setnicka/ulozto-downloader

Cloudflare version 2 Captcha challenge

Opened this issue · 20 comments

I am getting the following error message when trying to download anything:

[Link solve]    Network error get new TOR connection: Cloudflare scrapper error: Detected a Cloudflare version 2 Captcha challenge, This feature is not available in the opensource (free) version.. Try new TOR session.

I tried to replace ulozto.cz by zachowajto.pl, but the problem persists.

Using or not using --auto-captcha makes no difference.

Setting --parts to different numbers makes no difference.

My configuration:

$ulozto-downloader --version
3.5.1
$tor --version
Tor version 0.4.7.13.
Tor is running on Darwin with Libevent 2.1.12-stable, OpenSSL 1.1.1u, Zlib 1.2.11, Liblzma N/A, Libzstd N/A and Unknown N/A as libc.
Tor compiled with clang version 14.0.0
$uname -a
Darwin XXX-MacBook-Pro.local 22.4.0 Darwin Kernel Version 22.4.0: Mon Mar  6 20:59:28 PST 2023; root:xnu-8796.101.5~3/RELEASE_ARM64_T6000 arm64 arm Darwin
$python --version
Python 3.10.11

Full dump

Here's the full dump (I had to terminate the job, the error message was repeating indefinitely):

$ulozto-downloader --parts 10 'https://ulozto.cz/file/Dbnt1YRwZ9xi/tom-and-jerry-tee-for-two-avi#!ZGx1BGR2AmN5BJMzBGuzMJL4Lmp1MJy4BIbhHHxkBR44FzLjZj=='
Starting downloading for url 'https://ulozto.cz/file/Dbnt1YRwZ9xi/tom-and-jerry-tee-for-two-avi#!ZGx1BGR2AmN5BJMzBGuzMJL4Lmp1MJy4BIbhHHxkBR44FzLjZj=='
Getting info (filename, filesize, …)
Downloading into: './Tom and Jerry _tee_for_two.avi'
CAPTCHA protected download - CAPTCHA challenges will be displayed
[TOR]   TOR started
[Link solve]    TOR get new CAPTCHA (timeout 30)
[Link solve]    Network error get new TOR connection: Cloudflare scrapper error: Detected a Cloudflare version 2 Captcha challenge, This feature is not available in the opensource (free) version.. Try new TOR session.
[Link solve]    TOR get new CAPTCHA (timeout 30)
[Link solve]    Network error get new TOR connection: Cloudflare scrapper error: Detected a Cloudflare version 2 Captcha challenge, This feature is not available in the opensource (free) version.. Try new TOR session.
[Link solve]    TOR get new CAPTCHA (timeout 30)
[Link solve]    Network error get new TOR connection: Cloudflare scrapper error: Detected a Cloudflare version 2 Captcha challenge, This feature is not available in the opensource (free) version.. Try new TOR session.

Experiencing the same problem.
From what i've found is that cloudscraper is obsolete because of the new cloudflare protection.

I have exactly the same problem.

Configuration:

Linux 6.3.6-arch1-1 x86_64 GNU/Linux
ulozto-downloader 3.5.1
Python            3.11.3
Tor version       0.4.7.13

The same problem for me. Downloader was working ok few weeks ago.

  • Ubuntu 22.04
  • Python 3.9.7
  • Tor 0.4.2.7
  • ulozto-downloader 3.5.0

Try downgrading to version 3.3.1

Try downgrading to version 3.3.1

That doesn't work either. It just solves the captcha but don't actually download the file

Why is there cloudscraper? In my web browser, there is not any cloudflare on ulozto.cz. (If it is because of tor, why not to use some proxy instead? Or just change the browser "identification" things. EXAMPLE: I can download the same file in my normal window and in my anonymous window of the same web browser at the same time)

I think, that implemetning some changer of browser identification will solve problem. Cloudscraper is not needed and tor is also not needed. In end, it is still downloading in "not tor" mode (if I understand it correctly?). With tor, it just gets the link. And after the implementation of this "browser-changing-identification-thing/plugin" it will not need tor and cloudscraper => much easier to install and not so problematic.

But it is just an idea... Idk, if @setnicka wants to so this thing... It would be more work i guess.

Cloudscraper is showing just when identify suspicious activity. Try to download tor browser or just open uloz.to over VPN. Then you will see cloudflare. But older version of ulozto-downloader is working. So probably this version could be used for it.

But older version of ulozto-downloader is working.

Which older versions are working? I downgraded to 3.3.1 and it didn't work.

But older version of ulozto-downloader is working.

Which older versions are working? I downgraded to 3.3.1 and it didn't work.

For me and others person is working. It doesn't work 100% but it runs. You don't need to fill a captcha and you can download more files at the same time. Just speed is slower.

Interesting. For me it's just throwing bunch of captcha solves and it doesn't start to download. I was on this version earlier this year and it worked fine. Could you share command (with flags) you're using to download some link?

Cloudscraper is showing just when identify suspicious activity. Try to download tor browser or just open uloz.to over VPN. Then you will see cloudflare. But older version of ulozto-downloader is working. So probably this version could be used for it.

And why is it using tor (which leads to suspicious activity)? It can get link normally. And for e.g. 30 parts it will not be problem, will it? And alsonI think, there are other methods to change IP address...

But older version of ulozto-downloader is working.

Which older versions are working? I downgraded to 3.3.1 and it didn't work.

For me and others person is working. It doesn't work 100% but it runs. You don't need to fill a captcha and you can download more files at the same time. Just speed is slower.

Does it download it with more parts? Or only 1? And how do I install it on Debian? (I just know how to install newest version, sorry)

version 3.3.1 sometimes downloading a more parts at the same time. But max speed which i had is 2MB/s.
If you will try to download something from uloz.to, then you are limited to speed and number of files. Tor changing you ip and you can bypass it with it. Another one way is VPN, but its the same. For downgrade just run command "pip install ulozto-downloader==3.3.1". Sometimes it has problem with captcha before start of downloading. Then try to kill it with ctrl+z and start again. or try to dele tor temp file from folder. And if speed is slow, then try to kill it and start once again. ulozto-downloader --parts %number_of_parts% --parts-progress "%url%".

Thank you. I will try it 👍

i have last version, it is able to solve max 1 or 2 captcha file. I am in Germany, so I have to use VPN. it is useless. I am not programer, but when i start normal slow download on ulozto, then check downloading file {RIGHT CLICK TO COPY DOWNLOADING FILE} it looks like this for example https://content-cache34.greencdn.link/Ps;Hs;up=0;cid=316549158;uip=169.150.218.27;aff=uloz.to;did=uloz-to;fide=rsDYF5d;fs=aJCDhGn36ivi;hid=mBcPEaz;rid=1964762282;tm=1688646986;ut=f;rs=0;fet=download_free;He;ch=d5aaf257ea9de9b9ff2fe25783e73495;Pe/file/aJCDhGn36ivi/invalida-2023-sk-mkv?bD&c=316549158&De anyone anywhere can download this link even in GERMANY without VPN. It is possible to set it up like this?

zbyna commented

version 3.3.1 sometimes downloading a more parts at the same time. But max speed which i had is 2MB/s. If you will try to download something from uloz.to, then you are limited to speed and number of files. Tor changing you ip and you can bypass it with it. Another one way is VPN, but its the same. For downgrade just run command "pip install ulozto-downloader==3.3.1". Sometimes it has problem with captcha before start of downloading. Then try to kill it with ctrl+z and start again. or try to dele tor temp file from folder. And if speed is slow, then try to kill it and start once again. ulozto-downloader --parts %number_of_parts% --parts-progress "%url%".

Version 3.3.1 does not work for me reliably too. It is like a Russian roulette 🙂 .

It was nice while it lasted

d56Wr commented

3.3.1 "works" for me, but usually it only gets one or two CAPTCHAs right
image

For me as well, it gets right like one out of hundred captchas right but at the end of the day it's still faster than normal download from uloz.to

Sounds more like ulozto updated their captcha images. So the next step would be take a look into some logs if present and make new training model for solving captcha.