search

erickutcher / httpdownloader

HTTP(S) download manager that uses input/output completion ports (IOCP).
https://erickutcher.github.io/#HTTP_Downloader
558 stars 62 forks source link

Stopped for big file size from github in 1.0.3.6 #148

Open rayman3003 opened 3 years ago

rayman3003 commented 3 years ago

First I must say that I didn't have this problem with older versions of HTTP Downloader.

Problem in this version: HTTP Downloader 1.0.3.6 x64 [Windows 10 x64]

My settings [Portable version]:

http_downloader_settings.zip

The problem:

Since 1.0.3.6, When I try to download files from github, Usually if the file size is more than 50-60 MB, The download stops at the middle of the download by saying "Stopped" & even if I try to "Start" them again, they won't continue the download & stalled there forever.

I captured a video to show the problem:

HTTP.zip

The links for the files that get stopped:

https://github.com/charlielee/boats-animator/releases/download/v0.10.0/boats-animator-0.10.0-win-x64.zip

https://github.com/Tribler/tribler/releases/download/v7.5.4/Tribler_7.5.4_x64.exe

https://github.com/joncampbell123/dosbox-x/releases/download/dosbox-x-v0.83.7/dosbox-x-windows-0.83.7-setup.exe

erickutcher commented 3 years ago

That's quite strange. Your settings show that you have it split the downloads into 2 parts. It seems that one of the parts is failing to start. It's most likely that the first part indicates it supports partial content (HTTP status 206), and the second part is returning something else. That would usually be a problem with the server.

Does it still happen? If it does, you might want to download from GitHub using 1 download part instead.

rayman3003 commented 3 years ago

Does it still happen?

Yes.

If it does, you might want to download from GitHub using 1 download part instead.

Thats a solution. But using 1 part, would decrease the speed.

erickutcher commented 3 years ago

Try this older version and let me know if it happens in it: https://github.com/erickutcher/httpdownloader/releases/download/v1.0.3.0/HTTP_Downloader_64.zip

Be sure to run it in portable mode and set the default download parts to 2.

rayman3003 commented 3 years ago

Try this older version and let me know if it happens in it:

I already said (In the first post) that there is no problem with old versions of HTTP Downloader.

This is a bug in 1.0.3.6.

I don't have this problem with other download managers too. For example I use 8 connections in IDM & it can download from github without any problem.

erickutcher commented 3 years ago

In the Advanced Options there's a checkbox called "Update redirected URL(s) in download list". Uncheck that. Then right click on one of the stopped GitHub downloads and select "Update Download...". Replace the URL with the original one, remove everything from the Cookies tab, and then press Update. That should reset the download allow it to resume from where it left off.

What I think is happening is that GitHub downloads are being redirected to an AWS server and a cookie is being set. The cookie expires after a short amount of time and prevents the download from continuing.

rayman3003 commented 3 years ago

In the Advanced Options there's a checkbox called "Update redirected URL(s) in download list". Uncheck that.

It seems this solution worked;

But unchecking "Update redirected URL(s) in download list", Won't produce download problems from other websites?

erickutcher commented 3 years ago

It shouldn't cause any issues to leave it unchecked. For as long as the program's been around, it's always been set to not update the URL(s).

rayman3003 commented 3 years ago

It shouldn't cause any issues to leave it unchecked. For as long as the program's been around, it's always been set to not update the URL(s).

OK, I will try other downloads & post here, If I see a problem. Thank u.