Convert not working after upgrading to Calibre-web 0.6.22 #3100
Comments
|
So, if I stop calibre-server, everything works fine... I don't get it, I've always had calibre-server and calibre-web working together without issues... Is there a change in this version that makes them not compatible? |
|
Could you please set the log level to debug and check it again.
maybe. Seems you have activated embed metadata on conversion/download… maybe this interferes somehow |
I'm afraid I don't know how to do that, but I'm willing to follow instructions. I can see there's a "Download debug package" in Administration, but it gives the same info as normal log files.
Again, if I've done that, I don't know how I've done it. If you could tell me how to deactivate that feature, I can try again and share the logs. Thanks four your help! |
|
Debug log: admin section -> basic configuration -> logfile configuration -> Log Level -> DEBUG |
Yes! That was it. As soon as I disable that option, converting books works fine again. Anyway, in case it is useful, these are the debug logs with "embed metadata..." enabled: Thanks a lot for your help!! |
|
Calibre is locking the database, so there is no workaround. The code now directly outputs the calibre error message: |
I also suffer from the problem running calibre-server in parallel. But does the message mean, that using the --with-library option together with the URL of the running calibre-server should work? |
|
You need to deactivate the embed metadata option, otherwise there is nothing you can do |
I understand that this is currently the case. But, if I understand it correctly, |
|
Okay, now I got it, I‘ll check |
|
A very quick and very dirty test seems to work in my case: --- embed_helper.py.240716 2024-07-16 09:04:17.511906420 +0200
+++ embed_helper.py 2024-07-16 09:04:56.515297154 +0200
@@ -37,7 +37,7 @@
my_env['CALIBRE_OVERRIDE_DATABASE_PATH'] = os.path.join(config.config_calibre_dir, "metadata.db")
library_path = config.config_calibre_split_dir
else:
- library_path = config.config_calibre_dir
+ library_path = "http://localhost:8080"
opf_command = [calibredb_binarypath, 'export', '--dont-write-opf', '--with-library', library_path,
'--to-dir', tmp_dir, '--formats', book_format, "--template", "{}".format(temp_file_name),
str(book_id)] |
|
I thought a while on it and came to the conclusion that I will not change it. As you mentioned it's an easy fix for you, but for me there are a lot of more use cases to consider: |
|
Just in case it is useful for anyone reading this, I've found a workaround. My use case: Calibre-web serves its contents to 10 friends who, from time to time, want to upload a bunch of files. As doing it through Calibre-web, one by one, would be tedious, I had Calibre-server running alongside, which imports all the uploaded files by my friends once a day. What I've done: besides disabling "embed metadata on conversion/download" (thanks), now Calibre-web is running 24/7 and Calibre-server is disabled. Then, I've created a cron job that, at 1:00 AM, stops cps.service, starts calibre-server, sleeps for 5 seconds, and then imports all the files from the designed folder (a nextcloud share). After that, the files are deleted from the shared folder, calibre-server is stopped and calibre-web is started again. Something like this:
|
Describe the bug/problem
Convert does not work after upgrading to Calibre-web 0.6.22.
Path to Calibre Binaries is correct (/opt/calibre) and indeed, converting from calibre-server works fine.
To Reproduce
Steps to reproduce the behavior:
Converting a books always throws an error.
Logfile
Environment (please complete the following information):
The text was updated successfully, but these errors were encountered: