New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Files can be assembled after the job is removed from the queue #1509
Labels
Comments
So ... what is the real problem? CPU power wasted? Time wasted until the next download starts? How much? And does it only happen with small files / downloads? |
Crash of the assembler (see the URL, a few lines down) resulting in the job disappearing after the restart. |
Safihre
added a commit
that referenced
this issue
Aug 30, 2020
Relates to #1509 Hopefully it stops the failing tests.
Safihre
added a commit
that referenced
this issue
Nov 1, 2020
Safihre
added a commit
that referenced
this issue
Dec 12, 2020
Removed the par2 files for the unicode job, they caused too much problems. It's a bad "fix" for #1509.
Safihre
added a commit
that referenced
this issue
Dec 12, 2020
Removed the par2 files for the unicode job, they caused too much problems. It's a bad "fix" for #1509.
This was referenced Feb 3, 2022
Safihre
added a commit
that referenced
this issue
Feb 7, 2022
Safihre
added a commit
that referenced
this issue
Feb 7, 2022
Closed by 9ad5bd4 in 3.5.1RC2. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Using the automated testing I found that in special circumstances it can be that files are still being decoded/assembled while the job has already finished.
It seems to occur because only after the first par2 is received, we postpone the rest of the par2 files we don't need. But in the case of the test and the small files, these were already being downloaded.
Solutions:
nzo.deleted
. Have to check the impact when extra par2 is added.At first I thought, based on the log output show, that end-of-job was triggered multiple times while parts are still being decoded. This turned out not to be the case, since
nzo.remove_article
is locked, so the end-of-job is really only triggered once. This took me hours of debugging to find out.The text was updated successfully, but these errors were encountered: