Question 1

Why does Komga return 400 Bad Request when syncing reading state for Kobo store books?

Accepted Answer

This is a known issue with Komga's Kobo proxy. The fix is available on the `koboproxy` branch: https://github.com/gotson/komga/tree/koboproxy. Build Komga from that branch to resolve the 400 Bad Request error. Alternatively, the fix from PR #2291 also works. A permanent fix will be included in a future release.

Question 2

Why do Komga series sharing labels disappear after restarting the Docker container when using an NFS mount for the library?

Accepted Answer

This is caused by an NFS mount race condition. If the NFS share is not fully mounted before Komga starts, the library scan sees an empty directory and treats it as a new, empty library. This triggers a full library rebuild with new series IDs, orphaning existing series sharing labels and other metadata linked to the old IDs. To prevent this, configure the library directory to a sub-directory within the NFS mount (e.g., /data/komga instead of the mount root /data). When the mount is missing, Komga will then detect the library folder as unavailable rather than empty, preserving existing series IDs and metadata.

Question 3

Why does Komga OPDS feed only show the first 20 publishers in clients like Panels?

Accepted Answer

This is a regression in Komga 1.29.x where the `/opds/v1.2/publishers` feed was paginated but missing the `rel="next"` link, so OPDS clients (e.g., Panels on iOS) could only see the first 20 publishers. **Fix:** upgrade to Komga 1.29.2 or later. **Workaround on 1.29.0/1.29.1:** manually append `?size=200` (or `?page=1`, `?page=2`) to the publishers URL to retrieve more entries, e.g., `GET /opds/v1.2/publishers?size=200`.

Question 4

Why does Komga KOReader progress sync fail to find some books (fileHashKoreader hash mismatch)?

Accepted Answer

This is a bug in Komga ≤1.29.1: when an EPUB is small enough that one of KOReader's sample offsets falls within 1024 bytes of the end of file, the hash computation included stale buffer bytes instead of only the bytes actually read, so Komga's `fileHashKoreader` differs from KOReader's `partialMD5` and progress sync can't match the book. Fix: upgrade to Komga 1.29.2 or later, where the hash is computed with `digest.update(buffer, 0, s)`. After upgrading, the affected books' hashes need to be recomputed (re-run the KOReader hash task on the library) so sync can find them.

Question 5

How to fix Komga OPDS v2 auth logo URL and Latest Series pagination pointing to books?

Accepted Answer

Upgrade Komga from 1.24.3 to 1.24.4 or later; both OPDS v2 issues were fixed there. As a temporary workaround, manually insert the missing slash in the logo URL (for example, `/komga/android-chrome-512x512.png`) and replace the incorrect pagination link `/opds/v2/libraries/books/latest?page=1` with `/opds/v2/libraries/series/latest?page=1`.

Komga

Core Features

What It Can't Do

Use Cases

Tags

Getting Started

Similar Projects