- This topic has 48 replies, 15 voices, and was last updated August 30, 2010 at 8:26 am by Debris.
-
AuthorPosts
-
August 21, 2010 at 9:08 am #1049575
I’m working to improve our network configuration with a change coming to our web site very soon, if you’e interested in helping out please run the following commands from your command line (start > run > cmd) and paste the results below. Ideally the more samples I have the better, particularly from a wide variety of geographical locations:
ping 66.228.118.51
ping 67.228.118.21
ping 208.43.118.51Cheers!
August 21, 2010 at 9:12 am #1228488And please don’t paste the whole output, just the final statistics such as:
Ping statistics for 66.228.118.51:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 192ms, Maximum = 212ms, Average = 197msPing statistics for 67.228.118.21:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 184ms, Maximum = 194ms, Average = 187msPing statistics for 208.43.118.51:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 230ms, Maximum = 240ms, Average = 233msAugust 21, 2010 at 9:25 am #1228530Ping statistics for 66.228.118.51:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 145ms, Maximum = 154ms, Average = 147msPing statistics for 67.228.118.21:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 188ms, Maximum = 196ms, Average = 190msPing statistics for 208.43.118.51:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 104ms, Maximum = 117ms, Average = 107msAugust 21, 2010 at 9:27 am #1228513Ping statistics for 66.228.118.51:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 146ms, Maximum = 206ms, Average = 164msPing statistics for 67.228.118.21:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 190ms, Maximum = 195ms, Average = 191msPing statistics for 208.43.118.51:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 112ms, Maximum = 126ms, Average = 116msAugust 21, 2010 at 11:52 am #1228504as at 2010-08-21 12:50 uk local time the first of the 3 servers is inaccessible
— 67.228.118.51 ping statistics —
3 packets transmitted, 0 received, 100% packet loss, time 1999ms
with regard to the other two
— 67.228.118.21 ping statistics —
4 packets transmitted, 4 received, 0% packet loss, time 3002ms
rtt min/avg/max/mdev = 172.845/173.754/174.573/0.614 ms, pipe 2
— 208.43.118.51 ping statistics —
4 packets transmitted, 4 received, 0% packet loss, time 3003ms
rtt min/avg/max/mdev = 104.773/105.320/105.944/0.476 ms, pipe 2August 21, 2010 at 1:05 pm #1228489Interesting can you try the failed one again?
August 21, 2010 at 1:24 pm #1228505no luck – 67.228.118.51 is inaccessible from both my home IP address and logging in remotely to my office PC (7 miles away)
August 21, 2010 at 1:29 pm #1228510whats my command line and where can i find it?
August 21, 2010 at 1:38 pm #1228506I think you’re using XP so its Start Menu > Run then type in “cmd” in the box
type the ping command with IP address into the black “old style computer screen” box what appears.
August 21, 2010 at 1:40 pm #1228515Ping statistics for 66.228.118.51:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 132ms, Maximum = 159ms, Average = 143msPing statistics for 67.228.118.21:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 171ms, Maximum = 202ms, Average = 186msPing statistics for 208.43.118.51:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 114ms, Maximum = 146ms, Average = 133msAugust 21, 2010 at 2:06 pm #1228511how do i cut and paste from there? it wont let me highlight…
August 21, 2010 at 2:09 pm #1228516right click>slect all>ctrl+c
August 21, 2010 at 2:10 pm #1228517@General Lighting 395802 wrote:
no luck – 67.228.118.51 is inaccessible from both my home IP address and logging in remotely to my office PC (7 miles away)
Mby an exchange is down in your area?
August 21, 2010 at 2:12 pm #1228490@General Lighting 395802 wrote:
no luck – 67.228.118.51 is inaccessible from both my home IP address and logging in remotely to my office PC (7 miles away)
Could you traceroute the ip and post the result?
August 21, 2010 at 2:20 pm #1228512@DaftFader 395816 wrote:
right click>slect all>ctrl+c
safe
Ping statistics for 66.228.118.51:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 145ms, Maximum = 147ms, Average = 146msargh! got it to work once, but it wont do it again and im doing the same thing, sorry doc.
August 21, 2010 at 7:24 pm #1228533It’s not network performance test but a ISP performance test.
August 21, 2010 at 10:41 pm #1228507normally it would be but Dr B is trying to balance the load across 3 servers which are presumably all in the same data-centre and on the same network – except for some reason I cannot access to 67.228.118.51.
Can anyone else get to it?
I could try with a mobile phone and my netbook but the mobile I would use has run out of battery charge anyway and TBH I have drunk too much beer now to be bothered to fetch all this equipment, particularly if the routing fault is in the USA anyway!
August 21, 2010 at 11:25 pm #1228519Here’s some MTR’s from a box in maidenhead. If you want any advise give me a shout. You shouldn’t need to load balance anything though, given the amount of traffic you get should be handled no problem with one box… curious why it’s balanced over 3 servers. Anyway here’s the output, given it’s in the US though you’ll get not-so-useful results from the UK. All these go through hurricane electric who handle my transit from London to NYC so the traffic stays on the same network across the water. I’d probably recommend moving to the UK if you want any noticeable improvements from a network POV, especially for latency and jitter.
General Lighting FYI: 67.228.118.51 blocks ICMP echo requests hence why you can’t ping it however the MTR results below use ICMP Time Exceeded which they haven’t blocked.
The traffic in terms of geography follows this path for me:
Hops 1-2: Maidenhead (our datacentre)
Hops 3-4: London (our POP in telehouse east, peering with HE)
Hops 5-7: New York City (HE)
Hops 8-11: Dallas (It would appear softlayer peer with HE also, so it’s actually a pretty efficient route in this case)66.228.118.51:
earth (0.0.0.0) Sun Aug 22 00:14:58 2010
Keys: Help Display mode Restart statistics Order of fields quit
Packets Pings
Host Loss% Snt Last Avg Best Wrst StDev
1. ar3.blsq.xcalibre.net.uk 0.0% 43 0.3 0.4 0.2 2.7 0.4
2. tr1.blsq.xcalibre.net.uk 0.0% 43 0.7 0.8 0.4 4.2 0.5
3. 0-519-br2.the.flexiant.net 0.0% 43 3.6 3.6 3.3 4.8 0.3
4. gige-g4-14.core1.lon1.he.net 0.0% 42 3.9 6.2 3.4 17.6 4.2
5. 10gigabitethernet2-3.core1.nyc4.he.net 0.0% 42 71.8 75.0 71.4 95.0 5.2
6. 10.53.11.2 0.0% 42 71.9 73.7 71.6 86.9 3.9
7. softlayer.tienyc.telx.com 0.0% 42 71.9 75.5 71.7 184.5 17.6
8. po3.bbr01.eq01.dal01.networklayer.com 0.0% 42 114.1 166.8 114.0 564.6 113.2
9. po5.dar01.sr01.dal01.networklayer.com 0.0% 42 115.6 124.6 115.6 296.7 31.4
10. po1.slr01.sr01.dal01.networklayer.com 0.0% 42 116.4 116.6 115.6 129.7 2.6
11. SoftLayer® Technologies – Infrastructure as a Service including Dedicated, Virtualized, and Cloud Hosting 0.0% 42 116.3 116.8 116.0 126.2 2.167.228.118.21:
earth (0.0.0.0) Sun Aug 22 00:18:21 2010
Keys: Help Display mode Restart statistics Order of fields quit
Packets Pings
Host Loss% Snt Last Avg Best Wrst StDev
1. ar3.blsq.xcalibre.net.uk 0.0% 66 0.2 0.7 0.2 14.1 2.2
2. tr1.blsq.xcalibre.net.uk 0.0% 66 0.6 0.8 0.4 10.5 1.2
3. 0-519-br2.the.flexiant.net 0.0% 66 3.8 4.1 3.2 15.6 1.9
4. gige-g4-14.core1.lon1.he.net 0.0% 66 3.9 5.9 3.6 16.0 3.6
5. 10gigabitethernet4-4.core1.nyc4.he.net 0.0% 65 71.9 74.4 71.5 87.8 3.8
6. 10gigabitethernet1-2.core1.nyc1.he.net 0.0% 65 71.8 74.7 71.6 85.2 4.1
7. te1-7.bbr01.tl01.nyc01.networklayer.com 0.0% 65 77.7 73.5 71.6 115.2 6.8
8. po4.bbr01.wb01.sea01.networklayer.com 0.0% 65 144.1 148.9 144.0 275.2 21.6
9. po3.bbr01.sr01.sea01.networklayer.com 0.0% 65 145.0 145.5 144.3 164.3 3.0
10. po1.cer02.sr01.sea01.networklayer.com 0.0% 65 144.5 148.1 144.3 259.9 15.9
11. po2.fcr01.sr01.sea01.networklayer.com 0.0% 65 156.3 148.8 144.3 182.7 8.5
12. Welcome to SoftLayer 0.0% 65 144.6 145.0 144.1 158.8 2.1208.43.118.51:
earth (0.0.0.0) Sun Aug 22 00:20:00 2010
Keys: Help Display mode Restart statistics Order of fields quit
Packets Pings
Host Loss% Snt Last Avg Best Wrst StDev
1. ar3.blsq.xcalibre.net.uk 0.0% 46 10.0 1.2 0.2 17.2 3.3
2. tr1.blsq.xcalibre.net.uk 0.0% 46 0.8 1.4 0.4 16.8 2.7
3. 0-519-br2.the.flexiant.net 0.0% 46 3.5 4.4 3.3 19.2 3.4
4. gige-g4-14.core1.lon1.he.net 0.0% 46 9.5 6.7 3.5 21.3 4.8
5. 10gigabitethernet4-4.core1.nyc4.he.net 0.0% 46 71.9 75.3 71.6 89.5 4.7
6. 10gigabitethernet1-2.core1.nyc1.he.net 0.0% 45 72.0 75.4 71.6 89.3 4.5
7. te1-7.bbr01.tl01.nyc01.networklayer.com 0.0% 45 72.3 81.4 71.8 343.3 43.2
8. po4.bbr01.eq01.wdc01.networklayer.com 0.0% 45 78.3 81.4 77.9 191.4 16.9
9. po1.bbr01.sr01.wdc01.networklayer.com 0.0% 45 78.7 79.2 78.3 87.2 1.8
10. po1.cer01.sr01.wdc01.networklayer.com 0.0% 45 78.7 81.7 78.1 203.2 18.6
11. po1.fcr01.sr01.wdc01.networklayer.com 0.0% 45 101.9 85.1 81.9 105.4 6.2
12. Welcome to SoftLayer 0.0% 45 99.2 83.4 82.0 101.4 3.8August 21, 2010 at 11:57 pm #1228491I’m looking at a move to a new datacenter not load balancing, interestingly preliminary results are better results from from their Seattle datacenter which is unexpected…
August 21, 2010 at 11:59 pm #1228492@Debris 395865 wrote:
It’s not network performance test but a ISP performance test.
Care to post your results?
August 22, 2010 at 12:00 am #1228520@Dr Bunsen 395888 wrote:
I’m looking at a move to a new datacenter not load balancing…
The results I got are pretty normal for Dallas, unless you’re moving to the UK there won’t be much difference for UK visitors.
August 22, 2010 at 12:01 am #1228493@madscientist 395885 wrote:
Here’s some MTR’s from a box in maidenhead. If you want any advise give me a shout.
Much appreciated, thanks…
August 22, 2010 at 12:02 am #1228494@madscientist 395890 wrote:
The results I got are pretty normal for Dallas, unless you’re moving to the UK there won’t be much difference for UK visitors.
So far average responce times are ~40% less from Seattle than Dallas…
August 22, 2010 at 12:06 am #1228521When you’re talking distances like this, especially for streaming etc then jitter rather than latency is the biggest problem, although it’s usually only protocols like SIP for VoIP etc that are affected.
Almost all transit from the UK to the US will go via NYC, so the closer you get to NYC the better, when you jump to Dallas expect at least +30-40ms on latency.
August 22, 2010 at 12:10 am #1228508when I use mtr 67.228.118.51 (tries) to route to Seattle here from BT!! something is hosed somewhere along the line..
11 t2c1-p4-0-0.us-ash.eu.bt.net (166.49.164.225) 112.940 ms 112.169 ms 112.237 ms
12 te3-5.bbr01.eq01.wdc01.networklayer.com (206.223.115.185) 105.027 ms 104.221 ms 105.405 ms
13 po3.bbr02.eq01.chi01.networklayer.com (173.192.18.155) 121.577 ms 120.696 ms 121.321 ms
MPLS Label=124501 CoS=6 TTL=1 S=0
14 po4.bbr02.cf01.den01.networklayer.com (173.192.18.131) 146.880 ms 147.000 ms 146.801 ms
MPLS Label=102290 CoS=6 TTL=1 S=0
15 po4.bbr01.wb01.sea01.networklayer.com (173.192.18.144) 172.690 ms 172.494 ms 173.317 ms -
AuthorPosts
- You must be logged in to reply to this topic.