IPv6 Day

[GreyGeek]

thank you ......... one more thing that can effect consistence is having to many splitters and week connections in the line between what you show in the picture and the modem.

Ya, I used to have four rooms wired for cable TV. When I dropped Cable TV I disconnected all the splitters and connected the cable directly to the cable modem, which is the only device the cable now feeds.

I learned that my rtl8192ce wireless chip does not support the 5GHz band width, even though my new Linksys E2500 does allow me to configure both bands. I reconfigured the wireless router to give the 5GHz band an ESSID of "GreyGeek5" and the 2.4GHz band an ESSID of "GreyGeek2". I saved the configuration to the wireless router but did not reboot it. I used NetworkManager's Connection Editor create a new connection, GreyGeek2. The map feature showed TWO ESSIDS, "GreyGeek2" and "GreyGeek", but I had changed that "GreyGeek" connection to "GreyGeek2", and the 5GHz band was named "GreyGeek5". I connected to GreyGeek2. When I showed the configuration window it indicated that the ESSID was "GreyGeek2" but showed "GreyGeek" as the SSID, even though GreyGeek didn't exist. The nm-tool showed that both "GreyGeek2" AND "GreyGeek" were connected APs and GreyGeek2 was giving me a 90% signal strength. My wife's Acer One 521, using a 2.4GHz ath9k driver was and continues to get 100%.

I rebooted the wireless and my Acer 7739. When they came back up I ran the nm-tool and it showed there was just one ESSID, "GreyGeek2" BUT, its signal strength was only in the 70s, and varied between 70 and 78. I can see 10 APs, and all are in the 90's except one, which matches my 74%. But, I'm 10 feet from my wireless and GarberNetgear has to be at least 100 feet from me.

Wireless Access Points (* = current AP)
    WIN_f0a6:        Infra, 2C:E4:12:46:F0:A5, Freq 2412 MHz, Rate 54 Mb/s, Strength 97 WPA WPA2
    GarberNetgear:   Infra, 2C:B0:5D:81:97:BE, Freq 2422 MHz, Rate 54 Mb/s, Strength [COLOR=#ff0000]74[/COLOR] WPA2
    linksys:         Infra, 00:13:10:89:75:0D, Freq 2437 MHz, Rate 11 Mb/s, Strength 95
    *[COLOR=#ff0000]GreyGeek2:      Infra, 20:AA:4B:3A:85:E3, Freq 2442 MHz, Rate 54 Mb/s, Strength 74 WPA2[/COLOR]
    WIN_A9E8:        Infra, 4C:17:EB:20:A9:E7, Freq 2462 MHz, Rate 54 Mb/s, Strength 97 WPA WPA2
    BettyBoop:       Infra, C0:C1:C0:24:95:3C, Freq 2412 MHz, Rate 54 Mb/s, Strength 97 WPA2
    Baby:            Infra, 08:86:3B:07:2D:87, Freq 2462 MHz, Rate 54 Mb/s, Strength 97 WEP
    5dfe:            Infra, 20:4E:7F:7A:09:02, Freq 2412 MHz, Rate 54 Mb/s, Strength 97 WPA WPA2
    NETGEAR:         Infra, E0:46:9A:6F:83:70, Freq 2437 MHz, Rate 54 Mb/s, Strength 95

I've been blaming the cable modem but on the suggestion of Dibl I checked the cable connection outside. A green platina covered the central Copper wire on one connection and the barrel connector was corroded. I replaced the barrel connector and polished the platina off of the Copper wire. I thought I had it licked when I got 90% on the mixed mode connection. If I could repeat that setup I would.

I am going to RadioShack tomorrow and get two new connectors and barrel, and make a fresh connection.

[vinnywright]

thank you ......... one more thing that can effect consistence is having to many splitters and week connections in the line between what you show in the picture and the modem.

VINNY

[GreyGeek]

@GreyGeek what command are you using to get those outputs you posted ?

VINNY

It's called "nm-tool"!

[vinnywright]

@GreyGeek what command are you using to get those outputs you posted ?

VINNY

[GreyGeek]

I think I remember reading that you are having "line-dropping" kinds of issues too -- is that correct? Because, if it is, when it happened to me it turned out to be an issue with the physical cable to the house. And of course first I had to replace the cable modem, before I was able to convince TW that there was something wrong on their side of the modem. They have an ability (if you can get to the correct geek) to test the "reflection" of a test signal to your modem, which gives an indication of the conductivity of the cable. In my case, the guy said it was obvious that the reflected signal was only half as strong as it should be, indicating a lot of excess signal loss over the cable. Also it was not stable -- it was rising and falling in strength as he was observing it. I got a new cable in a few days, and among other things the installer found that a cover/shield on the utility box had been installed in such a way as to pinch the cable to my house, probably compromising the insulation and/or or the shield layer, but not the center conductor.

I checked that cable connection I posted a photo of and found that the Copper center wire on the service side had a green patina on it. :eek:. And, the Aluminum barrel connector between the two plugs was very oxidized. I replaced the barrel connector with a gold plated one and I used a plastic scratch pad to get bare Copper on the center wire. After putting it all together I went in and turned on the modem and wireless. BOOM! I got ESSID's for "GreyGeek" and "GreyGeek2", but not "GreyGeek5". I think GreyGeek is coming from a cache. In checking, both appear to be used because "GreyGeek2" is listed as the ESSID but "GreyGeek" is listed as the SSID.

  Wireless Access Points (* = current AP)                                                                                                                                  
    linksys:         Infra, 00:13:10:89:75:0D, Freq 2437 MHz, Rate 11 Mb/s, Strength 95                                                                                    
    GarberNetgear:   Infra, 2C:B0:5D:81:97:BE, Freq 2422 MHz, Rate 54 Mb/s, Strength 85 WPA2                                                                               
    NETGEAR:         Infra, E0:46:9A:79:5B:2E, Freq 2412 MHz, Rate 54 Mb/s, Strength 90 WEP                                                                                 
    5dfe:            Infra, 20:4E:7F:7A:09:02, Freq 2412 MHz, Rate 54 Mb/s, Strength 77 WPA WPA2                                                                            
    WIN_f0a6:        Infra, 2C:E4:12:46:F0:A5, Freq 2412 MHz, Rate 54 Mb/s, Strength 95 WPA WPA2                                                                            
    WIN_A9E8:        Infra, 4C:17:EB:20:A9:E7, Freq 2462 MHz, Rate 54 Mb/s, Strength 74 WPA WPA2                                                                            
    belkin.fec:      Infra, 08:86:3B:29:8F:EC, Freq 2462 MHz, Rate 54 Mb/s, Strength 95 WPA WPA2
    Baby:            Infra, 08:86:3B:07:2D:87, Freq 2462 MHz, Rate 54 Mb/s, Strength 92 WEP
    The Alois Family:Infra, 00:1E:58:32:BB:35, Freq 2412 MHz, Rate 54 Mb/s, Strength 90 WPA
  [COLOR=#ff0000]  *GreyGeek2:      Infra, 20:AA:4B:3A:85:E3, Freq 2442 MHz, Rate 54 Mb/s, Strength 90 WPA2
    *GreyGeek:       Infra, 20:AA:4B:3A:85:E3, Freq 2442 MHz, Rate 54 Mb/s, Strength 77 WPA2[/COLOR]
    PS3-9210729:     Infra, 8C:7C:B5:3C:90:B4, Freq 2437 MHz, Rate 54 Mb/s, Strength 90 WPA

My signal strength is 90%, which is much better than 60%.

I fired up my wife's Acer One, also running Kubuntu Precise, and she reported 100% signal strength. The wireless chip in her box is the AR9285 and it is driven by the ath9k. It cannot connect to the 5GHz band either.

... the following info is about 30 minutes later... things don't look so good ..........

I closed the GreyGeek2 connection and deleted it. Then I created it again. The scan now shows:

  Wireless Access Points (* = current AP)
    linksys:         Infra, 00:13:10:89:75:0D, Freq 2437 MHz, Rate 11 Mb/s, Strength 95
    WIN_f0a6:        Infra, 2C:E4:12:46:F0:A5, Freq 2412 MHz, Rate 54 Mb/s, Strength 94 WPA WPA2
    5dfe:            Infra, 20:4E:7F:7A:09:02, Freq 2412 MHz, Rate 54 Mb/s, Strength 90 WPA WPA2
    GarberNetgear:   Infra, 2C:B0:5D:81:97:BE, Freq 2422 MHz, Rate 54 Mb/s, Strength 89 WPA2
    WIN_A9E8:        Infra, 4C:17:EB:20:A9:E7, Freq 2462 MHz, Rate 54 Mb/s, Strength 95 WPA WPA2
    BettyBoop:       Infra, C0:C1:C0:24:95:3C, Freq 2412 MHz, Rate 54 Mb/s, Strength 90 WPA2
    belkin.7da:      Infra, 08:86:3B:16:E7:DA, Freq 2417 MHz, Rate 54 Mb/s, Strength 84 WPA WPA2
    belkin.fec:      Infra, 08:86:3B:29:8F:EC, Freq 2462 MHz, Rate 54 Mb/s, Strength 84 WPA WPA2
   [COLOR=#ff0000] *GreyGeek2:      Infra, 20:AA:4B:3A:85:E3, Freq 2442 MHz, Rate 54 Mb/s, Strength 74 WPA2[/COLOR]

and GreyGeek no longer shows and GreyGeek5 doesn't show because my rtl8192ce chip isn't 5GHz compatible.
I rebooted the wireless and only GreyGeek2 shows.

  Wireless Access Points (* = current AP)
    linksys:         Infra, 00:13:10:89:75:0D, Freq 2437 MHz, Rate 11 Mb/s, Strength 95
    WIN_f0a6:        Infra, 2C:E4:12:46:F0:A5, Freq 2412 MHz, Rate 54 Mb/s, Strength 94 WPA WPA2
    5dfe:            Infra, 20:4E:7F:7A:09:02, Freq 2412 MHz, Rate 54 Mb/s, Strength 94 WPA WPA2
    WIN_A9E8:        Infra, 4C:17:EB:20:A9:E7, Freq 2462 MHz, Rate 54 Mb/s, Strength 97 WPA WPA2
    belkin.fec:      Infra, 08:86:3B:29:8F:EC, Freq 2462 MHz, Rate 54 Mb/s, Strength 84 WPA WPA2
    Baby:            Infra, 08:86:3B:07:2D:87, Freq 2462 MHz, Rate 54 Mb/s, Strength 97 WEP
    GarberNetgear:   Infra, 2C:B0:5D:81:97:BE, Freq 2422 MHz, Rate 54 Mb/s, Strength 92 WPA2
   [COLOR=#ff0000] *GreyGeek2:      Infra, 20:AA:4B:3A:85:E3, Freq 2442 MHz, Rate 54 Mb/s, Strength 75 WPA2[/COLOR]
    The Alois Family:Infra, 00:1E:58:32:BB:35, Freq 2412 MHz, Rate 54 Mb/s, Strength 89 WPA
    BettyBoop:       Infra, C0:C1:C0:24:95:3C, Freq 2412 MHz, Rate 54 Mb/s, Strength 89 WPA2

It looks like my connection strength is still yo-yo-ing, but it is better than before. My next task is to cut off the two connectors and show fresh cable, insulation and shielding, and add new hardware. I wonder how I got 90% when I was cross-connected with GreyGeek2 and "GreyGeek". IF I could duplicate that and get 90% back that would be neat.

[GreyGeek]

On my journey trying to get a 10/10 IPv6 rating, I looked into the DNS hosting of SiXS. They will not allow using their DNS facility unless you are accessing it through one of their tunnels. EVEN IF you are using one of their tunnels they do not recommend using their DNS if you are using a wireless connection (i.e., using aicuu and not their static IP tool).

So, for wireless connections, which I must use, SixXS offers no DNS ability. Ergo, IMO, the miredo tunnel is as good as SixXS, so I plan to remove my SixXS tunnel and just use the miredo. I'd get a 9/10 anyway, so what's to lose?

Anyone think of reasons why I should stay with SixXS?

[GreyGeek]

Just installed miredo and went to http://test-ipv6.com/ and these are my results.

Added:
Just signed up for my SixXS IPv6 account. Awaiting the 'human' verification of my registration email response. Thinking about the results above, I wonder if a SixXS IPv6 account is even necessary. Guess it won't hurt to have it. I still don't get the color Star Wars console test.

Snowhog, did you hook up with SixXS yet? IF so, has your 10/10 score changed when you use the sixxs device and not teredo?

[GreyGeek]

Unfortunately, it is. But, that cable has been connected ONLY to our modem for Internet use, since we dropped cable tv years ago.

I think I will put new connectors on fresh ends of the cables and join them, then weather proof the joint.

[dibl]

That does not appear to be a weatherproof connector, and the cable looks pretty wimpy too. Are you sure that cable is carrying your internet service? If so, you're lucky it lasted that long. Mine is a fatter orange cable, and carries the TV/ISP "bundle" from a service box in the back of my property. There are no such joints on it. It comes to a weatherproof box on the back of my house where the internet service is split off the CATV service.

[GreyGeek]

You may be on to something there, Dibl. I went outside and noticed that a root of a tree along the path where the cable was laid was pushing out of the ground next to the trunk, and embedded in the root, almost entirely, was a cable. But, it was too big for the typical coax and is probably the fiber optic cable that was laid in the early 1990's and is still dark.


I followed the cable to the pedestal and saw this:


All out in the open, no protection from rain, dust, ice, etc...

[GreyGeek]

You may be on to something there, Dibl. I went outside and noticed that a root of a tree along the path where the cable was laid was pushing out of the ground next to the trunk, and embedded in the root, almost entirely, was a cable. But, it was too big for the typical coax and is probably the fiber optic cable that was laid in the early 1990's and is still dark.


I followed the cable to the pedestal and saw this:


all out in the open, no protection from dust, dirt, rain, ice or snow. The last time that connection was touched was when we dropped cable tv five years ago.

[dibl]

... mine has the lowest signal strength at 67%, and I am only 10' from my wireless router! This is the same problem that my other wireless, the WR1043ND, gave me. I am beginning to suspect that it is a problem with the cable modem from TimeWarner.

I think I remember reading that you are having "line-dropping" kinds of issues too -- is that correct? Because, if it is, when it happened to me it turned out to be an issue with the physical cable to the house. And of course first I had to replace the cable modem, before I was able to convince TW that there was something wrong on their side of the modem. They have an ability (if you can get to the correct geek) to test the "reflection" of a test signal to your modem, which gives an indication of the conductivity of the cable. In my case, the guy said it was obvious that the reflected signal was only half as strong as it should be, indicating a lot of excess signal loss over the cable. Also it was not stable -- it was rising and falling in strength as he was observing it. I got a new cable in a few days, and among other things the installer found that a cover/shield on the utility box had been installed in such a way as to pinch the cable to my house, probably compromising the insulation and/or or the shield layer, but not the center conductor.

[SteveRiley]

My pleasure.

[GreyGeek]

That clears up a lot of fog in my brain about IPv6!!! Thanks, Steve!

[SteveRiley]

Don't know what those two "unreachable" defaults are, but I'm assuming that there can be only one default.

It's intentional. Remember that in IPv6, automatic address assignment doesn't include a default gateway. Instead, IPv6 issues neighbor discovery probes, looking for routers on the same link as your interface. As the stack finds routers, it builds a routing table.

Now imagine that something happened, and all your neighbor routers died. Your routing table would be empty, and applications would send datagrams, receive no responses, and just keep trying over and over again. To prevent this, IPv6 automatically creates a default "unreachable" route on each local interface and assigns it a metric of -1, which means "always least preferred" -- you can see this in your ip r s t all output.

Under ordinary situations, when your computer has found a router, its entry in your route table will include a better metric, and "unreachable" is ignored. But when all your links die, and there are no routers to reach, your IPv6 stack falls back to the "unreachable" route. Thus, it can inform applications that the destination is, well, unreachable. At this point, it's up to the application to determine what to do, but at least it doesn't have to sit there and keep guessing all day.

You'll notice that in the "nm-tool" listing are the other "AP"'s local to me, and among them, mine has the lowest signal strength at 67%, and I am only 10' from my wireless router! This is the same problem that my other wireless, the WR1043ND, gave me. I am beginning to suspect that it is a problem with the cable modem from TimeWarner.

See my reply in the other thread. I was asking where you got that list from; now I know. I also wrote some more about signal strength, wireless frequencies, and NIC capabilities.