|
IPnG unreach?
![[be]](/s/countries/be.gif) Carmen Sandiego on Saturday, 13 September 2003 18:57:04
Anyone having problems using IPnG POP, IPv6 connectivity seems to have gone down somewhere last evening (12/09), no sign of life atm.
Will update post if connectivity should come back up.
IPnG unreach?
I wonder how I can be typing this then :)
Show us the debug! (traceroute + traceroute6 + ping + ping6 + .. + ...)
Did you check POP status btw?
That does give a live feed from the POP to this site, if there is no communication possible between those two sites then the POP is marked down.
IPnG unreach?
Carmen Sandiego on Saturday, 13 September 2003 23:56:36
traceroute to 212.19.192.219 (212.19.192.219), 30 hops max, 38 byte packets
1 212.71.29.153 (212.71.29.153) 0.606 ms 0.403 ms 0.416 ms
2 br01.edpnet.net (212.71.4.4) 3.078 ms 2.961 ms 3.352 ms
3 bnix01.edpnet.net (212.71.4.3) 3.720 ms 4.493 ms 3.665 ms
4 lambdanet.bnix.net (194.53.172.118) 4.225 ms 5.314 ms 7.689 ms
5 217.71.96.214 (217.71.96.214) 8.160 ms 8.373 ms 8.134 ms
6 intouch-nv.nl.lambdanet.net (217.71.99.106) 8.217 ms 9.431 ms 8.294 ms
7 217.8.101.242 (217.8.101.242) 10.899 ms 9.003 ms 8.983 ms
8 * * *
9 tunnelserver.ipng.nl (212.19.192.219) 13.807 ms 9.347 ms 12.315 ms
$ ping -R 212.19.192.219
PING 212.19.192.219 (212.19.192.219) from 212.71.29.155 : 56(124) bytes of data.
64 bytes from 212.19.192.219: icmp_seq=0 ttl=247 time=12.835 msec
RR: 212.71.29.155
194.53.172.89
80.245.35.11
217.71.99.105
217.8.101.245
212.26.220.175
212.19.192.217
212.19.192.219
212.19.192.219
64 bytes from 212.19.192.219: icmp_seq=1 ttl=247 time=12.582 msec
RR: 212.71.29.155
194.53.172.89
80.245.35.11
217.71.99.105
217.8.101.241
212.26.220.175
212.19.192.217
212.19.192.219
212.19.192.219
Bloody weird I can ping it tho .. coz tunnel connectivity seems down
Weird coz I didn't change any config or anything ... tried to reset tunnel up .. nothing, also graphs on homepage show blanks, Graphs shows it going down at 4:00 this morning
IPnG unreach?
Carmen Sandiego on Sunday, 14 September 2003 00:10:28
14/09, the wolves have woken up .. connectivity is back online;
Ping & Traceroute still show the same results .. but it's working now ..
I'm sure the graphs will start updating soon again :)
Now stop playing around Jeroen ;)
IPnG unreach?
Nopes, nothing changed at all, you have to realize that the internet(tm) is a ever changing world of routes.
IPnG unreach?
![[fr]](/s/countries/fr.gif) Carmen Sandiego on Saturday, 13 September 2003 22:45:07
Hi,
Have the same probleme IPng POP (212.19.192.219) seem down last 12/09 at 00:45 :
PING 212.19.192.219 (212.19.192.219) from 62.x.x.x : 56(84) bytes of data.
--- 212.19.192.219 ping statistics ---
5 packets transmitted, 0 received, 100% loss, time 4012ms
traceroute to 212.19.192.219 (212.19.192.219), 30 hops max, 38 byte packets
1 loopback1-lns101-tip-courbevoie.nerim.net (62.4.16.246) 57.469 ms 59.330 ms 59.606 ms
2 hsrp0-courbevoie.nerim.net (62.4.16.3) 55.982 ms 57.647 ms 57.386 ms
3 geth0-2-holger.nerim.net (62.4.16.33) 59.952 ms 55.392 ms 55.681 ms
4 lambdanet-france.sfinx.tm.fr (194.68.129.164) 57.970 ms 59.751 ms 61.257 ms
5 LDCH-1-pos022.fr.lambdanet.net (217.71.96.77) 59.945 ms 57.677 ms 59.647 ms
6 TE2-1-pos030.nl.lambdanet.net (217.71.96.102) 78.032 ms 77.380 ms 75.763 ms
7 InTouch-NV.nl.lambdanet.net (217.71.99.106) 75.845 ms 75.932 ms 77.811 ms
8 *
PING 3ffe:8114:1000::594(3ffe:8114:1000::594) from 3ffe:8114:2000:1890:30::1 : 56 data bytes
--- 3ffe:8114:1000::594 ping statistics ---
3 packets transmitted, 0 received, 100% loss, time 2011ms
IPnG unreach?
This could be a number of things, but apparently there is no route to your network from the POP. Maybe there are some problems at the transit providers.
IPnG unreach?
![[gb]](/s/countries/gb.gif) Shadow Hawkins on Sunday, 14 September 2003 03:39:23
Looks like packets destined for the IPng PoP are taking a slow boat to China :-)
[tez@snuffleupagus tez]$ /usr/sbin/traceroute 212.19.192.219
traceroute to 212.19.192.219 (212.19.192.219), 30 hops max, 38 byte packets
1 demon-gw.spilsby.net (193.195.23.161) 3.116 ms 2.460 ms 2.643 ms
2 * demon-du.demon.co.uk (158.152.1.222) 45.295 ms 34.144 ms
3 anchor-ndu-as20-l0.access.demon.net (194.159.171.60) 31.883 ms 29.203 ms 28.904 ms
4 anchor-core-21-fxp3.router.demon.net (195.173.57.251) 225.908 ms 561.217 ms 780.806 ms
5 anchor-core-11-242.router.demon.net (194.159.187.18) 1067.734 ms 969.548 ms *
6 anchor-border-1-4-0-2-549.router.demon.net (194.159.187.1) 744.528 ms 673.494 ms 979.770 ms
7 * LINX.LON-1-eth110.uk.lambdanet.net (195.66.224.99) 815.185 ms 964.284 ms
8 TE2-1-pos030.nl.lambdanet.net (217.71.96.102) 976.756 ms 967.659 ms *
9 217.8.101.241 (217.8.101.241) 218.688 ms 231.907 ms 803.809 ms
10 * * *
11 * * *
12 * * *
Doing a lookup on hop 9 tells me that the registered netblock owner is Outblaze (a company that does mail forwarding, etc) which if the RIPE WHOIS entry is to be believed - is in China!
[tez@cybernoid tez]$ whois 217.8.101.241
[Querying whois.ripe.net]
[whois.ripe.net]
% This is the RIPE Whois server.
% The objects are in RPSL format.
%
% Rights restricted by copyright.
% See http://www.ripe.net/ripencc/pub-services/db/copyright.html
inetnum: 217.8.101.0 - 217.8.101.255
netname: OUTBLAZE2002-NET
descr: OUTBLAZE Ltd.
country: CN
admin-c: MW1067-RIPE
tech-c: RO83-RIPE
tech-c: FVE2-RIPE
status: ASSIGNED PA
mnt-by: AS8954-MNT
notify: willy@intouch.nl
changed: willy@intouch.nl 20020610
source: RIPE
Jeroen, I assume that users of the IPng PoP won't get credits deducted due to this outage ?
IPnG unreach?
Apparently there are routing issues when traffic go over Lambdanet, we can't do anything about that and the traffic to 99% of the rest of the world does arrive.
You might notice that hop 3 to 4 is already taking a 300ms hit, which is a lot.
Check the Latency pages showing that the POP can perfectly reach the rest of the world.
IPnG unreach?
![[dk]](/s/countries/dk.gif) Carmen Sandiego on Sunday, 14 September 2003 10:48:16
I've had the same problem since midnight friday/saturday.
Matt's traceroute from here (217.157.1.202) to tunnelserver.ipng.nl says:
$ mtr -c 10 -n tunnelserver.ipng.nl
HOST LOSS RCVD SENT BEST AVG WORST
217.157.1.201 0% 10 10 1.01 1.06 1.09
212.242.3.113 0% 10 10 8.14 9.12 12.49
212.242.18.98 0% 10 10 8.18 9.30 11.19
64.211.166.101 0% 10 10 8.89 10.55 18.17
67.17.65.181 0% 10 10 9.19 10.19 12.60
67.17.92.213 0% 10 10 23.16 24.00 24.83
67.17.65.82 0% 10 10 23.48 24.12 25.01
80.81.192.74 0% 10 10 23.78 24.51 25.36
217.71.105.41 0% 10 10 23.84 24.63 28.01
217.71.105.81 0% 10 10 27.36 27.86 28.93
217.71.96.89 0% 10 10 25.45 26.16 27.01
217.71.99.106 0% 10 10 25.54 26.58 27.77
??? 100% 0 10 0.00 0.00 0.00
$
Both mtr6 and traceroute6 say "sendto: Network is unreachable". Of course ping6 shows 100% packet loss.
The tunnel resides on a Debian/Linux box, and I'm using interfaces(5) to bring the tunnel up (not that it helps doing it "by hand", using e.g. iproute2). I have a /60 subnet. When I use "netmask 60" in the interfaces file, the tunnel is created, but seem unoperational. The funny thing is, when I use "netmask 127" in the interfaces file, ifup barfs "Host is unreachable". It used to work with the "netmask 60", i.e. until midnight between friday and saturday.
What else can I try in order to bring the tunnel back up?
IPnG unreach?
Also coming over lambdanet, must be the same problem as the other two.
The tunnel resides on a Debian/Linux box, and I'm using interfaces(5) to bring the tunnel up (not that it helps doing it "by hand", using e.g. iproute2). I have a /60 subnet.
Tunnel != subnet, the tunnel is for the transit, the tunnel is what behind it.
When I use "netmask 60" in the interfaces file, the tunnel is created, but seem unoperational. The funny thing is, when I use "netmask 127" in the interfaces file, ifup barfs "Host is unreachable".
Let me guess, you upgraded to 2.4.21 ? See the rest of the forum about that.
It used to work with the "netmask 60",
Which is totally wrong and just luck as you couldn't reach any of your
neighbouring tunnel users which you covered in the /60 "tunnel".
i.e. until midnight between friday and saturday.
Apparently when Lambdanet decided to start misrouting packets...
IPnG unreach?
Carmen Sandiego on Sunday, 14 September 2003 14:06:55
Thanks for correcting me on the /127 and /60 issue.
I upgraded to Linux 2.4.21 at least a month ago, and my SixXS tunnel was up and working just until a couple of days ago (at 2003-09-13 00:04:11, according to the tunnel information page). I have tried "ip -6 route add default [...]" as well as "ip -6 route add 2000::/3 [...]" -- both without luck.
Monitoring my interfaces with iptraf shows incoming ICMP6-packets from tunnelserver.ipng.nl and outgoing ICMP6-packets from my host. I'm guess my box' replies don't reach the tunnel server, because ping6 says 100% packet loss.
Assuming that Lambdanet is indeed the culprit, will I lose karma (credits) until the tunnel starts working again?
Is there anything I can do, except waiting and trying to be patient?
// Klaus
IPnG unreach?
Monitoring my interfaces with iptraf shows incoming ICMP6-packets from tunnelserver.ipng.nl and outgoing ICMP6-packets from my host. I'm guess my box' replies don't reach the tunnel server, because ping6 says 100% packet loss.
I guess you have a problem of asynchronous routing:
From noc.sixxs.net:
64 bytes from 3ffe:8114:1000::3bf: icmp_seq=1 ttl=53 time=276 ms
It might be that the packets are traveling backwards over another path
and as you are logging into the SixXS site over 6to4... check your routes :)
Could you show a tcpdump of the packets you mention in the above text?
A small tcpdump suggest this too:
15:00:01.380163 tunnelserver.ipng.nl > 217.157.1.202: fbruckman-pt.tunnel.tserv1.fmt.ipv6.he.net.ntp > 3ffe:8114:2000:fb0:7:95ff:feab:e487.ntp: v4 client strat 3 poll 10 prec -17 (DF)
15:00:18.034593 tunnelserver.ipng.nl > 217.157.1.202: gw-480.ams-02.nl.sixxs.net > cl-480.ams-02.nl.sixxs.net: icmp6: echo request (DF)
But no traffic coming back...
Assuming that Lambdanet is indeed the culprit, will I lose karma (credits) until the tunnel starts working again? Is there anything I can do, except waiting and trying to be patient?
Traceroutes to your tunnel endpoint don't go over Lambdanet at the moment, they do go over glbx.net and you seem to be unreachable over IPv4.
Seems like some odd problem making intouch unreachable...
IPnG unreach?
Carmen Sandiego on Sunday, 14 September 2003 23:46:10
The reason you saw me logging into SiXXS over 6to4 is that my SixXS tunnel doesn't work. My browser wants to use IPv6 if there's an IPv6 address for a site, and since there were no route to SixXS, I had to wait for the TCP timeout for the browser to tell me that the host was unreachable. In order to have at least something on IPv6, I told my box to route everything over 6to4.
I'm not sure I can show you a tcpdump, I'm not used to using that tool...
The fbruckman-pt.tunnel.tserv1.fmt.ipv6.he.net is an NTP client using my NTP server that is listening to both IPv4 and IPv6.
IPnG unreach?
Carmen Sandiego on Sunday, 14 September 2003 23:58:18
Just after I finished my previous message, I noticed that tunnelserver.ipng.nl was pingable over IPv4, so I decided to try an restart all interfaces and flush the routing cache. I believe my SixXS tunnel is up and running correctly now.
IPnG unreach?
Note that for applications that don't support fallback or address family selection there are the following hostnames:
www.sixxs.net - AAAA + A
www.ipv4.sixxs.net - only AAAA
www.ipv4.sixxs.net - only A
Same scheme for all other hostnames.
IPnG unreach?
Carmen Sandiego on Monday, 15 September 2003 15:02:51
Thanks, I didn't know that.
Jeroen, could you do me a favour and see if my tunnel (3ffe:8114:1000::3bf) looks okay from your end of the world? If I look at the output from, say, "tcpdump 'ip proto 41'", it seems to me that my ICMPv6 requests and replies are symmetrical. Yet the tunnel information page fails to acknowledge that my tunnel is alive.
I have no problems using my SixXS connection now, and I have tried to trace my hosts via various endpoints on http://www.traceroute6.org/ - everything looks fine to me.
Have I overlooked something?
Cheers,
// Klaus
IPnG unreach?
--- 3ffe:8114:1000::3bf ping6 statistics ---
50 packets transmitted, 0 packets received, 100% packet loss
Packets are going out but nothing is coming back...
16:58:20.625187 tunnelserver.ipng.nl > hq.szn.dk: gw-480.ams-02.nl.sixxs.net > cl-480.ams-02.nl.sixxs.net: icmp6: echo request (DF)
16:58:21.624941 tunnelserver.ipng.nl > hq.szn.dk: gw-480.ams-02.nl.sixxs.net > cl-480.ams-02.nl.sixxs.net: icmp6: echo request (DF)
Check your routing.
IPnG unreach?
Before anybody asks, IPv6 works:
jeroen@purgatory:~$ traceroute6 noc.sixxs.net
traceroute to noc.sixxs.net (2001:838:1:1:210:dcff:fe20:7c7c) from 3ffe:8114:2000:240:290:27ff:fe24:c19f, 30 hops max, 16 byte packets
1 gw-20.ams-02.nl.sixxs.net (3ffe:8114:1000::26) 20.424 ms 19.747 ms 19.495 ms
2 Amsterdam.core.ipv6.intouch.net (2001:6e0::2) 20.099 ms 20.203 ms 20.198 ms
3 ams-ix.ipv6.concepts.nl (2001:7f8:1::a501:2871:1) 21.035 ms 20.346 ms 20.251 ms
4 se1.breda.ipv6.concepts-ict.net (2001:838:0:10::2) 24.087 ms 26.191 ms 23.578 ms
5 noc.sixxs.net (2001:838:1:1:210:dcff:fe20:7c7c) 23.161 ms 22.905 ms 23.807 ms
(Hop 1 is the IPng POP :)
And, thus, so does IPv4:
jeroen@purgatory:~$ traceroute ipng.nl
traceroute to ipng.nl (212.19.192.219), 64 hops max, 40 byte packets
1 nrp-1.bras-1.ams-tel.cistron.net (195.64.92.1) 19 ms 19 ms 19 ms
2 ve-4.rtr-1.ams-sar.cistron.net (62.216.29.1) 19 ms 19 ms 18 ms
3 ams-ix.intouch.net (195.69.144.93) 17 ms 18 ms 18 ms
4 0161.nikhef.intouch.terrabone.net (212.26.220.161) 19 ms 19 ms 19 ms
5 tunnelserver.ipng.nl (212.19.192.219) 20 ms 19 ms 19 ms
Transit over Lambdanet just seems to be peeped up. Let's hope, for you that they fix it soon. BTW check the Traceroute tool or the IPng traceroute.
IPnG unreach?
![[nl]](/s/countries/nl.gif) Shadow Hawkins on Wednesday, 17 September 2003 22:55:27
This 'outage' appears to be related to the renumbering effort at the AMS-IX during the last few weeks. Within that time-frame the entire exchange had to renumber their peering interfaces which caused quite a number of unexpected problems.
Although my provider has direct peering to intouch, I too could not reach ipng and saw an IPv4 route via lambdanet (hmm, a common denominator) which did not work.
So all in all, there (probably) was no outage on either side, but just an unexpected side-effect of renumbering the ams-ix. Nothing too bad, just internet.
IPnG unreach?
![[fi]](/s/countries/fi.gif) Carmen Sandiego on Sunday, 14 September 2003 18:02:50
i have same problem.
from my home it works, but from tunnelrouter does not work and from other location not working too.
-- DEBUG FROM HOME: (ip works)
traceroute to 212.19.192.219 (212.19.192.219), 30 hops max, 38 byte packets
1 baana224-1.baanapalvelut.net (213.186.224.1) 8.859 ms 9.409 ms 8.498 ms
2 172.29.0.25 (172.29.0.25) 8.970 ms 8.555 ms 9.813 ms
3 * * *
4 hki2v6.kv9.net (217.78.198.53) 235.428 ms 179.491 ms 96.342 ms
5 hki1r3.kv9.net (217.78.198.29) 98.069 ms * 18.126 ms
6 POS0-1.GW3.HEL2.ALTER.NET (146.188.50.57) 21.016 ms 18.214 ms POS6-1.GW3.HEL2.ALTER.NET (146.188.53.85) 19.584 ms
7 so-1-0-0.XR1.HEL2.ALTER.NET (146.188.12.106) 18.649 ms 16.614 ms 19.883 ms
8 so-4-0-0.TR1.STK2.ALTER.NET (146.188.2.205) 29.642 ms 28.673 ms 29.109 ms
9 so-5-1-0.TR2.AMS2.ALTER.NET (146.188.3.21) 58.415 ms 62.259 ms 63.686 ms
10 POS1-0.BR1.AMS3.ALTER.NET (146.188.3.214) 61.860 ms 57.857 ms 59.913 ms
11 AMS-IX.AMS-1-eth110.nl.lambdanet.net (195.69.144.212) 57.732 ms 64.043 ms 62.489 ms
12 InTouch-NV.nl.lambdanet.net (217.71.99.106) 59.097 ms 60.211 ms 57.282 ms
13 217.8.101.242 (217.8.101.242) 62.130 ms 61.396 ms 63.995 ms
14 * * *
15 tunnelserver.ipng.nl (212.19.192.219) 64.324 ms 62.091 ms 58.607 ms
-- DEBUG FROM TUNNELSERVER: (not working)
traceroute to 212.19.192.219 (212.19.192.219), 30 hops max, 40 byte packets
1 router.pyh.titanix.net (10.140.23.65) 0.952 ms 0.885 ms 0.963 ms
2 10.98.56.1 (10.98.56.1) 18.702 ms 27.001 ms 14.309 ms
3 192.168.51.153 (192.168.51.153) 11.566 ms 17.650 ms 16.162 ms
4 elisa.gw.netsonic.fi (194.29.197.229) 19.152 ms 12.818 ms 23.757 ms
5 adslgw4.netsonic.fi (194.29.192.73) 22.862 ms 16.772 ms 16.304 ms
6 jippii.gw.netsonic.fi (81.17.196.4) 18.014 ms 17.337 ms 18.792 ms
7 ge0-1-0-752.br1.esp1.fi.eunetip.net (213.192.191.85) 13.994 ms 13.996 ms 22.265 ms
8 so1-0-0-0.r1.ffm1.de.eunetip.net (213.192.191.145) 42.809 ms 41.597 ms 46.800 ms
9 de-cix.d-f-8-ge130.de.lambdanet.net (80.81.192.74) 47.662 ms 59.045 ms 46.894 ms
10 F-4-eth200-0.de.lambdanet.net (217.71.105.41) 43.921 ms 52.161 ms 48.770 ms
11 D-2-eth310-0.de.lambdanet.net (217.71.105.81) 50.149 ms 47.068 ms 44.865 ms
12 TE2-1-pos130.nl.lambdanet.net (217.71.96.89) 61.480 ms 144.463 ms 81.074 ms
13 InTouch-NV.nl.lambdanet.net (217.71.99.106) 60.024 ms 59.149 ms 62.263 ms
14 * * *
15 * * *
16 * * *
-- DEBUG FROM OTHER PLACE: (not working)
traceroute to 212.19.192.219 (212.19.192.219), 30 hops max, 38 byte packets
1 salpaus-gw.edu.lahti.fi (143.51.164.1) 0.570 ms 0.499 ms 0.435 ms
2 193.209.2.185 (193.209.2.185) 1.060 ms 0.677 ms 0.611 ms
3 193.209.2.177 (193.209.2.177) 1.850 ms 2.815 ms 1.676 ms
4 172.22.195.98 (172.22.195.98) 4.721 ms 4.034 ms 2.803 ms
5 172.22.195.102 (172.22.195.102) 3.414 ms 3.115 ms 3.032 ms
6 hkicore3-ge3-0-0-v20.datanet.tele.fi (193.227.225.213) 4.439 ms 3.853 ms 3.977 ms
7 hkicore1-ge2-2-1-1.datanet.tele.fi (193.227.224.150) 4.372 ms 4.164 ms 4.512 ms
8 s-bb1-pos1-2-0.telia.net (213.248.66.1) 11.984 ms 11.726 ms 11.526 ms
9 hbg-bb1-pos0-2-0.telia.net (213.248.64.78) 32.175 ms * *
10 * hbg-b1-pos2-0.telia.net (213.248.65.46) 32.098 ms *
11 * lambdanet-5001-hbg-b1.c.telia.net (213.248.77.178) 29.605 ms 27.804 ms
12 D-2-pos120.de.lambdanet.net (217.71.105.69) 43.096 ms * 42.917 ms
13 TE2-1-pos130.nl.lambdanet.net (217.71.96.89) 42.608 ms 42.287 ms 42.300 ms
14 InTouch-NV.nl.lambdanet.net (217.71.99.106) 46.929 ms 46.770 ms 46.466 ms
15 * * *
16 * * *
17 *
looks like IPNG haves bad routing problem at moment. :(
-Riku
IPnG unreach?
looks like IPNG haves bad routing problem at moment.
Intouch is, IPng is just a client using their IP space.
IPnG unreach?
Carmen Sandiego on Monday, 15 September 2003 16:31:13
What a bummer for my Uptime stats lol :)
Anyway just for info, connectivity was back up for a day or two, now again it's down, since monday 12:00.
My graphs also seem to confirm this. Time to open a POP in Belgium! *eg* ;)
(Belnet would be nice)
IPnG unreach?
Carmen Sandiego on Monday, 15 September 2003 16:51:16
And one in Denmark, too! :9-
IPnG unreach?
There is a rumor going on about something in that direction.
IPnG unreach?
Redirect them to POP requirements and it can be arranged.
|
![[ch]](/s/countries/ch.gif)
on Sunday, 14 September 2003 02:16:47
Posting is only allowed when you are