TSTT Forum: Server Routing Problem - TSTT Forum

Jump to content

Page 1 of 1

Server Routing Problem cannot reach server

#1 User is offline   admin Icon

  • Administrator
  • PipPipPipPipPipPipPip
  • Group: Admin
  • Posts: 817
  • Joined: 28-July 03

Posted 10 December 2010 - 01:23 AM

About a bit over one and a half hours ago all my ssh sessions to my main server (TSTTproblems.com and others) dropped. In addition I couldn't receive email and then I realised I couldn't access the server at all via Blink. But I had no problems accessing the same server via Flow. There must be a routing problem to the US, which affects only a few IP addresses. Tested accessing the same server via a different interface (IP) and it worked fine.

Traceroute via Blink:

Quote

Fri Dec 10 00:35:01 AST 2010
traceroute to www.tsttproblems.com (64.34.161.248), 30 hops max, 40 byte packets
1 firewall
2 186-45-112-1.dynamic.tstt.net.tt (186.45.112.1) 95.432 ms 97.862 ms 105.642 ms
3 201.238.123.13.business.static.tstt.net.tt (201.238.123.13) 100.485 ms 103.205 ms 108.022 ms
4 201.238.77.21.business.static.tstt.net.tt (201.238.77.21) 110.655 ms 128.849 ms 131.360 ms
5 gige7-13.usa.nmi-edge05.columbus-networks.com (63.245.68.81) 174.482 ms 182.727 ms 190.515 ms
6 * * *
7 xe-3-1-0.usa.nam-core02.columbus-networks.com (63.245.5.71) 194.612 ms 129.957 ms 140.659 ms
8 * * *
9 oc48-po6-1.atl-telx-cor-1.peer1.net (216.187.124.166) 149.268 ms 154.542 ms 139.206 ms
10 * * *
11 * * *
12 * * *
13 216.187.120.230 (216.187.120.230) 123.023 ms 116.032 ms 118.740 ms
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * * ... max number of hops reached ...


Same traceroute via Flow:

Quote

traceroute to www.tsttproblems.com (64.34.161.248), 30 hops max, 40 byte packets
1 firewall
2 * * *
3 200.1.111.177 (200.1.111.177) 101.503 ms 106.665 ms 114.306 ms
4 200.1.111.78 (200.1.111.78) 104.210 ms 109.092 ms 117.096 ms
5 200.1.111.63 (200.1.111.63) 111.758 ms 119.555 ms 121.983 ms
6 gige7-12.usa.nmi-edge05.columbus-networks.com (63.245.68.69) 170.488 ms 169.229 ms 169.216 ms
7 xe-1-1-3.usa.nmi-teracore02.columbus-networks.com (63.245.5.146) 168.871 ms 99.741 ms 104.969 ms
8 xe-4-0-0.usa.nam-core02.columbus-networks.com (63.245.5.68) 105.265 ms 105.343 ms 120.154 ms
9 (198.32.124.125) 120.571 ms 120.284 ms 135.263 ms
10 oc48-po6-1.atl-telx-cor-1.peer1.net (216.187.124.166) 135.721 ms 226.639 ms 232.047 ms
11 10ge-ten1-1.atl-101mar-cor-1.peer1.net (216.187.120.226) 232.123 ms 232.095 ms 211.634 ms
12 10ge.xe-1-0-0.wdc-eqx-dis-1.peer1.net (216.187.115.37) 231.623 ms 231.608 ms 231.800 ms
13 10ge.ten1-2.wdc-sp2-cor-1.peer1.net (216.187.115.234) 221.898 ms 221.383 ms 221.306 ms
14 216.187.120.238 (216.187.120.238) 180.821 ms 180.396 ms 133.869 ms
15 www.tsttproblems.com (64.34.161.248) 137.579 ms 137.300 ms 137.027 ms


Doing a traceroute via Blink to a different interface on the same server is fine:

Quote

traceroute to 64.34.211.163 (64.34.211.163), 30 hops max, 40 byte packets
1 my.firewall (192.168.5.254) 3.419 ms 6.467 ms 9.448 ms
2 186-45-112-1.dynamic.tstt.net.tt (186.45.112.1) 144.115 ms 146.566 ms 149.175 ms
3 201.238.123.5.business.static.tstt.net.tt (201.238.123.5) 82.429 ms 84.806 ms 113.313 ms
4 201.238.77.21.business.static.tstt.net.tt (201.238.77.21) 116.009 ms 118.692 ms 121.095 ms
5 gige7-13.usa.nmi-edge05.columbus-networks.com (63.245.68.81) 175.252 ms 183.441 ms 191.753 ms
6 * * *
7 * * *
8 * * *
9 oc48-po6-1.atl-telx-cor-1.peer1.net (216.187.124.166) 136.009 ms 121.163 ms 127.490 ms
10 * * *
11 * * *
12 * * *
13 216.187.120.234 (216.187.120.234) 128.167 ms 141.269 ms 143.736 ms
14 ns3.ultragate.com (64.34.211.163) 157.331 ms 162.539 ms 167.667 ms


Hope this routing problem is fixed in the morning, otherwise quite a number of customers will not be able receiving their emails and accessing their sites.
Contacted the hosting company ...

Update: Did more testing and it seems that only certain TSTT IP addresses are affected. Not even a range of addresses, but particular addresses. Strange I would say ...

... maybe wikileaks related :ph34r:
0

#2 User is offline   admin Icon

  • Administrator
  • PipPipPipPipPipPipPip
  • Group: Admin
  • Posts: 817
  • Joined: 28-July 03

Posted 10 December 2010 - 06:14 PM

This is what the hosting company told me:

Quote

We heard back from Sprint and they found that the traffic is being dropped within Telecom Services of Trinidad & Tobago. Our Noc Team has confirmed from the Routeviews BGPlay that AS5639 appears to be having issues and was withdrawing the 186.45.112.0/21 prefix yesterday. It appears that the issues are within the Telecom Services of Trinidad & Tobago network and we would recommend that you follow up with them if this issue arises again.


Looks like TSTT is/was having some routing issues.
0

#3 User is offline   greall Icon

  • Telecom Master
  • PipPipPipPipPipPipPip
  • Group: Members
  • Posts: 1,433
  • Joined: 28-July 05
  • Location:Bel Air,La Romain

Posted 10 December 2010 - 09:52 PM

View Postadmin, on 10 December 2010 - 06:14 PM, said:

This is what the hosting company told me:



Looks like TSTT is/was having some routing issues.


That would explain the hair pulling browsing experience tonight...

Greg
0

#4 User is offline   admin Icon

  • Administrator
  • PipPipPipPipPipPipPip
  • Group: Admin
  • Posts: 817
  • Joined: 28-July 03

Posted 24 December 2010 - 11:36 PM

Looks like I am having an intermittent Internet routing problem again.
My ssh sessions to a particular server drop frequently. If I log in via a different route or via flow, I can reach my server without problems.
Beside my server, many other sites including www.google.com are affected.

Quote

traceroute www.google.com
traceroute to www.google.com (74.125.67.105), 30 hops max, 40 byte packets
1 my.firewall (192.168.5.254) 3.519 ms 6.579 ms 9.706 ms
2 186-45-80-1.dynamic.tstt.net.tt (186.45.80.1) 106.494 ms 111.534 ms 114.041 ms
3 201.238.123.5.business.static.tstt.net.tt (201.238.123.5) 108.975 ms 116.513 ms 119.021 ms
4 201.238.77.21.business.static.tstt.net.tt (201.238.77.21) 121.994 ms 124.507 ms 127.000 ms
5 gige7-13.usa.nmi-edge05.columbus-networks.com (63.245.68.81) 181.003 ms 183.832 ms 186.826 ms
6 xe-1-0-1.usa.nmi-teracore02.columbus-networks.com (63.245.5.96) 189.368 ms 188.131 ms 187.923 ms
7 63.245.6.46 (63.245.6.46) 221.403 ms 176.579 ms 177.020 ms
8 209.85.253.120 (209.85.253.120) 144.779 ms 144.704 ms 209.85.253.116 (209.85.253.116) 130.689 ms
9 216.239.48.192 (216.239.48.192) 155.068 ms 166.228 ms 209.85.254.252 (209.85.254.252) 178.070 ms
10 * * *
11 209.85.255.198 (209.85.255.198) 138.008 ms * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *


Here it works a minute later ... and then it doesn't ... and ...:

Quote

traceroute www.google.com
traceroute to www.google.com (74.125.67.99), 30 hops max, 40 byte packets
1 my.firewall (192.168.5.254) 3.504 ms 6.623 ms 9.701 ms
2 186-45-80-1.dynamic.tstt.net.tt (186.45.80.1) 88.626 ms 91.115 ms 98.894 ms
3 201.238.123.13.business.static.tstt.net.tt (201.238.123.13) 93.859 ms 98.862 ms 101.350 ms
4 201.238.77.21.business.static.tstt.net.tt (201.238.77.21) 154.663 ms 157.489 ms 159.995 ms
5 gige7-13.usa.nmi-edge05.columbus-networks.com (63.245.68.81) 162.989 ms 182.791 ms 185.506 ms
6 xe-0-0-1.usa.nmi-teracore02.columbus-networks.com (63.245.5.84) 194.129 ms 124.010 ms 137.955 ms
7 63.245.6.34 (63.245.6.34) 140.515 ms 145.265 ms 145.573 ms
8 209.85.253.120 (209.85.253.120) 145.557 ms 209.85.253.116 (209.85.253.116) 167.172 ms 169.692 ms
9 209.85.254.252 (209.85.254.252) 171.145 ms 216.239.48.192 (216.239.48.192) 127.410 ms 130.153 ms
10 209.85.254.249 (209.85.254.249) 132.471 ms * *
11 209.85.255.190 (209.85.255.190) 132.060 ms 209.85.255.198 (209.85.255.198) 131.224 ms 209.85.255.190 (209.85.255.190) 125.295 ms
12 gw-in-f99.1e100.net (74.125.67.99) 124.215 ms 133.332 ms 129.100 ms

0

#5 User is offline   BaM Icon

  • Newbie
  • Pip
  • Group: Members
  • Posts: 5
  • Joined: 09-June 10

Posted 12 June 2011 - 12:51 PM

Maybe it has something to do with the IPv6 switchover?
Not sure really.
It was working fine for about 10 minutes this morning, and now it isnt working again.

Trace route come up as almost the same as yours.

FML, looks like i will be using the Vidiala network till it is fixed.
0

Page 1 of 1


Fast Reply

  

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users