Cool Ideas Fibre ISP – Feedback Thread 2

DJZassie

Well-Known Member
Joined
Dec 10, 2011
Messages
234
So, we dropped Hurricane Electric in lieu of Cogent in the UK.

Seems like HE had some sort of attack/issue tonight. If it doesn't rain....

I trace clean to 5.79.80.227 now.
seems MUUUUCH better, and looks alot like the route i normally get! Thanks a bunch guys!


Code:
|------------------------------------------------------------------------------------------|
|                                      WinMTR statistics                                   |
|                       Host              -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
|                            172.18.7.225 -    0 |  417 |  417 |    0 |    0 |    0 |    0 |
|                            155.93.231.1 -    0 |  417 |  417 |    2 |    3 |   27 |    2 |
|                           155.93.240.34 -    0 |  417 |  417 |    2 |    3 |   32 |    2 |
|                           155.93.240.25 -    0 |  414 |  414 |    0 |    1 |  443 |    0 |
|                uwu-cust.coolideas.co.za -    0 |  291 |  291 |  144 |  144 |  146 |  144 |
|                              154.0.5.18 -    0 |  291 |  291 |  144 |  144 |  150 |  144 |
|            peering.thn.lon.leaseweb.net -    0 |  289 |  289 |  144 |  145 |  315 |  145 |
|       et-27-1.agg01.ams-01.leaseweb.net -    0 |  284 |  284 |  145 |  147 |  262 |  149 |
|      be-101.br02.ams-01.nl.leaseweb.net -    0 |  279 |  279 |  150 |  150 |  157 |  150 |
|      te-8-1.ce38.ams-01.nl.leaseweb.net -    0 |  277 |  277 |  150 |  152 |  591 |  150 |
|                             5.79.80.227 -    0 |  278 |  278 |  145 |  151 |  574 |  150 |
|________________________________________________|______|______|______|______|______|______|
   WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider
 

TheRoDent

Expert Member
Joined
Aug 6, 2003
Messages
3,663
Steam is dead as a robert mugabe.

Tracing route to steam.com [69.36.241.237]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms router.lan [192.168.88.1]
2 3 ms 3 ms 2 ms 154.0.0.246
3 2 ms 2 ms 2 ms 154.0.4.161
4 146 ms 146 ms 146 ms 154.0.4.2
5 147 ms 146 ms 146 ms v1000.core1.lon3.he.net [216.66.80.169]
6 163 ms 146 ms 148 ms 100ge14-1.core1.lon2.he.net [184.105.64.237]
7 214 ms 213 ms 213 ms 100ge13-2.core1.nyc4.he.net [72.52.92.166]
8 213 ms 213 ms 213 ms 100ge16-1.core1.ash1.he.net [184.105.223.165]
9 220 ms 218 ms 218 ms core1.ash.layer42.net [206.126.236.8]
10 281 ms 281 ms 281 ms xe1-4.core2.sv1.layer42.net [65.50.198.81]
11 281 ms 281 ms 282 ms be-5.cr1-sanant-a.bb.as11404.net [65.50.198.42]
12 281 ms 281 ms 281 ms be-6.cr2-losangeles-b.bb.as11404.net [65.50.198.45]
13 280 ms 280 ms 275 ms cr1-scl-be-4.bb.as11404.net [192.175.29.172]
14 282 ms 282 ms 282 ms cr1-529bryant-hu-0-3-0-20-0.bb.as11404.net [192.175.28.139]
15 282 ms 283 ms 282 ms core1-te-0-2-0-4-2.sv8.layer42.net [208.76.184.59]
16 278 ms 278 ms 279 ms te-0-0-0-7.core2.sv8.bb.as11404.net [65.50.198.190]
17 281 ms 281 ms 281 ms 69.36.239.148
18 a66-109-98-50.mtv.toaster.net [66.109.98.50] reports: Destination host unreachable.

Trace complete.
It's steampowered.com ;)
 

DJZassie

Well-Known Member
Joined
Dec 10, 2011
Messages
234
So, we dropped Hurricane Electric in lieu of Cogent in the UK.

Seems like HE had some sort of attack/issue tonight. If it doesn't rain....

I trace clean to 5.79.80.227 now.
Cogent died again xD


Code:
|------------------------------------------------------------------------------------------|
|                                      WinMTR statistics                                   |
|                       Host              -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
|                            172.18.7.225 -    0 |  612 |  612 |    0 |    0 |    0 |    0 |
|                            155.93.231.1 -    0 |  612 |  612 |    2 |    3 |   36 |    2 |
|                           155.93.240.34 -    0 |  611 |  611 |    2 |    3 |   50 |    2 |
|                           155.93.240.25 -    0 |  613 |  613 |    0 |    0 |    1 |    0 |
|                uwu-cust.coolideas.co.za -    0 |  430 |  430 |  140 |  142 |  145 |  140 |
|te0-1-0-7-4.1200.ccr21.lon01.atlas.cogentco.com -    0 |  431 |  431 |  140 |  143 |  148 |  140 |
|   be2871.ccr42.lon13.atlas.cogentco.com -    0 |  432 |  432 |  140 |  142 |  148 |  145 |
|  be12488.ccr42.ams03.atlas.cogentco.com -    1 |  379 |  378 |  148 |  149 |  154 |  148 |
|   be3434.rcr21.ams06.atlas.cogentco.com -    1 |  374 |  373 |  152 |  152 |  155 |  153 |
|                             149.6.0.202 -   60 |   20 |    8 |  179 |  196 |  208 |  197 |
|      te-8-3.ce38.ams-01.nl.leaseweb.net -   60 |   20 |    8 |    0 |  191 |  230 |  166 |
|                             5.79.80.227 -   47 |   26 |   14 |  149 |  196 |  239 |  190 |
|________________________________________________|______|______|______|______|______|______|
   WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider
 

John Tempus

Expert Member
Joined
Aug 8, 2017
Messages
2,593
There are still some major issues with international.

Since this issue most international downloads or speedtests that does direct connections without local proxying is running at 10Mbit if I am lucky or not connecting at all.
 

MolzZ

New Member
Joined
Jun 8, 2019
Messages
6
Still slow international and constant dc's. Why is this **** more unstable than the early days of ADSL?
 

TheRoDent

Expert Member
Joined
Aug 6, 2003
Messages
3,663
The WACS cable and WIOCC seems to be having endless problems tonight.

We are switching between backhauling it ourselves to Cape Town, to it being up in Joburg.

Right now, it looks like WACS is dead.
 

MolzZ

New Member
Joined
Jun 8, 2019
Messages
6
What magic are the other ISP's using that stops them from being impacted as badly or as often as CISP?
 

John Tempus

Expert Member
Joined
Aug 8, 2017
Messages
2,593
So when will this be resolved ?

Getting 0.5Mbit to international sites, this is beyond fked up guys and the only report since this started is that it has been "resolved".

Can't even avoid this issue via VPN since it cant establish a stable VPN connection either.

No way.
 

jannier

Expert Member
Joined
Jul 31, 2005
Messages
1,165
International is horrendously slow, disconnects, timeouts, stalls, pages not loading, buffering.


 

John Tempus

Expert Member
Joined
Aug 8, 2017
Messages
2,593
You think they *never* have problems?

Our network has layers upon layers of redundancy. We just openly admit when there are issues.

Maybe we should stop doing that.

I doubt anyone could claim no network having any issues ever but CISP have had countless issue after issue either daily or every other day starting at least 3 months ago.

For the first year not a single issue like this ever happened and then suddenly its a daily occurrence ?

Something changed recently because you guys simply cannot go from no issues at all for over a year and then issues every other day, it makes no sense.

Unless the claim is that everyone were just lucky to not have issues for first year and we have to deal with these issues now as a norm or there is some other unknown explanation we are not getting for some reason.
 

TheRoDent

Expert Member
Joined
Aug 6, 2003
Messages
3,663
Unless the claim is that everyone were just lucky to not have issues for first year and we have to deal with these issues now as a norm or there is some other unknown explanation we are not getting for some reason.
I think we're pretty transparent about issues. You don't see Internet Solutions/MWeb or Afrihost engineers debugging traceroute's and reacting to them on this forum.

In the last three weeks, we've had outages to KZN, on dual paths. Due to (what we hear from our vendors)

1. Fires at Machadodorp
2. Vandalisation of cables near Springs.
3. Vandalisation at a bridge.

And that's just the KZN route.

We also have two routes to Cape Town, which has had similar issues.

And we buy our Joburg transit in Joburg from a vendor that's also got the same underlying issues.

We don't and, won't ever buy from Seacom because they are high latency routes. But if you buy from an ISP such as IS/Mweb/Afrihost, then you will have different issues. And higher latency.

We prefer the routes we have because we believe in providing the lowest latency service we can.

We are the only ISP using the SACS cable to Brazil, and from thereon to the US. Other's have discarded it as "too complicated".

99% of the problems we have though, is getting to the coast, and from there on, or dealing with vendor networks that have outages due to either "finger troubles" or other issues.

We have built a POP in the UK, with multiple upstream carriers. We haul that back to ZA with as many low latency/diverse undersea cables as we can.

We do remote peering with Brazil, and three different USA based exchanges. We peer with more networks than anyone else in South Africa ( feel free to lookup the stats at https://bgp.he.net/country/ZA), more than any other ISP. Small ISP's just fling you to a default gateway at a cheap price.

We have more diversity on any given IP route than most other ISP's in South Africa. Our peering shows that.

When it comes to trundling your IP packets from A to B, I think we do a pretty good job.
 
Last edited:

under_sky

Expert Member
Joined
Dec 26, 2013
Messages
1,145
I think we're pretty transparent about issues. You don't see Internet Solutions/MWeb or Afrihost engineers debugging traceroute's and reacting to them on this forum.

In the last three weeks, we've had outages to KZN, on dual paths. Due to (what we hear from our vendors)

1. Fires at Machadodorp
2. Vandalisation of cables near Springs.
3. Vandalisation at a bridge.

And that's just the KZN route.

We also have two routes to Cape Town, which has had similar issues.

And we buy our Joburg transit in Joburg from a vendor that's also got the same underlying issues.

We don't and, won't ever buy from Seacom because they are high latency routes. But if you buy from an ISP such as IS/Mweb/Afrihost, then you will have different issues. And higher latency.

We prefer the routes we have because we believe in providing the lowest latency service we can.

We are the only ISP using the SACS cable to Brazil, and from thereon to the US. Other's have discarded it as "too complicated".

99% of the problems we have though, is getting to the coast, and from there on, or dealing with vendor networks that have outages due to either "finger troubles" or other issues.

We have built a POP in the UK, with multiple upstream carriers. We haul that back to ZA with as many low latency/diverse undersea cables as we can.

We do remote peering with Brazil, and three different USA based exchanges. We peer with more networks than anyone else in South Africa ( feel free to lookup the stats at https://bgp.he.net/country/ZA), more than any other ISP. Small ISP's just fling you to a default gateway at a cheap price.

We have more diversity on any given IP route than most other ISP's in South Africa. Our peering shows that.

When it comes to trundling your IP packets from A to B, I think we do a pretty good job.
Can 100% agree with you but issues for most people are on the local FNO, which is to an extent out of your hands and you can only relay the issue to them.

I for one have had maybe 2 to 3 problems over the past 1-2 years of which it was just silly routing issue but you go further than any other ISP hence why i only ever recommend people to you when they can afford it
 

Yster21

Senior Member
Joined
Feb 3, 2009
Messages
763
I doubt anyone could claim no network having any issues ever but CISP have had countless issue after issue either daily or every other day starting at least 3 months ago.

For the first year not a single issue like this ever happened and then suddenly its a daily occurrence ?

Something changed recently because you guys simply cannot go from no issues at all for over a year and then issues every other day, it makes no sense.

Unless the claim is that everyone were just lucky to not have issues for first year and we have to deal with these issues now as a norm or there is some other unknown explanation we are not getting for some reason.
What magic are the other ISP's using that stops them from being impacted as badly or as often as CISP?
I have not had a single CISP issue for a VERY long time, and when I do, 90% of the time it's a Vumatel Aerial issue. I was actually just thinking this weekend how good it's been for the past couple of months.
 
Top