|
Ticket ID: SIXXS #8562490 Ticket Status: User PoP: brudi01 - (Uberlandia)
T42593 uptime pings error
![[uy]](/s/countries/uy.gif) Shadow Hawkins on Monday, 07 January 2013 02:37:22
Greetings. I am having some issues with the tunnelrobot that pings my tunnel to make sure it's "up". The tunnel has been up for countless days and I even started sending out constant ping packets to make sure that traffic flows through the tunnel. Now my question is, how can the graphs and tunnelrobot say that my tunnel is down, and doesn't reply to pings (has happened twice already), when the constant ping to the tunnel replies just fine? I've even checked just now that the tunnel has gotten 26 MB just from pings. Is this an error on the tunnelrobot or am I missing something here? The tunnel is currently setup in a VPS in Chicago, due to my ISK count being in the negatives, so as long as the VPS is up, the tunnel should as well. I may have some elevated latency, but that doesn't stop the tunnel from replying to pings. The graphs show a 100% loss, but that's impossible since the constant pings come back 99%.
T42593 uptime pings error
Shadow Hawkins on Monday, 07 January 2013 02:54:54
VPS is Debian 6.0.6, 2.6.32-5-amd64, current setup is static via network interfaces:
auto sixxs
iface sixxs inet6 v4tunnel
address 2001:1291:200:da::2
netmask 64
endpoint 201.48.254.14
ttl 64
up ip link set mtu 1480 dev sixxs
up ip route add 2001:1291::/32 via 2001:1291:200:da::2 dev sixxs
up ip addr change 2001:1291:200:da::2 dev sixxs preferred_lft 0
Routing table:
root@Anubis:~# route -nv6
Kernel IPv6 routing table
Destination Next Hop Flag Met Ref Use If
2001:1291:200:da::/64 :: Un 256 0 2995 sixxs
2605:f700:xxx:1::xxx:8cf9/128 :: U 256 0 0 eth0
2605:f700:xxx:1::xxx:fec8/128 :: U 256 0 0 eth0
2605:f700:xxx:1::xxx:c18f/128 :: U 256 0 0 eth0
2605:f700:xxx:1::/64 :: U 256 0 1 eth0
fe80::/64 :: Un 256 0 0 sixxs
fe80::/64 :: U 256 0 0 eth0
::/0 2605:f700:xxx:1::1 UG 1 6 1536 eth0
::/0 :: !n -1 1585237 lo
::1/128 :: Un 0 1 2186 lo
2001:1291:200:da::2/128 :: Un 0 1202513 lo
2605:f700:xxx:1::xxx:8cf9/128 :: Un 0 1 935 lo
2605:f700:xxx:1::xxx:fec8/128 :: Un 0 2 23753 lo
2605:f700:xxx:1::xxx:c18f/128 :: Un 0 3 13460 lo
2605:f700:xxx:1::xxx:1d76/128 :: Un 0 6 33024 lo
ff00::/8 :: U 256 0 0 sixxs
ff00::/8 :: U 256 0 0 eth0
root@Anubis:~#
(trimmed/edited out sensitive information)
Traceroute to the v4 PoP:
root@Anubis:~# mtr -4 -w -r -c 10 brudi01.sixxs.net
HOST: Anubis Loss% Snt Last Avg Best Wrst StDev
1. 67.202.105.129 0.0% 10 0.3 0.4 0.3 0.6 0.1
2. ip2.50-200-2.static.steadfastdns.net 0.0% 10 0.8 3.7 0.4 22.3 7.2
3. ip45.208-100-32.static.steadfastdns.net 0.0% 10 162.3 226.0 128.6 317.0 67.8
4. xe-0-3-0-6.r06.chcgil09.us.bb.gin.ntt.net 0.0% 10 0.7 0.8 0.7 1.4 0.2
5. xe-1.globalcrossing.chcgil09.us.bb.gin.ntt.net 0.0% 10 0.5 0.6 0.5 0.6 0.0
6. ctbc-telecom.ge-1-3-2.ar4.gru1.gblx.net 0.0% 10 201.6 202.0 196.0 207.9 3.5
7. xe-1-2-0-0.core-b.spo511.ctbc.com.br 0.0% 10 228.4 222.1 212.7 240.2 8.6
8. ae1-0.core-b.ula001.ctbc.com.br 0.0% 10 216.4 215.9 210.8 222.3 3.5
9. ge-1-0-0-0.edge-b.ula001.ctbc.com.br 10.0% 10 211.3 229.6 211.3 288.2 28.8
10. brudi01.sixxs.net 10.0% 10 207.8 211.4 206.6 219.0 4.5
root@Anubis:~# mtr -6 -w -r -c 10 brudi01.sixxs.net
HOST: Anubis Loss% Snt Last Avg Best Wrst StDev
1. 2605:f700:xxx:1::1 0.0% 10 2.6 0.6 0.3 2.6 0.7
2. 2607:f128:45::1 0.0% 10 0.6 0.9 0.5 3.1 0.8
3. 2607:f128:3:6::2 0.0% 10 0.6 1.7 0.6 11.0 3.3
4. 2607:f128:3:5::1 0.0% 10 0.4 2.7 0.3 12.3 4.6
5. 10gigabitethernet4-1.core1.chi1.he.net 0.0% 10 1.7 2.4 1.3 5.9 1.6
6. 10gigabitethernet1-2.core1.ash1.he.net 0.0% 10 17.0 18.5 16.6 27.6 3.4
7. eqix-dc5.ctbs.com 0.0% 10 191.0 193.5 190.6 204.3 4.4
8. xe-1-1-0-0.core-a.ula001.ipv6.ctbc.com.br 0.0% 10 173.6 164.3 159.3 176.1 7.0
9. ge-1-3-0-0.edge-b.ula001.ipv6.ctbc.com.br 10.0% 10 158.0 158.1 157.5 158.8 0.4
10. brudi01.sixxs.net 10.0% 10 160.1 159.7 159.0 160.3 0.4
root@Anubis:~#
If I am missing something, please let me know.
T42593 uptime pings error
::/0 2605:f700:xxx:1::1 UG 1 6 1536 eth0
Seems you have a default route going somewhere else. One of likely many many problems.
Looks like something native already on the host. If that is the case, then why bother using a tunnel when that is not needed?
State change: user
The state of this ticket has been changed to user
T42593 uptime pings error
I even started sending out constant ping packets to make sure that traffic flows through the tunnel.
The ping test does not care about that. Though a "solution" (mind the quotes as it only hides a problem) like that only keeps the connection going, it should not be needed if you setup your tunnel properly.
when the constant ping to the tunnel replies just fine?
Because of all the factors listed in the FAQ.
I've even checked just now that the tunnel has gotten 26 MB just from pings
Likely only your constant pings though. The PoP really does not send that much.
Is this an error on the tunnelrobot or am I missing something here?
You are missing the portion where you need to read the FAQ.
The graphs show a 100% loss,
Thus clearly you are not responding to the ping tests, read the FAQ to realize why.
but that's impossible since the constant pings come back 99%.
No, that is a completely different kind of ping, that is you pinging something random, not the PoP pinging you and the PoP actually receiving that answer. Read the FAQ!
T42593 uptime pings error
Shadow Hawkins on Monday, 14 January 2013 03:31:46
I fixed the problem temporarily.
The tunnel is on the server because it's an old account of mine, and in my past I've had negative ISK, and I set the tunnel on the server, since it stays up all the time, so the ISK count can go up. The tunnel before had like 6 weeks of uptime, at one point.
I've fixed the problem by moving the endpoint to another server more geographically close to the PoP and it seems to be working fine. I hope that the ISK count starts to go up again..
Thanks, and my bad.
|
![[ch]](/s/countries/ch.gif)
on Thursday, 10 January 2013 13:58:59
Posting is only allowed when you are