Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: NANOG: users

IPv6 support via Charter | Ideas on BGP Tunnel via HE

 

 

NANOG users RSS feed   Index | Next | Previous | View Threaded


me at anuragbhatia

Apr 11, 2012, 11:16 AM

Post #1 of 12 (740 views)
Permalink
IPv6 support via Charter | Ideas on BGP Tunnel via HE

Hello,


Does anyone here has clues on IPv6 support via Charter? We recently got BGP
up on the connection and they denied for IPv6 support for now. Support
engineer gave expected time of something like end of year which seems very
late as per our plans.


Is situation same for everyone who sits in downstream of Charter?


Also, does it makes sense to go for BGP Tunnel for now? I just setup IPv6
Tunnel via Hurricane Electric. Latency seems pretty much OK ~ 10-15ms of
overhead. Yet to test other parameters. I heard Tunnels are usually bad.
Can someone tell how to test this tunnel setup to confirm if there is a
performance issue or not? I am thinking of writing a quick bash script and
run via cron to test latency, packet loss and bandwidth throughput for
couple of days. If anyone has better idea, please let me know.



Thanks.

--

Anurag Bhatia
anuragbhatia.com
or simply - http://[2001:470:26:78f::5] if you are on IPv6 connected
network!

Twitter: @anurag_bhatia <https://twitter.com/#!/anurag_bhatia>
Linkedin: http://linkedin.anuragbhatia.com


dxs at twitter

Apr 11, 2012, 11:27 AM

Post #2 of 12 (719 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

Hurricane Electric have a presentation on testing their tunnels using Traceroute6, Tracepath6, and mtr:
http://ipv6.he.net/presentations/trace6.pdf

iperf now supports IPv6 and works well for testing tunnels as well.

I have previously gotten good results from Hurricane Electric tunnels.

--
Dan Sneddon
Network Engineering & Network Security
twitter | AS13414
dxs [at] twitter | Follow me! @dansneddon


On Wednesday, April 11, 2012 at 11:16 AM, Anurag Bhatia wrote:

> Hello,
>
>
> Does anyone here has clues on IPv6 support via Charter? We recently got BGP
> up on the connection and they denied for IPv6 support for now. Support
> engineer gave expected time of something like end of year which seems very
> late as per our plans.
>
>
> Is situation same for everyone who sits in downstream of Charter?
>
>
> Also, does it makes sense to go for BGP Tunnel for now? I just setup IPv6
> Tunnel via Hurricane Electric. Latency seems pretty much OK ~ 10-15ms of
> overhead. Yet to test other parameters. I heard Tunnels are usually bad.
> Can someone tell how to test this tunnel setup to confirm if there is a
> performance issue or not? I am thinking of writing a quick bash script and
> run via cron to test latency, packet loss and bandwidth throughput for
> couple of days. If anyone has better idea, please let me know.
>
>
>
> Thanks.
>
> --
>
> Anurag Bhatia
> anuragbhatia.com
> or simply - http://[2001:470:26:78f::5] if you are on IPv6 connected
> network!
>
> Twitter: @anurag_bhatia <https://twitter.com/#!/anurag_bhatia>
> Linkedin: http://linkedin.anuragbhatia.com
>
>


jim.rampley at chartercom

Apr 11, 2012, 11:30 AM

Post #3 of 12 (717 views)
Permalink
RE: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

Anurag,

We (Charter) are planning on starting early field trials with our business customers with IPv6 real soon (within Q2). We have a few customers already identified, but would you be interested in participating with us?


Jim Rampley | Principal Engineer | 314-543-2505
12405 Powerscourt Drive, St. Louis, MO 63131



-----Original Message-----
From: Anurag Bhatia [mailto:me [at] anuragbhatia]
Sent: Wednesday, April 11, 2012 1:16 PM
To: NANOG Mailing List
Subject: IPv6 support via Charter | Ideas on BGP Tunnel via HE

Hello,


Does anyone here has clues on IPv6 support via Charter? We recently got BGP
up on the connection and they denied for IPv6 support for now. Support
engineer gave expected time of something like end of year which seems very
late as per our plans.


Is situation same for everyone who sits in downstream of Charter?


Also, does it makes sense to go for BGP Tunnel for now? I just setup IPv6
Tunnel via Hurricane Electric. Latency seems pretty much OK ~ 10-15ms of
overhead. Yet to test other parameters. I heard Tunnels are usually bad.
Can someone tell how to test this tunnel setup to confirm if there is a
performance issue or not? I am thinking of writing a quick bash script and
run via cron to test latency, packet loss and bandwidth throughput for
couple of days. If anyone has better idea, please let me know.



Thanks.

--

Anurag Bhatia
anuragbhatia.com
or simply - http://[2001:470:26:78f::5] if you are on IPv6 connected
network!

Twitter: @anurag_bhatia <https://twitter.com/#!/anurag_bhatia>
Linkedin: http://linkedin.anuragbhatia.com

E-MAIL CONFIDENTIALITY NOTICE:
The contents of this e-mail message and any attachments are intended solely for the addressee(s) and may contain confidential and/or legally privileged information. If you are not the intended recipient of this message or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message and any attachments. If you are not the intended recipient, you are notified that any use, dissemination, distribution, copying, or storage of this message or any attachment is strictly prohibited.


seth.mos at dds

Apr 11, 2012, 12:34 PM

Post #4 of 12 (709 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

Hi,

Op 11 apr 2012, om 20:16 heeft Anurag Bhatia het volgende geschreven:

> Also, does it makes sense to go for BGP Tunnel for now? I just setup IPv6
> Tunnel via Hurricane Electric. Latency seems pretty much OK ~ 10-15ms of
> overhead. Yet to test other parameters. I heard Tunnels are usually bad.
> Can someone tell how to test this tunnel setup to confirm if there is a
> performance issue or not? I am thinking of writing a quick bash script and
> run via cron to test latency, packet loss and bandwidth throughput for
> couple of days. If anyone has better idea, please let me know.

Also using a HE.net BGP tunnel for our IPv6, simply because having just 1 native provider with Ipv6 isn't redundant. That and it's 8mbit.

The v4 connection which the tunnel connects over is 90mbit, and the tunnel needs to travel from NL to DE for the FRA BGP peering.

I'm getting about 40mbit through the IPv6 tunnel, so i'd say it works well, although the throughput has slowly been dropping to the 30's range over the last 6 months. But that's probably because of the latency.

For something that is provided for free I'm really glad we have it.

I should have peered with their UK PoP as it's much closer by latency, thus faster.

Cheers,

Seth


bill at herrin

Apr 11, 2012, 1:00 PM

Post #5 of 12 (714 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

On Wed, Apr 11, 2012 at 2:16 PM, Anurag Bhatia <me [at] anuragbhatia> wrote:
> Also, does it makes sense to go for BGP Tunnel for now? I just setup IPv6
> Tunnel via Hurricane Electric. Latency seems pretty much OK ~ 10-15ms of
> overhead. Yet to test other parameters. I heard Tunnels are usually bad.
> Can someone tell how to test this tunnel setup to confirm if there is a
> performance issue or not? I am thinking of writing a quick bash script and
> run via cron to test latency, packet loss and bandwidth throughput for
> couple of days. If anyone has better idea, please let me know.

HE does a fine job with their IPv6 tunnels. If they're you're only v6
connectivity or you need them to provide a backup IPv6 route for when
sole native v6 provider goes down, they're a superb choice.

However...

Do not, do not, do not, rig your system to prefer tunneled IPvanything
to native IPvanythingelse. For all of the obvious reasons.

If you publish an IPv6 address for www.anuragbhatia.com, clients with
IPv6 will use that IPv6 address in preference to the address published
for IPv4. If your sole IPv6 access is with a tunnel, don't publish an
IPv6 address for www. Publish the IPv6 address under
www6.anuragbhatia.com instead. And on your mail server, have the
second MX point to a name with a AAAA, and let the first MX stay on
v4.

Regards,
Bill Herrin


--
William D. Herrin ................ herrin [at] dirtside  bill [at] herrin
3005 Crane Dr. ...................... Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004


me at anuragbhatia

Apr 11, 2012, 1:03 PM

Post #6 of 12 (712 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

Hi Seth


I just did a test from Eu based server sitting below EU based HE Tunnel
node by downloading Ubuntu release file from US based server. This does not
tells about possible high speed but surely tells what is available atleast.
Server itself is sitting on M-Online with 100Mbps pipe.




IPv4:

traceroute to mirror.anl.gov (146.137.96.7), 30 hops max, 60 byte packets
1 gw.giga-dns.com (91.194.90.1) [AS51167] 16.876 ms 16.925 ms 16.915 ms
2 host-93-104-204-33.customer.m-online.net (93.104.204.33) [AS8767]
1.166 ms 1.449 ms 1.445 ms
3 xe-2-1-0.rt-decix-1.m-online.net (212.18.6.162) [AS8767] 8.889 ms
8.888 ms 8.880 ms
4 20gigabitethernet4-3.core1.fra1.he.net (80.81.192.172) [AS6695] 18.586
ms 19.831 ms 19.824 ms
5 10gigabitethernet1-4.core1.par2.he.net (184.105.213.162) [AS6939]
18.794 ms 18.789 ms 10gigabitethernet2-2.core1.par2.he.net (72.52.92.26)
[AS6939] 18.437 ms
6 10gigabitethernet1-1.core1.par2.he.net (184.105.213.90) [AS6939]
18.507 ms 10gigabitethernet7-1.core1.ash1.he.net (184.105.213.93) [AS6939]
96.880 ms 97.345 ms
7 esnet.gigabitethernet4-15.core1.ash1.he.net (216.66.70.18) [AS6939]
95.544 ms 10gigabitethernet7-1.core1.ash1.he.net (184.105.213.93) [AS6939]
97.616 ms esnet.gigabitethernet4-15.core1.ash1.he.net (216.66.70.18)
[AS6939] 95.354 ms
8 washcr1-te-eqxashrt1.es.net (134.55.221.145) [AS293] 97.835 ms
esnet.gigabitethernet4-15.core1.ash1.he.net (216.66.70.18) [AS6939] 95.727
ms washcr1-te-eqxashrt1.es.net (134.55.221.145) [AS293] 98.492 ms
9 washcr1-te-eqxashrt1.es.net (134.55.221.145) [AS293] 98.463 ms
washsdn1-sdn2-washcr1.es.net (134.55.220.54) [AS293] 110.668 ms 110.641 ms
10 starsdn1-ip-washsdn2.es.net (134.55.218.65) [AS293] 120.357 ms
120.844 ms washsdn1-sdn2-washcr1.es.net (134.55.220.54) [AS293] 110.834 ms
11 starcr1-ip-starsdn1.es.net (134.55.219.25) [AS293] 164.788 ms 164.548
ms 164.550 ms
12 starcr1-ip-starsdn1.es.net (134.55.219.25) [AS293] 164.758 ms
anlmr2-starcr1.es.net (134.55.219.53) [AS293] 128.288 ms 128.286 ms
13 guava-esnet.anchor.anl.gov (192.5.170.77) [AS683] 117.532 ms
anlmr2-starcr1.es.net (134.55.219.53) [AS293] 128.263 ms
guava-esnet.anchor.anl.gov (192.5.170.77) [AS683] 117.500 ms
14 * guava-esnet.anchor.anl.gov (192.5.170.77) [AS683] 117.687 ms
117.858 ms
15 * * *
16 * * *
17 * * *
18 * * *


root [at] server:/home/anurag/tmp# wget -4
http://mirror.anl.gov/pub/ubuntu-iso/DVDs/ubuntu/12.04/beta-2/ubuntu-12.04-beta2-dvd-i386.iso
--2012-04-11 21:56:46--
http://mirror.anl.gov/pub/ubuntu-iso/DVDs/ubuntu/12.04/beta-2/ubuntu-12.04-beta2-dvd-i386.iso
Resolving mirror.anl.gov... 146.137.96.7
Connecting to mirror.anl.gov|146.137.96.7|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1644474368 (1.5G) [application/octet-stream]
Saving to: `ubuntu-12.04-beta2-dvd-i386.iso.1'

100%[============================================================================================================================>]
1,644,474,368 4.78M/s in 5m 38s

2012-04-11 22:02:25 (4.64 MB/s) - `ubuntu-12.04-beta2-dvd-i386.iso.1' saved
[1644474368/1644474368]








IPv6:


traceroute to mirror.anl.gov (2620:0:dc0:1800:214:4fff:fe7d:1b9), 30 hops
max, 80 byte packets
1 2001:470:25:78f::1 (2001:470:25:78f::1) [AS6939] 18.918 ms 21.147 ms
23.357 ms
2 gige-g2-20.core1.zrh1.he.net (2001:470:0:11d::1) [AS6939] 23.341 ms
23.324 ms 23.797 ms
3 10gigabitethernet5-1.core1.fra1.he.net (2001:470:0:21c::1) [AS6939]
29.781 ms 30.252 ms 23.671 ms
4 10gigabitethernet5-3.core1.lon1.he.net (2001:470:0:1d2::1) [AS6939]
37.897 ms 37.880 ms 43.095 ms
5 10gigabitethernet7-4.core1.nyc4.he.net (2001:470:0:128::1) [AS6939]
104.552 ms 105.763 ms 105.742 ms
6 10gigabitethernet2-3.core1.ash1.he.net (2001:470:0:36::1) [AS6939]
113.963 ms 114.467 ms 111.478 ms
7
lawrence-berkeley-national-laboratory.gigabitethernet4-15.core1.ash1.he.net(2001:470:1:27f::2)
[AS6939] 109.467 ms 109.452 ms 109.435 ms
8 washcr1-te-eqxashrt1.es.net (2001:400:0:15a::1) [AS293] 115.929 ms
113.625 ms 115.896 ms
9 washsdn1-sdn2-washcr1.es.net (2001:400:0:e0::2) [AS293] 114.606 ms
112.068 ms 112.045 ms
10 starsdn1-ip-washsdn2.es.net (2001:400:0:ab::1) [AS293] 126.783 ms
130.008 ms 126.747 ms
11 starcr1-ip-starsdn1.es.net (2001:400:0:a2::2) [AS293] 127.268 ms
124.223 ms 124.125 ms
12 anlmr2-starcr1.es.net (2001:400:0:c0::1) [AS293] 128.066 ms 130.529
ms 130.513 ms
13 2001:400:2202:8::2 (2001:400:2202:8::2) [AS293] 130.976 ms 128.915 ms
128.892 ms
14 * * *
15 * * *








root [at] server:/home/anurag/tmp# wget -6
http://mirror.anl.gov/pub/ubuntu-iso/DVDs/ubuntu/12.04/beta-2/ubuntu-12.04-beta2-dvd-i386.iso
--2012-04-11 21:45:52--
http://mirror.anl.gov/pub/ubuntu-iso/DVDs/ubuntu/12.04/beta-2/ubuntu-12.04-beta2-dvd-i386.iso
Resolving mirror.anl.gov... 2620:0:dc0:1800:214:4fff:fe7d:1b9
Connecting to mirror.anl.gov|2620:0:dc0:1800:214:4fff:fe7d:1b9|:80...
connected.
HTTP request sent, awaiting response... 200 OK
Length: 1644474368 (1.5G) [application/octet-stream]
Saving to: `ubuntu-12.04-beta2-dvd-i386.iso'

100%[============================================================================================================================>]
1,644,474,368 *3.66M/s in 6m 11s*

2012-04-11 21:52:04 (4.22 MB/s) - `ubuntu-12.04-beta2-dvd-i386.iso' saved
[1644474368/1644474368]


On Thu, Apr 12, 2012 at 1:04 AM, Seth Mos <seth.mos [at] dds> wrote:

> Hi,
>
> Op 11 apr 2012, om 20:16 heeft Anurag Bhatia het volgende geschreven:
>
> > Also, does it makes sense to go for BGP Tunnel for now? I just setup IPv6
> > Tunnel via Hurricane Electric. Latency seems pretty much OK ~ 10-15ms of
> > overhead. Yet to test other parameters. I heard Tunnels are usually bad.
> > Can someone tell how to test this tunnel setup to confirm if there is a
> > performance issue or not? I am thinking of writing a quick bash script
> and
> > run via cron to test latency, packet loss and bandwidth throughput for
> > couple of days. If anyone has better idea, please let me know.
>
> Also using a HE.net BGP tunnel for our IPv6, simply because having just 1
> native provider with Ipv6 isn't redundant. That and it's 8mbit.
>
> The v4 connection which the tunnel connects over is 90mbit, and the tunnel
> needs to travel from NL to DE for the FRA BGP peering.
>
> I'm getting about 40mbit through the IPv6 tunnel, so i'd say it works
> well, although the throughput has slowly been dropping to the 30's range
> over the last 6 months. But that's probably because of the latency.
>
> For something that is provided for free I'm really glad we have it.
>
> I should have peered with their UK PoP as it's much closer by latency,
> thus faster.
>
> Cheers,
>
> Seth
>



--

Anurag Bhatia
anuragbhatia.com
or simply - http://[2001:470:26:78f::5] if you are on IPv6 connected
network!

Twitter: @anurag_bhatia <https://twitter.com/#!/anurag_bhatia>
Linkedin: http://linkedin.anuragbhatia.com


charles-lists at knownelement

Apr 11, 2012, 2:14 PM

Post #7 of 12 (708 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

On 04/11/2012 02:34 PM, Seth Mos wrote:
>
>
> I'm getting about 40mbit through the IPv6 tunnel, so i'd say it works well, although the throughput has slowly been dropping to the 30's range over the last 6 months. But that's probably because of the latency.
>
> For something that is provided for free I'm really glad we have it.

Indeed. It's pretty amazing what HE has put together.

> I should have peered with their UK PoP as it's much closer by latency, thus faster.

Why don't you? Can you setup more then one peering?


paul4004 at gmail

Apr 11, 2012, 2:19 PM

Post #8 of 12 (723 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

He.net tunnels are also good to have because depending on your provider,
there's still many with incomplete views of the ipv6 routing table and he
might have a path. This is a more prevalent issue with ipv6 than v4 at the
moment.
On Apr 11, 2012 2:03 PM, "Anurag Bhatia" <me [at] anuragbhatia> wrote:

> Hi Seth
>
>
> I just did a test from Eu based server sitting below EU based HE Tunnel
> node by downloading Ubuntu release file from US based server. This does not
> tells about possible high speed but surely tells what is available atleast.
> Server itself is sitting on M-Online with 100Mbps pipe.
>
>
>
>
> IPv4:
>
> traceroute to mirror.anl.gov (146.137.96.7), 30 hops max, 60 byte packets
> 1 gw.giga-dns.com (91.194.90.1) [AS51167] 16.876 ms 16.925 ms 16.915
> ms
> 2 host-93-104-204-33.customer.m-online.net (93.104.204.33) [AS8767]
> 1.166 ms 1.449 ms 1.445 ms
> 3 xe-2-1-0.rt-decix-1.m-online.net (212.18.6.162) [AS8767] 8.889 ms
> 8.888 ms 8.880 ms
> 4 20gigabitethernet4-3.core1.fra1.he.net (80.81.192.172) [AS6695]
> 18.586
> ms 19.831 ms 19.824 ms
> 5 10gigabitethernet1-4.core1.par2.he.net (184.105.213.162) [AS6939]
> 18.794 ms 18.789 ms 10gigabitethernet2-2.core1.par2.he.net (72.52.92.26)
> [AS6939] 18.437 ms
> 6 10gigabitethernet1-1.core1.par2.he.net (184.105.213.90) [AS6939]
> 18.507 ms 10gigabitethernet7-1.core1.ash1.he.net (184.105.213.93)
> [AS6939]
> 96.880 ms 97.345 ms
> 7 esnet.gigabitethernet4-15.core1.ash1.he.net (216.66.70.18) [AS6939]
> 95.544 ms 10gigabitethernet7-1.core1.ash1.he.net (184.105.213.93)
> [AS6939]
> 97.616 ms esnet.gigabitethernet4-15.core1.ash1.he.net (216.66.70.18)
> [AS6939] 95.354 ms
> 8 washcr1-te-eqxashrt1.es.net (134.55.221.145) [AS293] 97.835 ms
> esnet.gigabitethernet4-15.core1.ash1.he.net (216.66.70.18) [AS6939]
> 95.727
> ms washcr1-te-eqxashrt1.es.net (134.55.221.145) [AS293] 98.492 ms
> 9 washcr1-te-eqxashrt1.es.net (134.55.221.145) [AS293] 98.463 ms
> washsdn1-sdn2-washcr1.es.net (134.55.220.54) [AS293] 110.668 ms 110.641
> ms
> 10 starsdn1-ip-washsdn2.es.net (134.55.218.65) [AS293] 120.357 ms
> 120.844 ms washsdn1-sdn2-washcr1.es.net (134.55.220.54) [AS293] 110.834
> ms
> 11 starcr1-ip-starsdn1.es.net (134.55.219.25) [AS293] 164.788 ms
> 164.548
> ms 164.550 ms
> 12 starcr1-ip-starsdn1.es.net (134.55.219.25) [AS293] 164.758 ms
> anlmr2-starcr1.es.net (134.55.219.53) [AS293] 128.288 ms 128.286 ms
> 13 guava-esnet.anchor.anl.gov (192.5.170.77) [AS683] 117.532 ms
> anlmr2-starcr1.es.net (134.55.219.53) [AS293] 128.263 ms
> guava-esnet.anchor.anl.gov (192.5.170.77) [AS683] 117.500 ms
> 14 * guava-esnet.anchor.anl.gov (192.5.170.77) [AS683] 117.687 ms
> 117.858 ms
> 15 * * *
> 16 * * *
> 17 * * *
> 18 * * *
>
>
> root [at] server:/home/anurag/tmp# wget -4
>
> http://mirror.anl.gov/pub/ubuntu-iso/DVDs/ubuntu/12.04/beta-2/ubuntu-12.04-beta2-dvd-i386.iso
> --2012-04-11 21:56:46--
>
> http://mirror.anl.gov/pub/ubuntu-iso/DVDs/ubuntu/12.04/beta-2/ubuntu-12.04-beta2-dvd-i386.iso
> Resolving mirror.anl.gov... 146.137.96.7
> Connecting to mirror.anl.gov|146.137.96.7|:80... connected.
> HTTP request sent, awaiting response... 200 OK
> Length: 1644474368 (1.5G) [application/octet-stream]
> Saving to: `ubuntu-12.04-beta2-dvd-i386.iso.1'
>
>
> 100%[============================================================================================================================>]
> 1,644,474,368 4.78M/s in 5m 38s
>
> 2012-04-11 22:02:25 (4.64 MB/s) - `ubuntu-12.04-beta2-dvd-i386.iso.1' saved
> [1644474368/1644474368]
>
>
>
>
>
>
>
>
> IPv6:
>
>
> traceroute to mirror.anl.gov (2620:0:dc0:1800:214:4fff:fe7d:1b9), 30 hops
> max, 80 byte packets
> 1 2001:470:25:78f::1 (2001:470:25:78f::1) [AS6939] 18.918 ms 21.147 ms
> 23.357 ms
> 2 gige-g2-20.core1.zrh1.he.net (2001:470:0:11d::1) [AS6939] 23.341 ms
> 23.324 ms 23.797 ms
> 3 10gigabitethernet5-1.core1.fra1.he.net (2001:470:0:21c::1) [AS6939]
> 29.781 ms 30.252 ms 23.671 ms
> 4 10gigabitethernet5-3.core1.lon1.he.net (2001:470:0:1d2::1) [AS6939]
> 37.897 ms 37.880 ms 43.095 ms
> 5 10gigabitethernet7-4.core1.nyc4.he.net (2001:470:0:128::1) [AS6939]
> 104.552 ms 105.763 ms 105.742 ms
> 6 10gigabitethernet2-3.core1.ash1.he.net (2001:470:0:36::1) [AS6939]
> 113.963 ms 114.467 ms 111.478 ms
> 7
> lawrence-berkeley-national-laboratory.gigabitethernet4-15.core1.ash1.he.net
> (2001:470:1:27f::2)
> [AS6939] 109.467 ms 109.452 ms 109.435 ms
> 8 washcr1-te-eqxashrt1.es.net (2001:400:0:15a::1) [AS293] 115.929 ms
> 113.625 ms 115.896 ms
> 9 washsdn1-sdn2-washcr1.es.net (2001:400:0:e0::2) [AS293] 114.606 ms
> 112.068 ms 112.045 ms
> 10 starsdn1-ip-washsdn2.es.net (2001:400:0:ab::1) [AS293] 126.783 ms
> 130.008 ms 126.747 ms
> 11 starcr1-ip-starsdn1.es.net (2001:400:0:a2::2) [AS293] 127.268 ms
> 124.223 ms 124.125 ms
> 12 anlmr2-starcr1.es.net (2001:400:0:c0::1) [AS293] 128.066 ms 130.529
> ms 130.513 ms
> 13 2001:400:2202:8::2 (2001:400:2202:8::2) [AS293] 130.976 ms 128.915 ms
> 128.892 ms
> 14 * * *
> 15 * * *
>
>
>
>
>
>
>
>
> root [at] server:/home/anurag/tmp# wget -6
>
> http://mirror.anl.gov/pub/ubuntu-iso/DVDs/ubuntu/12.04/beta-2/ubuntu-12.04-beta2-dvd-i386.iso
> --2012-04-11 21:45:52--
>
> http://mirror.anl.gov/pub/ubuntu-iso/DVDs/ubuntu/12.04/beta-2/ubuntu-12.04-beta2-dvd-i386.iso
> Resolving mirror.anl.gov... 2620:0:dc0:1800:214:4fff:fe7d:1b9
> Connecting to mirror.anl.gov|2620:0:dc0:1800:214:4fff:fe7d:1b9|:80...
> connected.
> HTTP request sent, awaiting response... 200 OK
> Length: 1644474368 (1.5G) [application/octet-stream]
> Saving to: `ubuntu-12.04-beta2-dvd-i386.iso'
>
>
> 100%[============================================================================================================================>]
> 1,644,474,368 *3.66M/s in 6m 11s*
>
> 2012-04-11 21:52:04 (4.22 MB/s) - `ubuntu-12.04-beta2-dvd-i386.iso' saved
> [1644474368/1644474368]
>
>
> On Thu, Apr 12, 2012 at 1:04 AM, Seth Mos <seth.mos [at] dds> wrote:
>
> > Hi,
> >
> > Op 11 apr 2012, om 20:16 heeft Anurag Bhatia het volgende geschreven:
> >
> > > Also, does it makes sense to go for BGP Tunnel for now? I just setup
> IPv6
> > > Tunnel via Hurricane Electric. Latency seems pretty much OK ~ 10-15ms
> of
> > > overhead. Yet to test other parameters. I heard Tunnels are usually
> bad.
> > > Can someone tell how to test this tunnel setup to confirm if there is a
> > > performance issue or not? I am thinking of writing a quick bash script
> > and
> > > run via cron to test latency, packet loss and bandwidth throughput for
> > > couple of days. If anyone has better idea, please let me know.
> >
> > Also using a HE.net BGP tunnel for our IPv6, simply because having just 1
> > native provider with Ipv6 isn't redundant. That and it's 8mbit.
> >
> > The v4 connection which the tunnel connects over is 90mbit, and the
> tunnel
> > needs to travel from NL to DE for the FRA BGP peering.
> >
> > I'm getting about 40mbit through the IPv6 tunnel, so i'd say it works
> > well, although the throughput has slowly been dropping to the 30's range
> > over the last 6 months. But that's probably because of the latency.
> >
> > For something that is provided for free I'm really glad we have it.
> >
> > I should have peered with their UK PoP as it's much closer by latency,
> > thus faster.
> >
> > Cheers,
> >
> > Seth
> >
>
>
>
> --
>
> Anurag Bhatia
> anuragbhatia.com
> or simply - http://[2001:470:26:78f::5] if you are on IPv6 connected
> network!
>
> Twitter: @anurag_bhatia <https://twitter.com/#!/anurag_bhatia>
> Linkedin: http://linkedin.anuragbhatia.com
>


jared at puck

Apr 11, 2012, 2:41 PM

Post #9 of 12 (703 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

On Apr 11, 2012, at 5:19 PM, PC wrote:

> He.net tunnels are also good to have because depending on your provider,
> there's still many with incomplete views of the ipv6 routing table and he
> might have a path. This is a more prevalent issue with ipv6 than v4 at the
> moment.

This is a big problem for the two providers involved in this "spat" having
inconsistent IPv4/IPv6 business relationships (peering, etc).

There are many professional service providers that will happily dual-stack
your internet port with consistent business relationships. Don't let these
two parties that so far have agreed to disagree prevent you from using IPv6
to its fullest. Select another carrier.

- Jared


bill at herrin

Apr 11, 2012, 3:19 PM

Post #10 of 12 (699 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

On Wed, Apr 11, 2012 at 5:41 PM, Jared Mauch <jared [at] puck> wrote:
> This is a big problem for the two providers involved in this "spat" having
> inconsistent IPv4/IPv6 business relationships (peering, etc).
>
> There are many professional service providers that will happily dual-stack
> your internet port with consistent business relationships.  Don't let these
> two parties that so far have agreed to disagree prevent you from using IPv6
> to its fullest.  Select another carrier.

Hi Jared,

Is it really fair to say there are "two" parties in a peering spat
when Cogent is one of them?

Regards,
Bill Herrin



--
William D. Herrin ................ herrin [at] dirtside  bill [at] herrin
3005 Crane Dr. ...................... Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004


jared at puck

Apr 11, 2012, 3:35 PM

Post #11 of 12 (707 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

On Apr 11, 2012, at 6:19 PM, William Herrin wrote:

> On Wed, Apr 11, 2012 at 5:41 PM, Jared Mauch <jared [at] puck> wrote:
>> This is a big problem for the two providers involved in this "spat" having
>> inconsistent IPv4/IPv6 business relationships (peering, etc).
>>
>> There are many professional service providers that will happily dual-stack
>> your internet port with consistent business relationships. Don't let these
>> two parties that so far have agreed to disagree prevent you from using IPv6
>> to its fullest. Select another carrier.
>
> Hi Jared,
>
> Is it really fair to say there are "two" parties in a peering spat
> when Cogent is one of them?

I know that the following IPv4 as-paths appear to enumerate transit paths where the providers can do IPv6 transit as well.

If HE does not take advantage of those existing IP transit connections for whichever IP version I'm not sure where I would cast blame. Perhaps those ports don't have IPv6. I have my own opinions about peering disputes which you can obtain privately.

route-views>sh ip bgp 216.218.186.2
BGP routing table entry for 216.218.128.0/17, version 417045
Paths: (35 available, best #26, table Default-IP-Routing-Table)
...snip...
1239 3549 6939 6939
144.228.241.130 from 144.228.241.130 (144.228.241.130)
Origin IGP, localpref 100, valid, external
2914 1299 6939 6939
129.250.0.11 from 129.250.0.11 (129.250.0.12)
Origin IGP, metric 5, localpref 100, valid, external
Community: 2914:420 2914:1008 2914:2000 2914:3000 65504:1299
3356 3549 6939 6939
4.69.184.193 from 4.69.184.193 (4.68.3.50)
Origin IGP, metric 0, localpref 100, valid, external
Community: 3356:3 3356:22 3356:86 3356:575 3356:666 3356:2012 3549:4143 3549:30840
701 1299 6939 6939
157.130.10.233 from 157.130.10.233 (137.39.3.60)
Origin IGP, localpref 100, valid, external
1668 3549 6939 6939
66.185.128.48 from 66.185.128.48 (66.185.128.48)
Origin IGP, metric 7, localpref 100, valid, external
7018 1299 6939 6939
12.0.1.63 from 12.0.1.63 (12.0.1.63)
Origin IGP, localpref 100, valid, external
Community: 7018:5000
3257 1299 6939 6939
89.149.178.10 from 89.149.178.10 (213.200.87.91)
Origin IGP, metric 10, localpref 100, valid, external
Community: 3257:8100 3257:30052 3257:50001 3257:54900 3257:54901
3561 3549 6939 6939
206.24.210.102 from 206.24.210.102 (206.24.210.102)
Origin IGP, localpref 100, valid, external
6453 3549 6939 6939
66.110.0.86 from 66.110.0.86 (66.110.0.86)
Origin IGP, localpref 100, valid, external


owen at delong

Apr 11, 2012, 4:31 PM

Post #12 of 12 (694 views)
Permalink
Re: IPv6 support via Charter | Ideas on BGP Tunnel via HE [In reply to]

On Apr 11, 2012, at 2:14 PM, Charles N Wyble wrote:

> On 04/11/2012 02:34 PM, Seth Mos wrote:
>>
>>
>> I'm getting about 40mbit through the IPv6 tunnel, so i'd say it works well, although the throughput has slowly been dropping to the 30's range over the last 6 months. But that's probably because of the latency.
>>
>> For something that is provided for free I'm really glad we have it.
>
> Indeed. It's pretty amazing what HE has put together.
>
>> I should have peered with their UK PoP as it's much closer by latency, thus faster.
>
> Why don't you? Can you setup more then one peering?
>
>

Yes... Many of our customers set up multiple peerings.

Owen

NANOG users RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.