Cool Ideas Fibre ISP - Feedback

Status
Not open for further replies.
jannier said:
Vodafone London


VOX London


Cybersmart London


Speed seems OK for international, you will seldom get full UP/DN speeds on international due to Latency and single thread, that is where download manger works so well, open about 10 threads and you should saturate your DN link most of the time on international, but speed depends heavily on latency the server you connecting too's performance and the routing international.
Click to expand...

Just to show you, tests to the exact same server

Last night during that peak time http://www.speedtest.net/result/i/2731962877
This morning http://www.speedtest.net/result/7461304796

As you can see there is a hughe difference in download speeds, I tested other servers in London as well and got the same results, as just after 10pm I did another test and downloads speeds were fine again and buffering stopped so pretty sure there is something up with the backhaul provider then.

At the rate Vuma is adding new customers in our areas it's very likely to happen if they do not keep their backhauls up to date to handle all the new traffic. AFAIK all these new areas, ie Brackenfell, Bellville, Kuilsrivier are routing via the Kendridge POP before heading off to the data center.
 
Kloon said:
Just to show you, tests to the exact same server

Last night during that peak time http://www.speedtest.net/result/i/2731962877
This morning http://www.speedtest.net/result/7461304796

As you can see there is a hughe difference in download speeds, I tested other servers in London as well and got the same results, as just after 10pm I did another test and downloads speeds were fine again and buffering stopped so pretty sure there is something up with the backhaul provider then.

At the rate Vuma is adding new customers in our areas it's very likely to happen if they do not keep their backhauls up to date to handle all the new traffic. AFAIK all these new areas, ie Brackenfell, Bellville, Kuilsrivier are routing via the Kendridge POP before heading off to the data center.
Click to expand...



I am getting higher ping and terrible dl and up speeds too.

http://www.speedtest.net/result/7461399628

[MENTION=394133]PBCool[/MENTION] is there a way to improve this?

Thanks
 
I will check the CloudFlare routing as well.
 
Nicholas Marzio said:
Hi PBCool

Would there be a reason why CISP CPT Vumatel clients in Tamboerskloof would be routing via JHB where as CISP CPT Octotel clients in Green point would be routing via cpt?

I.E.
My friend on Octotel CISP (Green Point Node).
Code: 
Tracing route to 1dot1dot1dot1.cloudflare-dns.com [1.1.1.1]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  10.11.12.1
  2     1 ms     1 ms     1 ms  155.93.250.1
  3     2 ms     1 ms     1 ms  cloudflare.ixp.capetown [196.10.140.198]
  4     1 ms     1 ms     1 ms  1dot1dot1dot1.cloudflare-dns.com [1.1.1.1]

Vs
Me Vumatel AE CISP (Tamboerskloof Node)

Code: 
[admin@MARZIO-CORE] > tool traceroute 1.1.1.1
 # ADDRESS                          LOSS SENT    LAST     AVG    BEST   WORST STD-DEV STATUS                                                          
 1 155.93.243.1                       0%  234   0.4ms     2.4     0.3   196.2      14                                                                 
 2 154.0.1.125                        0%  234   0.3ms     0.3     0.2     0.9       0                                                                 
 3 154.0.1.13                         0%  234   0.5ms     0.7     0.4     9.2     0.9                                                                 
 4 196.10.140.198                     0%  234  19.5ms    18.2    17.2    23.6     1.2                                                                 
 5 1.1.1.1                            0%  234  19.4ms    17.9    17.2    21.3       1                                                                 
-- [Q quit|D dump|C-z pause]

Also what is the JHB CISP speedtest capacity, friend whats to try find a local 10G speedtest server?
Click to expand...
Where is your friend located? In JHB?
 
johannkoegelenberg said:
Quote Originally Posted by johannkoegelenberg View Post
[MENTION=394133]PBCool[/MENTION]

COOL-20180514-102837

I am struggling and not getting any feedback since May.

Speedtest now from Cool Ideas 155.93.246.207 - CI Randburg - 70mbps down / 20.62 up.

I'm paying for 100/100.

CI - CI Durban 49.75 / 20.61
CI - Saix Cpt 11.95 / 3.67
CI - Vodacom 55 / 16.09
CI - RSAWeb - 75.05 - 33.62

Does not matter if this is done directly from CPE or with a couple of machines on my network.

I own another property a couple of houses down the street also on a 100/100mbps line but from Rsaweb, there I am getting very close to 100/100 to pretty much any speedtest server.

Please advise as no-one comes back to me.
Click to expand...
Just to confirm our technician came out today and the tests were normal?
 
[MENTION=394133]PBCool[/MENTION] as a matter of interest will you be updating the geoip records for your new CPT IP range on MTN be recognised as being in CT? All geoip based sites think im in JHB now.
 
Kloon said:
Just to show you, tests to the exact same server

Last night during that peak time http://www.speedtest.net/result/i/2731962877
This morning http://www.speedtest.net/result/7461304796

As you can see there is a hughe difference in download speeds, I tested other servers in London as well and got the same results, as just after 10pm I did another test and downloads speeds were fine again and buffering stopped so pretty sure there is something up with the backhaul provider then.

At the rate Vuma is adding new customers in our areas it's very likely to happen if they do not keep their backhauls up to date to handle all the new traffic. AFAIK all these new areas, ie Brackenfell, Bellville, Kuilsrivier are routing via the Kendridge POP before heading off to the data center.
Click to expand...

UP/DN Speeds seems all over the place atm. when I run a test to the same server, latency jumps between 140 - 170ms.







 
Last edited:
PBCool said:
Just to confirm our technician came out today and the tests were normal?
Click to expand...

[MENTION=394133]PBCool[/MENTION]

Thanks, it is sorted now. Tests were not normal. He shifted me onto another server. Getting normal speeds now after the change.

Great that you sent out someone so quickly. That is good service from you and much appreciated, but why could your normal helpdesk not make the call and do something about the problem. Had to constantly call and follow up without any resolution or further feedback.

I'm happy to be sorted out, but I think there is much room for improvement at your service desk.

Just a bit of constructive criticism.

Regards
Johann
 
jannier said:
UP/DN Speeds seems all over the place atm. when I run a test to the same server, latency jumps between 140 - 170ms.





Click to expand...
Try run an MTR to an IP in London instead when focussing on latency, on the higher latencies the results will natively vary with TCP when it comes to throughput.
 
blunt said:
[MENTION=394133]PBCool[/MENTION] as a matter of interest will you be updating the geoip records for your new CPT IP range on MTN be recognised as being in CT? All geoip based sites think im in JHB now.
Click to expand...
We have requested the update, this happens over a short period of time.
 
johannkoegelenberg said:
[MENTION=394133]PBCool[/MENTION]

Thanks, it is sorted now. Tests were not normal. He shifted me onto another server. Getting normal speeds now after the change.

Great that you sent out someone so quickly. That is good service from you and much appreciated, but why could your normal helpdesk not make the call and do something about the problem. Had to constantly call and follow up without any resolution or further feedback.

I'm happy to be sorted out, but I think there is much room for improvement at your service desk.

Just a bit of constructive criticism.

Regards
Johann
Click to expand...
I am confused as to which server he moved you over to?
There were no changes made to the service?
 
PBCool said:
Try run an MTR to an IP in London instead when focussing on latency, on the higher latencies the results will natively vary with TCP when it comes to throughput.
Click to expand...


Tracing route to 50.7.152.4 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms 192.168.88.1
2 1 ms 1 ms 1 ms 155.93.243.1
3 1 ms <1 ms <1 ms 154.0.1.125
4 1 ms 1 ms 1 ms 154.0.1.13
5 141 ms 141 ms 141 ms 154.0.4.158
6 141 ms 143 ms 142 ms v1000.core1.lon3.he.net [216.66.80.169]
7 141 ms 149 ms 149 ms 10ge3-1.core1.lon1.he.net [184.105.64.233]
8 141 ms 143 ms 141 ms ldn-b5-link.telia.net [213.248.93.81]
9 141 ms 141 ms 141 ms ldn-bb3-link.telia.net [213.155.132.194]
10 142 ms 142 ms 142 ms ldn-b1-link.telia.net [62.115.114.235]
11 166 ms 166 ms 165 ms cogent-ic-326660-ldn-b1.c.telia.net [62.115.9.29]
12 142 ms 141 ms 141 ms be2871.ccr42.lon13.atlas.cogentco.com [154.54.58.185]
13 148 ms 147 ms 148 ms core01.aggr1.lon.fdcservers.net [149.11.45.50]
14 141 ms 141 ms 141 ms 50.7.152.4

Trace complete.

|------------------------------------------------------------------------------------------|

| WinMTR statistics |

| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |

|------------------------------------------------|------|------|------|------|------|------|

| 192.168.88.1 - 0 | 35 | 35 | 0 | 0 | 0 | 0 |

| 155.93.243.1 - 0 | 35 | 35 | 1 | 2 | 27 | 1 |

| 154.0.1.125 - 0 | 35 | 35 | 1 | 1 | 1 | 1 |

| 154.0.1.13 - 0 | 35 | 35 | 1 | 1 | 4 | 1 |

| uoa-cust.coolideas.co.za - 0 | 35 | 35 | 141 | 141 | 145 | 141 |

| v1000.core1.lon3.he.net - 0 | 35 | 35 | 141 | 141 | 144 | 141 |

| 10ge3-1.core1.lon1.he.net - 0 | 35 | 35 | 141 | 145 | 153 | 144 |

| ldn-b5-link.telia.net - 0 | 35 | 35 | 141 | 141 | 159 | 141 |

| ldn-bb3-link.telia.net - 0 | 35 | 35 | 141 | 143 | 154 | 142 |

| ldn-b1-link.telia.net - 0 | 35 | 35 | 141 | 142 | 148 | 142 |

| cogent-ic-326660-ldn-b1.c.telia.net - 0 | 35 | 35 | 165 | 165 | 167 | 165 |

| be2871.ccr42.lon13.atlas.cogentco.com - 0 | 35 | 35 | 141 | 141 | 143 | 142 |

| core01.aggr1.lon.fdcservers.net - 0 | 35 | 35 | 147 | 147 | 150 | 148 |

| 50.7.152.4 - 0 | 35 | 35 | 141 | 141 | 143 | 141 |

|________________________________________________|______|______|______|______|______|______|

WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider


Reverse MTR and Traceroute https://www.fdcservers.net/lg/index.php

London lg-lon.fdcservers.net


traceroute to 155.93.xxx.xxx (155.93.xxx.xxx), 30 hops max, 60 byte packets
1 core-jn1.lon01.ent.fdcservers.net (50.7.152.1) 1.223 ms 1.199 ms 1.205 ms
2 be4660.ccr42.lon13.atlas.cogentco.com (149.11.45.49) 0.686 ms be4087.ccr41.lon13.atlas.cogentco.com (149.11.160.77) 0.710 ms 0.679 ms
3 be2791.ccr21.lon02.atlas.cogentco.com (154.54.56.78) 1.051 ms be2792.ccr21.lon02.atlas.cogentco.com (154.54.57.106) 0.881 ms be2791.ccr21.lon02.atlas.cogentco.com (154.54.56.78) 1.033 ms
4 149.6.148.134 (149.6.148.134) 9.032 ms 9.025 ms 149.6.148.130 (149.6.148.130) 2.510 ms
5 mi-za-cpt-p7-ten0-0-1-1sub2105.ip.isnet.net (168.209.100.141) 163.590 ms * 168.209.201.22 (168.209.201.22) 140.732 ms
6 168.209.2.133 (168.209.2.133) 142.565 ms 163.918 ms 168.209.6.133 (168.209.6.133) 163.958 ms
7 196.34.14.197 (196.34.14.197) 163.506 ms 140.506 ms 163.288 ms
8 wc-trce-cr1-gi0-0-4-3206.enetworks.net (154.72.97.222) 140.208 ms 140.266 ms 140.252 ms
9 wc-grwf-cr1-gi0-0-3224.enetworks.net (45.220.130.30) 143.740 ms 143.708 ms 143.697 ms
10 wc-grwf-ars1-v3202.enetworks.net (154.72.97.169) 163.530 ms 163.657 ms 163.712 ms
11 uwt-cust.coolideas.co.za (154.0.4.157) 163.702 ms 140.773 ms 163.717 ms
12 ce-backbone.coolideas.co.za (154.0.1.14) 143.306 ms 143.303 ms 143.319 ms
13 c3i-backbone.coolideas.co.za (154.0.1.126) 141.435 ms 141.379 ms 163.988 ms
14 * * *
15 * * *

HOST: lg-lon.fdcservers.net Loss% Snt Last Avg Best Wrst StDev
1.|-- core-jn1.lon01.ent.fdcser 0.0% 5 0.8 0.8 0.7 0.9 0.0
2.|-- be4660.ccr42.lon13.atlas. 0.0% 5 0.7 0.7 0.6 0.8 0.0
3.|-- be2792.ccr21.lon02.atlas. 0.0% 5 0.9 0.9 0.8 1.0 0.0
4.|-- 149.6.148.134 0.0% 5 1.8 2.4 1.8 2.7 0.0
5.|-- 168.209.201.22 0.0% 5 140.8 140.7 140.6 140.9 0.0
6.|-- 168.209.2.133 0.0% 5 141.5 151.3 140.5 193.4 23.5
7.|-- 196.34.14.197 0.0% 5 140.4 140.4 140.4 140.5 0.0
8.|-- wc-trce-cr1-gi0-0-4-3206. 0.0% 5 140.2 140.2 140.2 140.3 0.0
9.|-- wc-grwf-cr1-gi0-0-3224.en 0.0% 5 143.7 143.8 143.7 143.9 0.0
10.|-- wc-grwf-ars1-v3202.enetwo 0.0% 5 140.4 140.6 140.4 141.1 0.0
11.|-- uwt-cust.coolideas.co.za 0.0% 5 140.7 140.9 140.6 141.6 0.0
12.|-- ce-backbone.coolideas.co. 0.0% 5 140.9 140.9 140.9 140.9 0.0
13.|-- c3i-backbone.coolideas.co 0.0% 5 141.3 141.1 140.9 141.3 0.0
14.|-- 155.93.xxx.xxx 0.0% 5 141.8 141.8 141.7 142.2 0.0
 
jannier said:
Tracing route to 50.7.152.4 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms 192.168.88.1
2 1 ms 1 ms 1 ms 155.93.243.1
3 1 ms <1 ms <1 ms 154.0.1.125
4 1 ms 1 ms 1 ms 154.0.1.13
5 141 ms 141 ms 141 ms 154.0.4.158
6 141 ms 143 ms 142 ms v1000.core1.lon3.he.net [216.66.80.169]
7 141 ms 149 ms 149 ms 10ge3-1.core1.lon1.he.net [184.105.64.233]
8 141 ms 143 ms 141 ms ldn-b5-link.telia.net [213.248.93.81]
9 141 ms 141 ms 141 ms ldn-bb3-link.telia.net [213.155.132.194]
10 142 ms 142 ms 142 ms ldn-b1-link.telia.net [62.115.114.235]
11 166 ms 166 ms 165 ms cogent-ic-326660-ldn-b1.c.telia.net [62.115.9.29]
12 142 ms 141 ms 141 ms be2871.ccr42.lon13.atlas.cogentco.com [154.54.58.185]
13 148 ms 147 ms 148 ms core01.aggr1.lon.fdcservers.net [149.11.45.50]
14 141 ms 141 ms 141 ms 50.7.152.4

Trace complete.

|------------------------------------------------------------------------------------------|

| WinMTR statistics |

| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |

|------------------------------------------------|------|------|------|------|------|------|

| 192.168.88.1 - 0 | 35 | 35 | 0 | 0 | 0 | 0 |

| 155.93.243.1 - 0 | 35 | 35 | 1 | 2 | 27 | 1 |

| 154.0.1.125 - 0 | 35 | 35 | 1 | 1 | 1 | 1 |

| 154.0.1.13 - 0 | 35 | 35 | 1 | 1 | 4 | 1 |

| uoa-cust.coolideas.co.za - 0 | 35 | 35 | 141 | 141 | 145 | 141 |

| v1000.core1.lon3.he.net - 0 | 35 | 35 | 141 | 141 | 144 | 141 |

| 10ge3-1.core1.lon1.he.net - 0 | 35 | 35 | 141 | 145 | 153 | 144 |

| ldn-b5-link.telia.net - 0 | 35 | 35 | 141 | 141 | 159 | 141 |

| ldn-bb3-link.telia.net - 0 | 35 | 35 | 141 | 143 | 154 | 142 |

| ldn-b1-link.telia.net - 0 | 35 | 35 | 141 | 142 | 148 | 142 |

| cogent-ic-326660-ldn-b1.c.telia.net - 0 | 35 | 35 | 165 | 165 | 167 | 165 |

| be2871.ccr42.lon13.atlas.cogentco.com - 0 | 35 | 35 | 141 | 141 | 143 | 142 |

| core01.aggr1.lon.fdcservers.net - 0 | 35 | 35 | 147 | 147 | 150 | 148 |

| 50.7.152.4 - 0 | 35 | 35 | 141 | 141 | 143 | 141 |

|________________________________________________|______|______|______|______|______|______|

WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider


Reverse MTR and Traceroute https://www.fdcservers.net/lg/index.php

London lg-lon.fdcservers.net


traceroute to 155.93.xxx.xxx (155.93.xxx.xxx), 30 hops max, 60 byte packets
1 core-jn1.lon01.ent.fdcservers.net (50.7.152.1) 1.223 ms 1.199 ms 1.205 ms
2 be4660.ccr42.lon13.atlas.cogentco.com (149.11.45.49) 0.686 ms be4087.ccr41.lon13.atlas.cogentco.com (149.11.160.77) 0.710 ms 0.679 ms
3 be2791.ccr21.lon02.atlas.cogentco.com (154.54.56.78) 1.051 ms be2792.ccr21.lon02.atlas.cogentco.com (154.54.57.106) 0.881 ms be2791.ccr21.lon02.atlas.cogentco.com (154.54.56.78) 1.033 ms
4 149.6.148.134 (149.6.148.134) 9.032 ms 9.025 ms 149.6.148.130 (149.6.148.130) 2.510 ms
5 mi-za-cpt-p7-ten0-0-1-1sub2105.ip.isnet.net (168.209.100.141) 163.590 ms * 168.209.201.22 (168.209.201.22) 140.732 ms
6 168.209.2.133 (168.209.2.133) 142.565 ms 163.918 ms 168.209.6.133 (168.209.6.133) 163.958 ms
7 196.34.14.197 (196.34.14.197) 163.506 ms 140.506 ms 163.288 ms
8 wc-trce-cr1-gi0-0-4-3206.enetworks.net (154.72.97.222) 140.208 ms 140.266 ms 140.252 ms
9 wc-grwf-cr1-gi0-0-3224.enetworks.net (45.220.130.30) 143.740 ms 143.708 ms 143.697 ms
10 wc-grwf-ars1-v3202.enetworks.net (154.72.97.169) 163.530 ms 163.657 ms 163.712 ms
11 uwt-cust.coolideas.co.za (154.0.4.157) 163.702 ms 140.773 ms 163.717 ms
12 ce-backbone.coolideas.co.za (154.0.1.14) 143.306 ms 143.303 ms 143.319 ms
13 c3i-backbone.coolideas.co.za (154.0.1.126) 141.435 ms 141.379 ms 163.988 ms
14 * * *
15 * * *

HOST: lg-lon.fdcservers.net Loss% Snt Last Avg Best Wrst StDev
1.|-- core-jn1.lon01.ent.fdcser 0.0% 5 0.8 0.8 0.7 0.9 0.0
2.|-- be4660.ccr42.lon13.atlas. 0.0% 5 0.7 0.7 0.6 0.8 0.0
3.|-- be2792.ccr21.lon02.atlas. 0.0% 5 0.9 0.9 0.8 1.0 0.0
4.|-- 149.6.148.134 0.0% 5 1.8 2.4 1.8 2.7 0.0
5.|-- 168.209.201.22 0.0% 5 140.8 140.7 140.6 140.9 0.0
6.|-- 168.209.2.133 0.0% 5 141.5 151.3 140.5 193.4 23.5
7.|-- 196.34.14.197 0.0% 5 140.4 140.4 140.4 140.5 0.0
8.|-- wc-trce-cr1-gi0-0-4-3206. 0.0% 5 140.2 140.2 140.2 140.3 0.0
9.|-- wc-grwf-cr1-gi0-0-3224.en 0.0% 5 143.7 143.8 143.7 143.9 0.0
10.|-- wc-grwf-ars1-v3202.enetwo 0.0% 5 140.4 140.6 140.4 141.1 0.0
11.|-- uwt-cust.coolideas.co.za 0.0% 5 140.7 140.9 140.6 141.6 0.0
12.|-- ce-backbone.coolideas.co. 0.0% 5 140.9 140.9 140.9 140.9 0.0
13.|-- c3i-backbone.coolideas.co 0.0% 5 141.3 141.1 140.9 141.3 0.0
14.|-- 155.93.xxx.xxx 0.0% 5 141.8 141.8 141.7 142.2 0.0
Click to expand...
So MTR looks very consistent?
 
PBCool said:
So MTR looks very consistent?
Click to expand...

Yes, just the odd servers on the route that has higher latency then others.

11.|-- cogent-ic-326660-ldn-b1.c.telia.net - 0 | 35 | 35 | 165 | 165 | 167 | 165 |

6.|-- 168.209.2.133 0.0% 5 141.5 151.3 140.5 193.4 23.5

When doing a testzip file DL from https://www.fdcservers.net/lg/index.php a single thread will give 2.0 - 3.0MB/s versus using 10 Threads with a DL Manager will saturate my line to 23.0 - 25.0MB/s

So it is not so much a ISP speed issue than it is more the hardware/link serving the file that throttles the speed per thread it seems.
 
Status
Not open for further replies.
Back
Top
Sign up to the MyBroadband newsletter
X