|
Ticket ID: SIXXS #844248 Ticket Status: User PoP:
Pop reports my endpoint down for over 11 hrs but it is up
![[pl]](/s/countries/pl.gif) Shadow Hawkins on Sunday, 09 November 2008 21:38:15
I have read and followed the "Reporting Problems" section on the Contact page and am providing the following details for this report based on the list of items stated there:
Hi
I've just checked details of my tunnel T17640 and noticed that it was last alive on 2008-11-09 10:12:07 CET (over 11 hrs ago). I've checked the site and it works - I can ping POP from there:
yenna:~> ping6 uslax01.sixxs.net
PING uslax01.sixxs.net(2001:4de0:1000:a5::2) 56 data bytes
64 bytes from 2001:4de0:1000:a5::2: icmp_seq=0 ttl=64 time=32.7 ms
64 bytes from 2001:4de0:1000:a5::2: icmp_seq=1 ttl=64 time=30.1 ms
64 bytes from 2001:4de0:1000:a5::2: icmp_seq=2 ttl=64 time=29.6 ms
and I can ping and traceroute my endpoint from my other site:
tkepczyx-mobl1:~> traceroute6 yenna.jot23.org
traceroute to yenna.jot23.org (2001:1938:82:d::2), 30 hops max, 80 byte packets
1 router.jot23.org (fdb1:8565:c849:1:21e:e5ff:fe43:b823) 0.571 ms 1.294 ms 1.258 ms
2 gw-24.waw-01.pl.sixxs.net (2001:6a0:200:17::1) 15.452 ms 15.929 ms 16.157 ms
3 icm-r1-2001-6a0-0-a000-0-0-0-d.rtr6.net.icm.edu.pl (2001:6a0:0:a000::d) 18.633 ms 18.863 ms 19.099 ms
4 2001:b10:c000:3::5 (2001:b10:c000:3::5) 19.081 ms 20.299 ms 20.282 ms
5 2001:b10:c000:1::14 (2001:b10:c000:1::14) 24.221 ms 24.453 ms 24.932 ms
6 pionier-bckp.rt1.poz.pl.geant2.net (2001:798:23:10aa::5) 25.154 ms 23.636 ms 24.093 ms
7 so-7-1-0.rt1.fra.de.geant2.net (2001:798:cc:1401:2301::1) 41.307 ms 40.146 ms 40.102 ms
8 so-5-0-0.rt1.ams.nl.geant2.net (2001:798:cc:1401:2201::2) 47.060 ms 46.904 ms 46.407 ms
9 so-4-0-0.rt1.lon.uk.geant2.net (2001:798:cc:2201:2801::2) 54.627 ms 54.855 ms 54.895 ms
10 2001:450:2002:6f::1 (2001:450:2002:6f::1) 55.120 ms 54.902 ms 55.099 ms
11 2001:450:2002:6e::2 (2001:450:2002:6e::2) 106.547 ms 107.271 ms 105.874 ms
12 5-3.r2.lo.v6.hwng.net (2001:4de0:1000:24::1) 105.846 ms 112.352 ms 104.347 ms
13 1-3.r3.lo.v6.hwng.net (2001:4de0:1000:25::1) 104.065 ms 103.865 ms 102.135 ms
14 5-2.r2.am.v6.hwng.net (2001:4de0:1000:26::2) 102.863 ms 111.373 ms 110.619 ms
15 5-1.r1.am.v6.hwng.net (2001:4de0:1000:1::1) 103.091 ms 103.320 ms 103.884 ms
16 1-1.r1.dc.v6.hwng.net (2001:4de0:1000:4::2) 197.829 ms 195.791 ms 196.066 ms
17 1-2.r1.ny.v6.hwng.net (2001:4de0:1000:11::1) 195.544 ms 2-3.r2.dc.v6.hwng.net (2001:4de0:1000:9::2) 196.521 ms 195.798 ms
18 3-1.r1.ch.v6.hwng.net (2001:4de0:1000:13::2) 215.330 ms 2-1.r1.sj.v6.hwng.net (2001:4de0:1000:19::1) 268.775 ms 3-1.r1.ch.v6.hwng.net (2001:4de0:1000:13::2) 217.564 ms
19 1-2.r1.la.v6.hwng.net (2001:4de0:1000:18::2) 272.045 ms 270.526 ms 1-1.r1.la.v6.hwng.net (2001:4de0:1000:17::2) 261.807 ms
20 2001:4de0:1000:a5::2 (2001:4de0:1000:a5::2) 261.813 ms 260.541 ms 271.015 ms
21 cl-14.lax-01.us.sixxs.net (2001:1938:82:d::2) 241.260 ms 240.029 ms 237.026 ms
Distributed traceroute also works:
Hop Node Loss% Sent Last Avg Best Worst StDev ASN Organisation
1. 2001:838:1:1::1 0.0% 5 0.5 0.5 0.5 0.5 0.0 [.nl] Netherlands, The 12871 Concepts ICT
ge-1-3-0.breda.ipv6.concepts-ict.net.
2. 2001:838:0:10::1 0.0% 5 2.5 2.6 2.4 3.1 0.3 [.nl] Netherlands, The 12871 Concepts ICT
3. 2001:7f8:1::a501:2989:1 0.0% 5 3.6 10.9 3.6 19.7 7.0 [.nl] Netherlands, The AMS-IX-IPV6
ams-ix.hwng.net.
4. 2001:4de0:1000:1::1 0.0% 5 13.9 5.4 3.2 13.9 4.8 [.nl] Netherlands, The 12989 Eweka Internet Services
5-1.r1.am.v6.hwng.net.
5. 2001:4de0:1000:4::2 0.0% 5 103.7 108.5 99.9 117.8 6.9 [.nl] Netherlands, The 12989 Eweka Internet Services
1-1.r1.dc.v6.hwng.net.
6. 2001:4de0:1000:9::2 0.0% 5 104.9 101.2 100.1 104.9 2.1 [.nl] Netherlands, The 12989 Eweka Internet Services
2-3.r2.dc.v6.hwng.net.
7. 2001:4de0:1000:19::1 0.0% 5 181.8 184.3 181.5 190.4 3.6 [.nl] Netherlands, The 12989 Eweka Internet Services
2-1.r1.sj.v6.hwng.net.
8. 2001:4de0:1000:18::2 0.0% 5 189.8 190.3 189.5 193.0 1.5 [.nl] Netherlands, The 12989 Eweka Internet Services
1-2.r1.la.v6.hwng.net.
9. 2001:4de0:1000:a5::2 0.0% 5 189.6 189.6 189.4 189.8 0.2 [.nl] Netherlands, The 12989 Eweka Internet Services
10. 2001:1938:82:d::2 20.0% 5 219.1 219.3 219.0 220.1 0.5 [.us] United States 12989 Highwinds Network Group Inc
cl-14.lax-01.us.sixxs.net.
There is no firewall on my site:
yenna:~> sudo ip6tables-save
yenna:~> sudo iptables-save
yenna:~>
So the question is - why is POP not detecting that my side of a tunnel is up?
Pop reports my endpoint down for over 11 hrs but it is up
Please read the FAQ
The first hop of your traceroute is:
1 router.jot23.org (fdb1:8565:c849:1:21e:e5ff:fe43:b823) 0.571 ms 1.294 ms 1.258 ms
That is a ULA address, that can't be correct, as that is not globally routeable; next to that ping tests are performed between your endpoint and the PoP and not around the world.
Pop reports my endpoint down for over 11 hrs but it is up
Shadow Hawkins on Monday, 10 November 2008 20:47:28
I do not control the address the kernel chooses to respond. And linux 2.4 has been known to me to choose non rfc conformant address.
But that remark misses the point - the address I had the problem with is: 2001:1938:82:d::2 (which I do hope I clearly stated). This address is not properly monitored from pop (Last Alive 2008-11-09 10:12:07 CET - I've just copied and pasted it). Seeing the last alive I should have been charged 5 credits which hasn't happend as well.
I've left tcpdump on my sit tunnel - I haven't seen any pings coming in (and they supposedly are coming every 30 minutes) but I've seen some multicast traffic.
So the problem is - uslax01 POP does not properly monitor tunnels from it.
Pop reports my endpoint down for over 11 hrs but it is up
I do not control the address the kernel chooses to respond.
Configure it correctly and it will respond from the correct address.
As the tunnel endpoints are being used to ping, the response should always be correct.
And linux 2.4 has been known to me to choose non rfc conformant address.
Where is the bug report about this?
So the problem is - uslax01 POP does not properly monitor tunnels from it.
It works fine.
Pop reports my endpoint down for over 11 hrs but it is up
Shadow Hawkins on Tuesday, 11 November 2008 16:51:22
1. router.jot23.org (fdb1:8565:c849:1:21e:e5ff:fe43:b823) doesn't really matter. The problem was on last hop, not the first.
2. Yes, uslax01 started pinging around 21:00-22:00 yesterday. So someone probably fixed it in the meantime.
Pop reports my endpoint down for over 11 hrs but it is up
ULA addresses are filtered at a lot of locations, thus all traffic (including ICMP or UDP used for those traceroutes) are dropped and thus might exactly cause what you are showing in your output.
State change: user
The state of this ticket has been changed to user
|
![[ch]](/s/countries/ch.gif)
on Monday, 10 November 2008 11:31:55
Posting is only allowed when you are