1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

IPv6rd Not Working (Centurylink)

Discussion in 'Tomato Firmware' started by JDog2pt0, Dec 20, 2017.

  1. JDog2pt0

    JDog2pt0 Network Newbie Member

    I am running Advanced Tomato 3.4-140 (shibby 140) on my Nighthawk R7000, bridged to my Technicolor C2100T. Centurylink offers IPv6 through 6rd, and though I've got it properly set up in the R7000, it doesn't work. By doesn't work I mean, I've got no IPv6 connection, period. I can't even ping IPv6 addresses from the router. Only other thing I can say is that before I bridged the two, the modem was receiving an IPv6 just fine from CL. Really not sure where to start on figuring this out, and the logs show absolutely nothing useful in terms of IPv6, basically other than it's not connected.

    I have had 6rd working before on a different router/modem/firmware combo.
     
  2. Mercjoe

    Mercjoe Network Guru Member

    What firmware fork?
     
  3. JDog2pt0

    JDog2pt0 Network Newbie Member

    Advanced Tomato R7000-AT-ARM-3.4-140-AIO-64K if that's what you were asking. It's a shibby build with a modified GUI
     
  4. Sean B.

    Sean B. LI Guru Member

    Post screenshots of the Basic->IPv6 and Advanced->DHCP/dns sections for the GUI please.
     
  5. JDog2pt0

    JDog2pt0 Network Newbie Member

  6. Sean B.

    Sean B. LI Guru Member

    Looks correct. Run the command " ifconfig " in Tools->System Commands or a telnet/ssh shell on the router, with IPv6 active as in your screenshots, and post output please.
     
  7. JDog2pt0

    JDog2pt0 Network Newbie Member

    Code:
    br0        Link encap:Ethernet  HWaddr 10:DA:43:D2:2A:4A   
               inet addr:192.168.1.1  Bcast:192.168.1.255  Mask:255.255.255.0
               inet6 addr: fe80::12da:43ff:fed2:2a4a/64 Scope:Link
               inet6 addr: 2602::1/64 Scope:Global
               UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
               RX packets:594362 errors:0 dropped:0 overruns:0 frame:0
               TX packets:937446 errors:0 dropped:0 overruns:0 carrier:0
               collisions:0 txqueuelen:0 
               RX bytes:69822082 (66.5 MiB)  TX bytes:1140226856 (1.0 GiB)
     
    eth0       Link encap:Ethernet  HWaddr 10:DA:43:D2:2A:4A   
               UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
               RX packets:956829 errors:0 dropped:0 overruns:0 frame:0
               TX packets:670367 errors:0 dropped:0 overruns:0 carrier:0
               collisions:0 txqueuelen:1000 
               RX bytes:1151157872 (1.0 GiB)  TX bytes:147824439 (140.9 MiB)
               Interrupt:179 Base address:0x4000 
     
    eth1       Link encap:Ethernet  HWaddr 10:DA:43:D2:2A:4C   
               UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
               RX packets:183351 errors:0 dropped:0 overruns:0 frame:264682
               TX packets:292213 errors:8 dropped:0 overruns:0 carrier:0
               collisions:0 txqueuelen:1000 
               RX bytes:40354130 (38.4 MiB)  TX bytes:288073527 (274.7 MiB)
               Interrupt:163 
     
    eth2       Link encap:Ethernet  HWaddr 10:DA:43:D2:2A:4D   
               inet6 addr: fe80::12da:43ff:fed2:2a4d/64 Scope:Link
               UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
               RX packets:368572 errors:0 dropped:0 overruns:0 frame:555003
               TX packets:599447 errors:23 dropped:0 overruns:0 carrier:0
               collisions:0 txqueuelen:1000 
               RX bytes:33728837 (32.1 MiB)  TX bytes:817037475 (779.1 MiB)
               Interrupt:169 
     
    lo         Link encap:Local Loopback   
               inet addr:127.0.0.1  Mask:255.0.0.0
               inet6 addr: ::1/128 Scope:Host
               UP LOOPBACK RUNNING MULTICAST  MTU:16436  Metric:1
               RX packets:14000 errors:0 dropped:0 overruns:0 frame:0
               TX packets:14000 errors:0 dropped:0 overruns:0 carrier:0
               collisions:0 txqueuelen:0 
               RX bytes:1553555 (1.4 MiB)  TX bytes:1553555 (1.4 MiB)
     
    ppp0       Link encap:Point-to-Point Protocol   
               inet addr:75.169.196.201  P-t-P:207.108.176.12  Mask:255.255.255.255
               UP POINTOPOINT RUNNING MULTICAST  MTU:1492  Metric:1
               RX packets:98341 errors:0 dropped:0 overruns:0 frame:0
               TX packets:49727 errors:0 dropped:0 overruns:0 carrier:0
               collisions:0 txqueuelen:3 
               RX bytes:120368733 (114.7 MiB)  TX bytes:4908665 (4.6 MiB)
     
    vlan1      Link encap:Ethernet  HWaddr 10:DA:43:D2:2A:4A   
               UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
               RX packets:49136 errors:0 dropped:0 overruns:0 frame:0
               TX packets:99306 errors:0 dropped:0 overruns:0 carrier:0
               collisions:0 txqueuelen:0 
               RX bytes:4607301 (4.3 MiB)  TX bytes:61761840 (58.9 MiB)
     
    vlan2      Link encap:Ethernet  HWaddr 10:DA:43:D2:2A:4B   
               inet addr:192.168.0.2  Bcast:0.0.0.0  Mask:255.255.255.255
               inet6 addr: fe80::12da:43ff:fed2:2a4b/64 Scope:Link
               UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
               RX packets:907693 errors:0 dropped:0 overruns:0 frame:0
               TX packets:571061 errors:0 dropped:0 overruns:0 carrier:0
               collisions:0 txqueuelen:0 
               RX bytes:1129327649 (1.0 GiB)  TX bytes:86062599 (82.0 MiB)
     
    wds0.1     Link encap:Ethernet  HWaddr 10:DA:43:D2:2A:4C   
               inet6 addr: fe80::12da:43ff:fed2:2a4c/64 Scope:Link
               UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
               RX packets:27 errors:0 dropped:0 overruns:0 frame:264682
               TX packets:21957 errors:31 dropped:0 overruns:0 carrier:0
               collisions:0 txqueuelen:1000 
               RX bytes:1982 (1.9 KiB)  TX bytes:3579713 (3.4 MiB) 
     
  8. Sean B.

    Sean B. LI Guru Member

    Your router's getting an IPv6 address, just not a valid one:

    Either some of the recent changes to IPv6 that have gone into Advanced Tomato are not operating correctly, or your modem is not passing the prefix correctly. I'll look into it more and see what I can find.

    I notice you said you had IPv6 working with other routers/modems. On this modem you did remember to enable IPv6/transparent bridging, correct?
     
    Last edited: Dec 23, 2017
  9. JDog2pt0

    JDog2pt0 Network Newbie Member

    IPv6 was working on the previous modem router combo, but then stopped. I suspect it stopped working after flashing to a different version of tomato, but it's been so long I can't remember. As for this modem, it is bridged, dmz'ed, firewall is completely open.
     
  10. Sean B.

    Sean B. LI Guru Member

    If it's an acceptable option for you, I'd recommend trying Toastmans ARM build as it's IPv6 support is known functional ( I used the MIPS build on an E3000 on Centurylink myself, and the ARM build currently for my RT-AC68P on Comcast, both fully functional IPv6 ). This would confirm or eliminate IPv6 changes in your current firmware as the issue.

    Here's a link to the R7000 full build ( as you've already done the initial ) if interested to try:

    https://www.4shared.com/file/HjhvIB79ce/tomato-R7000-90088Toastman-ARM.html
     
    JDog2pt0 and kille72 like this.
  11. JDog2pt0

    JDog2pt0 Network Newbie Member

    Toastman build is up and running. IPv6 works 10/10, and amazingly enough so does WDS. Wonder what's up in the Shibby/Advanced Tomato. No DNSCrypt though, which is disappointing unless it's located somewhere else.
     
  12. Mercjoe

    Mercjoe Network Guru Member

    I had 6RD IPv6 working with AT&T U-Verse on shibby 140 before I switched over to kille72's fork.

    At least now we have a baseline to compare against if it is working on one firmware fork and not another.
     
  13. Sean B.

    Sean B. LI Guru Member

    @kille72 , @Jacky444 , I would suggest taking a closer look at @tvlz 's IPv6 patches and any others that have come in sense 140.
     
  14. Sean B.

    Sean B. LI Guru Member

    Unfortunately no one has heard from Toastman in almost a year now. As such, his builds are falling behind. I still run it, as I'm not concerned by any shortcomings as of yet, and assisted with patching his builds before he vanished so will just change what I need and build myself. But up to you if you want to stick with it or not, was just a good choice for diagnosing your IPv6 issue.
     
  15. JDog2pt0

    JDog2pt0 Network Newbie Member

    If it ain't broke, don't fix it. I went ahead and flashed one of Toastman's builds on my WDS router as well. Not having DNSCrypt isn't a deal breaker, but the lack of functioning IPv6 was. Generally once I find a functional and stable build, I tend to leave things be unless something comes up. The help has been greatly appreciated.
     
    Sean B. likes this.
  16. Sean B.

    Sean B. LI Guru Member

    You're welcome. And keep in mind, most feature types can be added on via optware/entware using a dedicated flash drive ( a USB HDD would be better if there will be consistent writing to the /opt partition ). Allows for many packages to be installed and run with, or instead of, the firmware's own packages. Including dnscrypt:

    Code:
    root@Storage:/tmp/home/root# ipkg list | grep dnscrypt
    dnscrypt-proxy - 1.6.0-2 - A tool for securing communications between a client and a DNS resolver.
    root@Storage:/tmp/home/root#
     
    Last edited: Dec 24, 2017
  17. Jacky444

    Jacky444 LI Guru Member

    What? I have latest patches which I helped fixing/adapting lol?
     
  18. Sean B.

    Sean B. LI Guru Member

    As I'm sure you're well aware, use cases are different. As shown, OP was receiving an erroneous IPv6 IP for the LAN interface in spite of correctly configured IPv6 GUI settings:

    After flashing a Toastman-ARM build as a diagnostic step OP now has full IPv6 connectivity. Unless OP is blatantly providing misleading information, it's rather clear the firmware was causing the issue. Just letting you know, as a suggestion to take a look.
     
  19. JDog2pt0

    JDog2pt0 Network Newbie Member

    Not sure if I should open a thread? I don't know if this a Centurylink problem, or router (doubt it). Basically, any site or services I use that runs off IPv6 instead of IPv4 (say Youtube for example) runs with very poor throughput. Disabling IPv6 (forcing everything back to IPv4 returns speeds back to normal). I know it's a tunnel and all that, but is the overhead really that much?
     
  20. Sean B.

    Sean B. LI Guru Member

    Telnet/ssh into the router and run a traceroute over v6 to somewhere like google:

    Code:
    traceroute -6 www.google.com
    And post output. Also, have you ran an ipv6 speed test to a major backbone connected provider, such as Comcast?
     
  21. JDog2pt0

    JDog2pt0 Network Newbie Member

    I've used fast.com (It only measures DL) after learning about the speedtest.net bias ISP can do. Based on a FF addon I have that tells me whether I'm using IPv4 or 6 for whatever site I'm on, Fast.com runs over 6 when I have it on. Anyways, the results are...something. Basically with IPv6 my DL speeds take forever, almost the whole course of the test to actually ramp up to as fast as it's going to go (and it tends to fluctuate a decent amount when it gets going as fast as it's going to). With IPv6 disabled, the test gets to speed as soon at starts, say 25mbps, and stays there will very little fluctuation.

    Code:
    traceroute to www.google.com (2607:f8b0:4009:800::2004), 30 hops max, 16 byte packets
     1  *  *  *
     2  2001:428:0:1:205:171:253:81 (2001:428:0:1:205:171:253:81)  33.348 ms  33.070 ms  62.614 ms
     3  2001:428::205:171:200:218 (2001:428::205:171:200:218)  33.630 ms  34.106 ms  33.851 ms
     4  2001:428:3801:210:0:2c:0:2 (2001:428:3801:210:0:2c:0:2)  71.927 ms  81.125 ms  69.940 ms
     5  2001:4860::9:4000:e95b (2001:4860::9:4000:e95b)  34.399 ms  34.244 ms  35.629 ms
     6  2001:4860::c:4000:d209 (2001:4860::c:4000:d209)  38.894 ms  34.547 ms  34.353 ms
     7  2001:4860::8:4000:cbc2 (2001:4860::8:4000:cbc2)  53.087 ms  55.002 ms  52.594 ms
     8  2001:4860::c:4000:d2a0 (2001:4860::c:4000:d2a0)  55.827 ms  60.745 ms  68.355 ms
     9  2001:4860::9:4001:141c (2001:4860::9:4001:141c)  99.209 ms  99.017 ms  107.879 ms
    10  *  *  *
    11  2001:4860:0:1::98d (2001:4860:0:1::98d)  113.567 ms  107.792 ms  116.068 ms
    12  ord36s04-in-x04.1e100.net (2607:f8b0:4009:800::2004)  52.632 ms  52.496 ms  52.492 ms 
    Do you think it could be my IPv6 DNS? I changed off of Centurylink's IPv6 DNS servers, and use the IPv6 OpenDNS servers instead. I think I'll test that... and no change whatsoever. For now I'll just leave it with the CL IPv6 DNS servers for a baseline to test against.
     
  22. Sean B.

    Sean B. LI Guru Member

    There's some pretty bad latency just a few hops out from your connection:

    I'd be willing to be this is an issue on Centurylink's side, something with equipment or crosstalk in the lines etc. How's the speeds for client to client LAN transfer? Should be over link local IP's ( fe80:'s ) . Would be worth trying a lower tunnel MTU, something like 1450. IPv6 packets cannot be fragmented, therefore even with PATH MTU discovery, if a link in the chain chokes bit.. it can cause a lot more drop than you'd usually see in IPv4. Make sure if there's any settings in your modem that are still active in bridge mode relating to MTU or IPv6 that they are properly configured. Also, have you checked your line stats at all? Not sure about your specific modem, but for the C1000Z I had while using Century you could enable a telnet or ssh connection into the modem, once in a shell you could run extensive line diagnostic tests.. mainly looking for the signal to noise ratio on the line.

    And no, it's not your DNS. DNS would only stall you out at when first resolving a hostname. Once the DNS lookup succeeds and the hostname is resolved to IP the answer is cached by hosts for a fair amount of time so DNS isn't a factor for that time either.
     
  23. Sean B.

    Sean B. LI Guru Member


    ***EDIT*** Just thought to mention.. make sure ICMPv6 is fully allowed on everything in your network. Windows/linux computer firewalls, the modem if any settings are still active, and that no custom rules have been set on the router that could deny the protocol. ICMPv6 functionality is critical to the speed and stability of IPv6 connectivity.
     
  24. JDog2pt0

    JDog2pt0 Network Newbie Member

    I do have line issues, and I need to get CL out here to fix them once and for all (that's going to be a battle in and of itself), but that should affect my online performance as a whole (and does), not just IPv6 specifically though right? Funny you mention the MTU as well, as I did nudge it down just a bit and noticed things became a tad more responsive (I'll drop it to 1450). Now for ICMPv6, I haven't seen anything in my router for that. I do have my router set to respond to ICMP ping in the firewall though. The modem is pretty much as dumbed down as I can get it (it is a C2100T) with it obviously in transparent bridge mode, along with my router DMZ'd in it (figured that was safe since the router was doing everything anyways), and with the firewall completely open. I'll see about testing client to client speeds when I get a moment.
     
  25. Sean B.

    Sean B. LI Guru Member

    It can be worse for IPv6, due to IPv6's intolerance of packet fragmentation and SnR ( signal to noise ratio ) on a DSL directly equals bandwidth. The frequency range over which data can be transmitted/received between the modem and the providers DSLAM narrows as the SnR margin shrinks. I'm sure you've seen the classic spiked oscilloscope waveform of an RF transmission before.. that's what noise will look like. As the margin gets tighter all those transient spikes start causing hits well before the full "wall" so to speak actually crushes down. IPv4 packets can be fragmented and reassembled, then a dupack or fast retrans for a much smaller amount of lost data. With IPv6, the entire packet will just black hole, none of the data makes it.

    Certainly not saying its for sure your issue, but something to eyeball so to speak.
     

Share This Page