



  • @ekolis said:

    Not sure if this counts, but for a while, whenever I tried to turn on wifi on my Droid, my wifi router would reboot itself! Finally solved the problem by switching from WPA2 with mixed mode encryption to WPA with TKIP encryption... no idea why that fixed it though!

    I've got something similar - whenever I enable wifi on my phone at home, everything works ok for a while, but invariably the network eventually dies and nothing, including the phone, can get a wireless connection. Power cycling the router fixes it, for a while...



  • @Zemm said:

    I used to live in a share house with cable internet. One day I was playing around with my old Linux laptop and managed to kill wifi, so I trundled out to the router and plugged in with cat5. After getting my wifi to work I accidentally left the cat5 cable there, dangling on the floor. Someone (it could have been the landlady) plugged it into the wall. Unfortuneatly that was the phone socket, not a network socket. It was a 100Mbps router so the middle pins were shorted so it killed the phone line. (I mentioned cable because if it were DSL this action would have killed the net too. Us lowly tennants didn't have access to the phone as it was in her room. That didn't affect me since mobile phones were/are common, but back then mobile internet wasn't)

    Landlady was mad because she called out the phone company, and since it wasn't their fault they charged a callout fee. I refused to pay/contribute because I didn't plug it in. No-one else confessed. I ended up just leaving soon after.

     

    Why was your home phone socket RJ45?



  • @morbiuswilters said:

    Why was your home phone socket RJ45?
    ISDN?



  • Why was your home phone socket RJ45?

    That's exactly what I was wondering. You can plug an RJ-11 into an RJ-45, but not the other way around.



  • @Gazzonyx said:

    Why was your home phone socket RJ45?

    That's exactly what I was wondering. You can plug an RJ-11 into an RJ-45, but not the other way around.

    Ok me too! Thank God I was going to post but I assumed I missed something.

    All I can think of is VOIP "phone lines", but in that case you'd be plugging ethernet into ethernet and no harm would be done.



  • @blakeyrat said:

    @Gazzonyx said:
    Why was your home phone socket RJ45?

    That's exactly what I was wondering. You can plug an RJ-11 into an RJ-45, but not the other way around.

    Ok me too!


    I wondered that, too!



  • @Gazzonyx said:

    Why was your home phone socket RJ45?
    That's exactly what I was wondering. You can plug an RJ-11 into an RJ-45, but not the other way around.

    Sure you can, it just may take a hammer to get it seated in the socket.



  • @morbiuswilters said:

    @blakeyrat said:
    @Gazzonyx said:
    Why was your home phone socket RJ45?

    That's exactly what I was wondering. You can plug an RJ-11 into an RJ-45, but not the other way around.

    Ok me too!


    I wondered that, too!
     

    It is a Telstra WTF installing RJ45 sockets (technically 8P8C). They stopped installing the old "600 series" socket in the late 1980s and used RJ12 for a little while, now all new phone sockets get RJ45. Info here. Nice way to avoid confusion with ethernet!



  • @barrabus said:

    @ekolis said:
    Not sure if this counts, but for a while, whenever I tried to turn on wifi on my Droid, my wifi router would reboot itself! Finally solved the problem by switching from WPA2 with mixed mode encryption to WPA with TKIP encryption... no idea why that fixed it though!

    I've got something similar - whenever I enable wifi on my phone at home, everything works ok for a while, but invariably the network eventually dies and nothing, including the phone, can get a wireless connection. Power cycling the router fixes it, for a while...

     

    Same sort of odd problem here, only mine was true twilight zone material.

    Everything would work fine... until I booted up Sid Meier's Pirates (the latest remake). Not only my network connection would drop when I did that, oh no no... every network connection in the house would drop. But ONLY when playing that game, not when playing something as retardedly heavy duty as Crysis. I wouldn't have been bothered with it if I didn't love Pirates so much.

    I use one of those USB sticks to provide a wifi connection for my computer, so I figured there was some kind of conflict with it. I replaced it. Things improved slightly as the network connection stayed up a little longer, but after about half a minute it dropped anyway.

     

    As a developer I have learned to reason that when you start to describe problems as "weird", "strange" or "impossible", it usually something so tiny and silly you overlook it as a way of protecting your own ego. The router happened to be on my computer desk; the computer is in an opening inside the desk, under the router. While the network connection was down, I simply picked the router up... and the connection came back. The fact that the router was physically ON the desk made it fail while playing that game and that game only. It is now on a small table NEXT to my desk and no more problems since.

    What it could be? The only thing I could come up with is vibrations. I optimized my hardware for low heat output so it usually is very silent and the fans are running really low, even when playing AAA games. But while playing Pirates I noticed that the CPU fan started to run at a high speed; it probably has some sucky kind of busy waiting game loop; the game is years old and should not make any kind of dent on CPU usage. That is really the only thing I could think of to explain this oddity :/



  • @morbiuswilters said:

    @blakeyrat said:
    @Gazzonyx said:
    Why was your home phone socket RJ45?

    That's exactly what I was wondering. You can plug an RJ-11 into an RJ-45, but not the other way around.

    Ok me too!


    I wondered that, too!
    Apparently this is how they are doing it in new houses. My friend's house POTS lines are run on CAT5e with RJ45 wall terminations all connected to the 5V telephone block in the basement. He, like many now, doesn't have a land line and was trying to figure out why he couldn't just plug his FiOS router into it to get wired internet throughout his house.



  • @rad131304 said:

    @morbiuswilters said:
    @blakeyrat said:
    @Gazzonyx said:
    Why was your home phone socket RJ45?

    That's exactly what I was wondering. You can plug an RJ-11 into an RJ-45, but not the other way around.

    Ok me too!


    I wondered that, too!
    Apparently this is how they are doing it in new houses. My friend's house POTS lines are run on CAT5e with RJ45 wall terminations all connected to the 5V telephone block in the basement. He, like many now, doesn't have a land line and was trying to figure out why he couldn't just plug his FiOS router into it to get wired internet throughout his house.

    New houses have POTS lines? Seriously, though, that's just fucked up. Telephone can be run on Cat5e but the jacks should not be RJ45. How the hell do you even plug into that? You'd have to have a special RJ11 to RJ45 cable to connect anything.



  • @morbiuswilters said:

    How the hell do you even plug into that? You'd have to have a special RJ11 to RJ45 cable to connect anything.
    You've never seen RJ11 or RJ12 crammed to RJ45?



  • We have a VoIP system.

    But the analog faxes are connected this way: Analog RJ45 port in the (digital) PBX is connected to a CAT5e patch panel leading to a network rack, daisy-chained through one or more of these racks until you reach the one that contains patchpanel for the appropriate room, and finally to an RJ45 wall socket. The RJ11 cable for the fax locks neatly in the center of the RJ45 socket, as they were designed.

    What'd be a better and more cost-effective alternative in your opinion, assuming you can't get rid of the faxes and they have to be operational and scattered all around a big building?



  • @morbiuswilters said:

    How the hell do you even plug into that? You'd have to have a special RJ11 to RJ45 cable to connect anything.

    Telephone lines fit fine in ethernet ports.  There's just some space on either side.


  • Discourse touched me in a no-no place

    @Rhywden said:

    Which reminds me: Anyone interested in a "packet of doom"? Stumbled across that one when a resident's PC bluescreened and hardlocked into a weird crash state where it perpetually sent out a packet which crashed our Netgear managed switches on the first attempt. That one was quite interesting - because the switches crashed in such a way that they acted as if they were working (port lights alight/blinking), however, when you pulled out all the network cables, the port status lights were still on and blinking...
    Do you have a capture of that packet?

    I use managed Netgears in my home infrastructure (lol megacheap refurbs!)  and have seen this exact state once. Never did figure out what the hell happened.



  • @ender said:

    @morbiuswilters said:
    How the hell do you even plug into that? You'd have to have a special RJ11 to RJ45 cable to connect anything.
    You've never seen RJ11 or RJ12 crammed to RJ45?

    My ADSL filter has RJ45 sockets and plastic inserts so it will accept a RJ12 plug without being "jammed".



  • @erikal said:

    USB sticks to provide a wifi connection for my computer

    @erikal said:
    The router happened to be on my computer desk; the computer is in an opening inside the desk, under the router.

    What are you doing with a wifi connection with the router on your desk? A USB one no less? Run out of Ethernet ports?



  • @Weng said:

    @Rhywden said:

    Which reminds me: Anyone interested in a "packet of doom"? Stumbled across that one when a resident's PC bluescreened and hardlocked into a weird crash state where it perpetually sent out a packet which crashed our Netgear managed switches on the first attempt. That one was quite interesting - because the switches crashed in such a way that they acted as if they were working (port lights alight/blinking), however, when you pulled out all the network cables, the port status lights were still on and blinking...
    Do you have a capture of that packet?

    I use managed Netgears in my home infrastructure (lol megacheap refurbs!)  and have seen this exact state once. Never did figure out what the hell happened.

    I meant to ask - are you sure it was sending a packet, and not just over-volting the line? If I recall correctly, CMOS latch-up is commonly caused by slight over-voltage - and Netgear kit is known to be particularly vulnerable thanks to much of it being low-end stuff.



  • @Weng said:

    @Rhywden said:

    Which reminds me: Anyone interested in a "packet of doom"? Stumbled across that one when a resident's PC bluescreened and hardlocked into a weird crash state where it perpetually sent out a packet which crashed our Netgear managed switches on the first attempt. That one was quite interesting - because the switches crashed in such a way that they acted as if they were working (port lights alight/blinking), however, when you pulled out all the network cables, the port status lights were still on and blinking...
    Do you have a capture of that packet?

    I use managed Netgears in my home infrastructure (lol megacheap refurbs!)  and have seen this exact state once. Never did figure out what the hell happened.

    [url= http://www.amazon.com/gp/aw/cr/B00008SCFL/ref=aw_d_cr_electronics]Here[/url] is a Netgear router with a review that mentions this exact symptom. One star review from Matthew on March 26, 2005 talking about how he can open too many TCP connections and it stops routing packets even though das Blinkenlichten keep Blinken.



  • @Zemm said:

    @erikal said:
    USB sticks to provide a wifi connection for my computer

    @erikal said:
    The router happened to be on my computer desk; the computer is in an opening inside the desk, under the router.

    What are you doing with a wifi connection with the router on your desk? A USB one no less? Run out of Ethernet ports?

     

    You make a very valid point - it is for two lame reasons;

    - my pc used to be on the top floor (away from the router) and I didn't change it when I moved it downstairs

    - it is to prevent yet another inflexible cable from running behind my desk :)

     


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.