Web Squad ISP

[DeatheCore]

I don't need to max it out, but what's the point of having gigabit and only being able to reach 300mbps on downloads? That's like buying a ferrari, and someone asking "Why do you need to go fast? Driving at the speed of a bus will still get you there?"

Anyhow, the issue isn't maxing it out, but that ANY NNTP download for me gets me blackholed by the DDoS protection software. @websquadza has explained the reason behind it and it's understandable, I'm just not enjoying being cut off for a few minutes every time NNTP is active, hence the request for a solution or a downgrade.

Yeah that's a strange one tbh.. don't think SABnzbd manipulates packet size outside of the usual tcp norms, and has the ability to utilize more simultaneous connections/tcp threads based on your config. Do you trigger the DDoS protection when downloading something with a download manager, like IDM, with say 8 threads?

In my testing with SAB rate limiting, there is barely any difference in pps between 50MB/s @ 50 threads or 50MB/s @ 25 threads. Firstly both were able to saturature the 50MB/s limit. The difference I did notice however was the inconsistency in pps - higher initial burst, more spikes (higher and lower), and higher tcp error count @ 50 threads.

Lastly, downloading a youtube video with IDM, speed unrestricted, 8 threads yielded similar pps results to the SAB tests. You can check the graphs here if you want:

https://imgur.com/a/KYBz9OL

Obviously there are many differences between our two setups, like usenet server, ISP, FNO, LAN, etc.. also, I'm no expert in this stuff, just curious

 

[Seeyou]

It is interesting. I proposed a theory to websquadza about the possibility of it being related to the quality of our two FNO networks and the difference in the amount of packet loss on the line between them. Since SAB is already generating upwards of 20k pps on your line, my theory was that higher packet loss on my line would generate a whole lot more retries/resends and skyrocket the overall pps to the point where it triggers the DDoS detection. (I can't reach 50MB/sec reliably even with 25 connections, on two separate usenet providers I've tested with, which I'm attributing to loss)

I did actually rate limit SAB to 30MB/sec today and managed several downloads without triggering the protection. I think somewhere between 30-40MB/sec is where it steps over the boundary in my setup / line conditions, so I'll try find the sweet spot.

 

[Hingisp]

Having a lot of dropouts today, online then offline then reconnects after a few mins.. Is there maintenance going on? Vumatel trenched, durban

 

[Seeyou]

Having a lot of dropouts today, online then offline then reconnects after a few mins.. Is there maintenance going on? Vumatel trenched, durban

Same thing, Octotel Southern Suburbs - really annoying.

 

[websquadza]

Having a lot of dropouts today, online then offline then reconnects after a few mins.. Is there maintenance going on? Vumatel trenched, durban

Hi. Yes, there's emergency maintenance going on in Durban. Network status page was updated a bit earlier. It's running slightly over from what I believe.

Same thing, Octotel Southern Suburbs - really annoying.

Nothing going on in Cape Town. What issue you having?

 

[Seeyou]

Hi. Yes, there's emergency maintenance going on in Durban. Network status page was updated a bit earlier. It's running slightly over from what I believe.



Nothing going on in Cape Town. What issue you having?

My internet will just die. Completely - no traffic. Then it will come back again in a few mins. If I disconnect and reconnect PPPoE it'll come back normally. I'm not downloading or anything that might trigger the blackhole.

 

[websquadza]

My internet will just die. Completely - no traffic. Then it will come back again in a few mins. If I disconnect and reconnect PPPoE it'll come back normally. I'm not downloading or anything that might trigger the blackhole.

Looking into this.

 

[Seeyou]

Happening like clockwork, every few mins now.

 

[websquadza]

Happening like clockwork, every few mins now.

Looks like a line issue- seeing packet loss to you.. Please log a ticket.

 

[websquadza]

Happening like clockwork, every few mins now.

Have you tried running a WinMTR from your end?

 

[Seeyou]

Have you tried running a WinMTR from your end?

I've tried - everything is completely dead when it happens, so MTR doesn't even work.

 

[Seeyou]

Looks like a line issue- seeing packet loss to you.. Please log a ticket.

Always had loss on the line - Octotel are quite happy about the state of it. Last time I ticketed it through AH, Octotel showed up, said "Signal strength is acceptable" and left. Unless you're seeing new, ridiculous levels of packet loss that I should ticket about, I can't see it making much difference with them.

 

[websquadza]

Always had loss on the line - Octotel are quite happy about the state of it. Last time I ticketed it through AH, Octotel showed up, said "Signal strength is acceptable" and left. Unless you're seeing new, ridiculous levels of packet loss that I should ticket about, I can't see it making much difference with them.

Run an MTR from your side to Google or 8.8.8.8 - let's see if your results look similar to ours. If they do, then yes, there's an issue and we'll get on Octotel's case.

 

[websquadza]

I've tried - everything is completely dead when it happens, so MTR doesn't even work.

MTR will tell us which hop it dies at. If it's first hop, it's local, if it makes it to second hop with loss, it's Octotel

 

[Seeyou]

MTR will tell us which hop it dies at. If it's first hop, it's local, if it makes it to second hop with loss, it's Octotel

Appears to have stabilised, touch wood. If it happens again I'll MTR directly to 1.1.1.1 - I see I was running it to an address instead of an IP, so obviously it couldn't resolve the DNS.

Current MTR is clear:

|------------------------------------------------------------------------------------------|
|                                      WinMTR statistics                                   |
|                       Host              -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
|                             192.168.1.1 -    0 |   79 |   79 |    0 |    0 |    1 |    0 |
|             core.as-01.cp1.za.ws.net.za -    0 |   79 |   79 |    0 |    0 |    0 |    0 |
|             core.cr-01.cp1.za.ws.net.za -    0 |   79 |   79 |    0 |    0 |    1 |    0 |
|           165-69-148-197.as37497.za.net -    0 |   79 |   79 |    0 |    0 |    1 |    0 |
|           162-68-148-197.as37497.za.net -    0 |   79 |   79 |  141 |  142 |  183 |  141 |
|            54-71-148-197.as37497.za.net -    0 |   79 |   79 |  141 |  144 |  215 |  142 |
|           121-66-148-197.as37497.za.net -    0 |   79 |   79 |  141 |  141 |  145 |  141 |
|ge-0-1-0-68.peering1.the.lon1.coreix.net -    0 |   79 |   79 |  141 |  141 |  147 |  142 |
|        xe-5-1.edge4.enf.lon5.coreix.net -    0 |   79 |   79 |  141 |  142 |  156 |  141 |
|        85.13.221.178.reverse.coreix.net -    0 |   79 |   79 |  141 |  141 |  148 |  145 |
|________________________________________________|______|______|______|______|______|______|
   WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

 

[Hingisp]

Hi. Yes, there's emergency maintenance going on in Durban. Network status page was updated a bit earlier. It's running slightly over from what I believe.



Nothing going on in Cape Town. What issue you having?

Thanks

 

[JOEY_8]

Internet seems to really unstable for me at the moment in Durban.

 

[Hingisp]

Internet seems to really unstable for me at the moment in Durban.

Yea me to bud.

 

[websquadza]

Internet seems to really unstable for me at the moment in Durban.

Yea me to bud.

We're checking this out.. Anything in particular giving issues? Loss?

 

[Hingisp]

We're checking this out.. Anything in particular giving issues? Loss?

Here's an MTR for you.