Please help!

We are having problems accessing all our and our clients sites that we host in the US. problems started being reported by our clients about a week ago and were intermittent, now the problem is severe enough that we can no longer work on the sites.


Chrome browser reports the error as:
Error 101 (net::ERR_CONNECTION_RESET): Unknown error.

Best example of the problem can be seen here:

http://www.howickcollege.school.nz/content/view/104/193/

first row of 4 photos (directly under the text Daffodil Day - Friday 27 August)
currently 1st and 4th showing, 2nd and 3rd showing as broken images
refresh the page
1st, 2nd and 4th images show
3rd broken image
refresh the page again
1st, 2nd, 3rd and 4th images all show as broken etc
all image files are physically on the server

Try to open the page images while the page is loading and you get the connection reset error.... wait a few seconds after the page has finished and refresh the image and it's OK.


This appears to only affect our clients using Telecom Xtra,  
Response from our hosting company was as follows:



Again I confirm that the server itself appears fine and is operating
well within what is considered healthy. Traceroutes back to your
location running for long periods do reveal an issue however. One of
the global gateways at your ISP has severely high packet loss which is
affecting hops after it. Considering that no other simple explanation
can be found, this appears to be the best answer for the problem. I'm
currently pulling more than 90% packet loss through this hop, which is
creating little jitters in packet loss around it.

I would recommending contacting your ISP with this information and
having them investigate.

Specifically on hop 15.

Packets Pings
Host Loss% Last Avg Best Wrst StDev
1. c1.bd.84ae.static.theplanet.com 1.4% 0.5 2.4 0.3 16.0 4.2
2. te6-1.dsr02.dllstx2.theplanet.com 1.4% 0.3 2.4 0.3 80.3 11.4
3. te9-4.dsr02.dllstx3.theplanet.com 1.4% 0.6 2.0 0.5 69.4 8.4
4. et3-2.ibr03.dllstx3.theplanet.com 1.4% 0.5 0.7 0.4 9.2 1.4
5. xe-0-5-0-4.r07.dllstx09.us.bb.gin.ntt.net 1.4% 1.0 0.9 0.6 2.5 0.2
6. ae-2.r21.dllstx09.us.bb.gin.ntt.net 1.4% 0.7 2.9 0.6 25.5 5.4
7. ae-4.r21.snjsca04.us.bb.gin.ntt.net 1.4% 91.0 49.3 43.7 98.2 13.4
8. ae-0.r20.plalca01.us.bb.gin.ntt.net 1.4% 45.0 43.9 42.7 46.7 1.0
9. po-1.r03.plalca01.us.bb.gin.ntt.net 0.0% 42.8 49.8 42.5 254.2 27.6
10. ge-1-2.r03.plalca01.us.ce.gin.ntt.net 1.4% 43.7 44.0 42.6 53.0 1.7
11. ???
12. so5-0-1.akbr4.global-gateway.net.nz 0.0% 194.8 189.1 163.7 215.2 10.7
so5-0-2.akbr4.global-gateway.net.nz
so1-0-0.akbr4.global-gateway.net.nz
13. ae1-10.tkbr9.global-gateway.net.nz 1.4% 191.2 188.2 163.9 221.8 12.5
14. 210.55.203.202 1.4% 164.4 187.3 164.1 194.4 10.4
15. glf-fid-int.tkbr10.global-gateway.net.nz 95.7% 29808 29733 29655 29808 76.2
16. 222-154-96-1.jetstream.xtra.co.nz 1.4% 196.6 191.2 167.1 235.2 13.7
17. ???


For their testing purposes, the servers primary IP is 174.132.189.194,
and this traceroute was to your IP of 222.154.105.100

Cheers,

James Lavoy
Linux Systems Administrator



Can anyone help confirm the problem and if it's only xtra customers? I have contact Telecom and working my way up the support levels but it's taking forever to get to someone that can help or even understand the problem, after 20mins of being asked what make of modem I'm using each time by various call center staff and trying to explain it's not related, I need to know if there is anything we can be doing before we start losing clients.

Other sites affected (although much more intermittent include:

designpath.co.nz
bee.geek.nz

pings to the servers are not a problem even when the errors occur

- Andy