mirror of
https://github.com/qbittorrent/qBittorrent.git
synced 2026-03-02 22:57:32 -05:00
qBittorrent 4.1.9.1 is unable to re-check and restarts partially/completed downloads if you re-add a torrent. #9328
Labels
No labels
Accessibility
AppImage
Bounty
Build system
CI
Can't reproduce
Code cleanup
Confirmed bug
Confirmed bug
Core
Crash
Data loss
Discussion
Docker
Documentation
Duplicate
Feature
Feature request
Feature request
Feature request
Filters
Flatpak
GUI
Has workaround
I2P
Invalid
Libtorrent
Look and feel
Meta
NSIS
Network
Not an issue
OS: *BSD
OS: Linux
OS: Windows
OS: macOS
PPA
Performance
Project management
Proxy/VPN
Qt bugs
Qt6 compat
RSS
Search engine
Security
Temp folder
Themes
Translations
Triggers
Waiting diagnosis
Waiting info
Waiting upstream
Waiting web implementation
Watched folders
WebAPI
WebUI
autoCloseOldIssue
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
starred/qBittorrent#9328
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @ezxpro on GitHub (Nov 16, 2019).
qBittorrent version and Operating System
4.1.9.1 Windows 10 x64 (10.0.1xxxx)
If on linux, libtorrent-rasterbar and Qt version
(type here)
What is the problem
qBittorrent 4.1.9.1 recommences partly or complete downloaded torrents upon removing and re-adding them to the list (I made sure I was setting it to download in the same folder as before).
After having this issues, I tried other torrent clients that I know could continue the downloads started by qbittorrent, but I found out qbittorrent had really messed them up, so the alternative client would recheck and start from where qbittorrent left of, namely, from the beginning, since it somehow erased all progress previously made. (In other words, the altenative clients did their job right, but qbittorrent had already messed up)
What is the expected behavior
Upon re-adding a torrent that was already completed or being downloaded and selecting the same download folder, it is expected a re-check will take place and already downloaded pieces will be retained.
Steps to reproduce
a. currently being downloaded or
b. is already downloaded (the bug happenned to me in both these scenarios)
3, Uncheck (if checked) the checkbox that causes files to be deleted from disk.
4.Re-add torrent (either through torrent file or magnet link, I tried them both and had the same result)
Extra info(if any)
Before having this problem, I was on qBittorrent 4.1.8, that I was using since the first day it was released. Today, when I turned my computer up and opened qBittorrent, all torrents disappeared, although I could see the completed/seeding ones in the "all" section of the status bar. My downloading torrents, however, didn't show up at all, so I closed qbittorrent and relaunched it and then my torrents were all there again, however, the downloading torrents would now show up under the "completed" section and the progress bar would show as empty and with 0%
progress. I also checked that all downloaded pieces were there in the torrent statuses, so I performed a re-check on these torrents, but they would still appear under the "completed" section and refuse to continue downloading.
I also restarted qbittorrent 4.1.8 several times and performed several re-checks on these torrents, tried force-starting them, but had no results.
Therefore, I decided to de-install version 4.1.8 and installed version 4.1.9.1 and the problem persisted, so I decided to remove and re-add the torrents and found out that qbittorrent 4.1.9.1 is actually unable to recheck and resume torrents when you remove and re-add them. I did it with only two of my torrents, and in both cases it messes up and re-starts the torernt.
@0xjmux commented on GitHub (Nov 18, 2019):
I am also experiencing this exact same problem.
In my case, I paused the torrent due to disconnecting from a network, and when I came back to restart it later, qBitTorrent threw an IO error. Even with a force recheck, qBitTorrent is unable to find the files.
The directory the files are being downloaded into hasn't changed, and even with me changing it to a different directory and changing it back (a fix I saw for an older bug online) it still will not find the partially completed torrent.
In my case it was a rather large torrent (200+gb), and it was about halfway downloaded, but I don't know if that is a contributor to the problem or not.
@Seeker2 commented on GitHub (Nov 18, 2019):
I had a vaguely similar problem back in 2013:
https://github.com/qbittorrent/qBittorrent/issues/684#issuecomment-20799222
An automatic move-on-finish of the torrent by qBitTorrent might partially be to blame for my problem...or be completely unrelated!
@ezxpro commented on GitHub (Nov 18, 2019):
I lost a 25GB download. Wouldn't be a problem if it didn't have only one or two very slow seeds.
I'm using Vuze instead of qBitTorrent now. Last time I used Vuze it had just been released and I didn't like it. But now it is much better.
@Kolcha commented on GitHub (Nov 19, 2019):
data loss after re-adding very often is observed when "append !qB extension" option is enabled. with this option disabled have no issues with re-adding.
@0xjmux commented on GitHub (Nov 24, 2019):
Kolcha The "append !qB extension" setting has never been enabled on my machine. I've verified that the setting is disabled, and qBitTorrent still can't find the already downloaded files.
@ezxpro commented on GitHub (Nov 27, 2019):
Where does one disable/enable this setting?
By the way, I've never fiddled with any such setting before having the issue herein reported.
@Kolcha commented on GitHub (Nov 28, 2019):
open settings dialog, go to "Downloads" page, find checkbox called "Append .!qB extension to incomplete files".
@ezxpro commented on GitHub (Nov 28, 2019):
This feature was disabled, yet I had this Data Loss.
@jenxi commented on GitHub (Jan 9, 2020):
I'm on version 4.2.1 and I'm having this issue as well. I suddenly have the I/O error and now all my seeds are restarting from 0%. Force recheck does not resolve this issue.
@dimitarvp commented on GitHub (Jan 18, 2020):
I still have one fairly small torrent (~6GB) that starts downloading and fails with an I/O error, every time on startup, and refuses to resume until I restart the client. Never even gets to 0.1%.
It uses the same disk all others do, the disk is thoroughly checked and has zero defective sectors (trust me it took a while to check such a huge drive!), all settings seem very generours -- big disk caches, a lot of upload slots, unlimited speed, uTP + TCP are used, Encryption is not enforced, etc. I can't see anything that would cause the issue (I even read the docs of the libtorrent settings and mine seemed default and sane).
@Seeker2 commented on GitHub (Jan 18, 2020):
You getting lots of piece fails in qBitTorrent's logger?
A big disk cache can cause an I/O error in qBitTorrent due to its 32bit nature...assuming that's the error you're having.
@dimitarvp commented on GitHub (Feb 13, 2020):
@Seeker2 here's what I am getting today again:
I swapped out real values with
/anonymised/pathand<torrent_name>.I seriously have no clue what's going on.
macOS 10.15.3 Catalina
qBittorrent 4.2.1 (64-bit)
I have full ownership and security rights to both the external drives I am using, and they have exactly the same info in Disk Utility, minus the mount points and drive numbers.
Disk cache is 32MB.
One note though: for some reason the Mac's Finder shows "You have custom access" in the Sharing & Permissions section of the "Get Info" dialog. But I am presuming this is because I granted full rights to everyone using the Mac at the moment (namely
chmod 0777).@dimitarvp commented on GitHub (Feb 13, 2020):
No, I get one error like the one I pasted and redacted above, and then the downloading of the whole torrent stops immediately. This is about torrents with more than one file in them.
@dimitarvp commented on GitHub (Feb 13, 2020):
Preliminary findings after relentlessly chasing this for half a day:
Seems the problem is that the external disks were formatted with exFAT. Changed one of them to HFS+ and qBittorrent had zero problems writing to it. I suspect it will also work fine with APFS but not going to try.
This reddit thread pointed me in the right direction: https://www.reddit.com/r/torrents/comments/ewjec8/operation_not_supported_creating_hard_links_on_a/
Of course, I am not sure that qBittorrent actually needs to create hard links but it doesn't matter much since changing the file system of one of my external disks immediately solved the problem.
The reddit thread also contained a link to this file systems feature matrix: https://docs.microsoft.com/en-us/windows/win32/fileio/filesystem-functionality-comparison?redirectedfrom=MSDN
@Chocobo1 commented on GitHub (Feb 13, 2020):
@dimitarvp
Your latest findings are very interesting, the
file_fallocateerror almost sounds like something is wrong in libtorrent. I really encourage you to bring the info over to https://github.com/arvidn/libtorrent/issues.@xavier2k6 commented on GitHub (Mar 12, 2025):
This ticket has been closed due to being "out-of-date", and thus either most likely resolved in recent versions or no longer applicable.
If you experience the reported problem or similar in the latest version, please open a new issue report with the requested information in the issue template.
A new issue report with relevant updated data gathered from the latest version is preferable to necroing an old report with a comment like "still happens in version x.y.z", even if you think the bug is the same, or suspect of a regression.
Due to the changes made to the qBittorrent code and its dependencies over time, the exact cause of your problem could be totally different than the original one, despite the visible symptoms of the bug being similar.
Thus, providing relevant updated information is crucial to find and fix the root cause of a recurrent problem or regression.
Thank you for your contribution(s).