Bot + 3proxy issue

lucax3

Power Member
Joined
Apr 9, 2015
Messages
547
Reaction score
224
Hello everyone, I’m experiencing an 80% ban rate on Reddit with my setup.

I’ve tried troubleshooting. My hardware consists of 10 dongles with Vodafone SIMs and a hub to connect them to the PC.

I tested the bot with a direct hotspot connection, and everything works fine, so the problem is the proxies.

Ho effettuato i controlli della mia configurazione su questi siti:
https://fv.pro/ 5% fraud (brownser version)
https://ipbinding.online/ (passed)
https://browserleaks.com/ip (passed)
https://proxy.incolumitas.com/proxy_detect.html (passed)
https://pixelscan.net/ (passed)
https://whoer.net/de 100%
https://www.browserscan.net/it 97%

My configuration seems perfect, but something is still missing.
 
can you show full screenshot from browserleaks/ip ?
 
I dont think it's the proxies that get you flagged. The only thing you might want to fix is that flag of "Generic VPN or Tunnel", for that you will need to install something like a OSfooler-ng on your proxy host machine to change MTU and other values. Have no idea if it does have Windows support as it looks like your dongles are run on windows?

When you test via hotspot, do you run your automation as well or you test it manually? Because if you run chrome automation, it can and will leak only when run and attached to active browser context.

It might be as well that sim card IP ranges are already abused if those are pre-paid cards.
 
I dont think it's the proxies that get you flagged. The only thing you might want to fix is that flag of "Generic VPN or Tunnel", for that you will need to install something like a OSfooler-ng on your proxy host machine to change MTU and other values. Have no idea if it does have Windows support as it looks like your dongles are run on windows?

When you test via hotspot, do you run your automation as well or you test it manually? Because if you run chrome automation, it can and will leak only when run and attached to active browser context.

It might be as well that sim card IP ranges are already abused if those are pre-paid cards.
Ho usato Windows per il problema TCP/IP. Sì, ho testato il bot con un hotspot e ho ricevuto solo pochi ban. C'è un modo per controllare se gli IP sono contrassegnati?
The dongle allows me to change the SIM's MTU value, but how do I determine the correct one? And in your opinion, what other values should I change?
 
Is scamalytics a good tool to check the quality of the IPs?

Short answer: No

Long answer:

It can only give a generic overview of what scamalytics think of that IP or subnet range.

Same as with any other IP checker, they will give opinion taken from their own databases.

So the question that then follows is - would your target site or app uses the same database.

I would say that if you test your IP and it's all red, in IPQS, scamalytics - then we can assume that this specific IP has been recently incredibly abused.

And then follows next question, did abuse happen in the site you want to target?

Due to increase of use for VPNs, in browsers and even phones(iCloud+ comes with privacy tools enabled) more and more it will become about device and it's fingerprints, not the IP itself.

I could signup gmail account without triggering PVA from my hetzner's ipv6, so if your IP is clean in specific target you want to run it on, it will be more about how previously it's abused, not what type of IP it is or what any other site thinks of it.
 
Ho usato Windows per il problema TCP/IP. Sì, ho testato il bot con un hotspot e ho ricevuto solo pochi ban. C'è un modo per controllare se gli IP sono contrassegnati?
The dongle allows me to change the SIM's MTU value, but how do I determine the correct one? And in your opinion, what other values should I change?
Stupid question, can a website detect the presence of JavaScript code used to extract data (not to interact), and can it be a problem?
 
Short answer: No

Long answer:

It can only give a generic overview of what scamalytics think of that IP or subnet range.

Same as with any other IP checker, they will give opinion taken from their own databases.

So the question that then follows is - would your target site or app uses the same database.

I would say that if you test your IP and it's all red, in IPQS, scamalytics - then we can assume that this specific IP has been recently incredibly abused.

And then follows next question, did abuse happen in the site you want to target?

Due to increase of use for VPNs, in browsers and even phones(iCloud+ comes with privacy tools enabled) more and more it will become about device and it's fingerprints, not the IP itself.

I could signup gmail account without triggering PVA from my hetzner's ipv6, so if your IP is clean in specific target you want to run it on, it will be more about how previously it's abused, not what type of IP it is or what any other site thinks of it.
Wow! I love your perspective on this. Your knowledge and growth have only increased since I followed your last thread about Kira and your journey.
 
Stupid question, can a website detect the presence of JavaScript code used to extract data (not to interact), and can it be a problem?
It's not a stupid question.

In theory - I can't think of a way how it could be detected.

What do you use for automating your browser? Do you use stealth browser or something else?

Please give more information, right now it's not clear how exactly you do run your automation.

One concern for me here is that when using in combination of socks5, chrome doesn't support UDP/QUIC(Http/3).
 
Hello everyone, I’m experiencing an 80% ban rate on Reddit with my setup.

I’ve tried troubleshooting. My hardware consists of 10 dongles with Vodafone SIMs and a hub to connect them to the PC.

I tested the bot with a direct hotspot connection, and everything works fine, so the problem is the proxies.

Ho effettuato i controlli della mia configurazione su questi siti:
https://fv.pro/ 5% fraud (brownser version)
https://ipbinding.online/ (passed)
https://browserleaks.com/ip (passed)
https://proxy.incolumitas.com/proxy_detect.html (passed)
https://pixelscan.net/ (passed)
https://whoer.net/de 100%
https://www.browserscan.net/it 97%

My configuration seems perfect, but something is still missing.
Yes you got it right! There are some necessary things that are needed to be done
 
It's not a stupid question.

In theory - I can't think of a way how it could be detected.

What do you use for automating your browser? Do you use stealth browser or something else?

Please give more information, right now it's not clear how exactly you do run your automation.

One concern for me here is that when using in combination of socks5, chrome doesn't support UDP/QUIC(Http/3).
Thankyou for you help
I’m using BAS,
the UDP test should be this one with https://www.browserscan.net/it.
For HTTP3, I use https://cloudflare-quic.com/. I need to refresh the page, but it shows HTTP3.
 

Attachments

  • brownserescan.png
    brownserescan.png
    129.2 KB · Views: 13
  • http3.png
    http3.png
    28.6 KB · Views: 13
The strange thing is that the Reddit accounts get banned after 7-8 hours and not immediately

Then it can be quite a lot of things, I'm afraid I can't help much more here.
 
In your opinion, is the browser proxy configuration correct, so it could just be a bot automation issue?
 
Back
Top