Feature Request: Auto-skip unreadable/problematic titles and continue with the rest + end-of-rip report

Everything related to MakeMKV
Post Reply
NESSNES404
Posts: 8
Joined: Thu Dec 22, 2022 1:03 pm

Feature Request: Auto-skip unreadable/problematic titles and continue with the rest + end-of-rip report

Post by NESSNES404 »

Hi everyone,

I’ve been a long-time MakeMKV user and it’s generally fantastic, but I keep running into one very frustrating scenario with discs that contain multiple titles (TV seasons, box sets, movie + extras discs, etc.).

The problem:
A disc has, say, 8–10 titles. Title #1 (or any early title) has bad sectors or is otherwise unreadable after retries. MakeMKV spends hours hammering on that one bad title, effectively stalling the entire job. The other 7–9 titles on the same disc are probably perfectly fine, but I can’t get to them without wasting a huge amount of time (or babysitting and manually intervening).

From what I’ve seen in other threads, when a title fails hard, MakeMKV can abort the remaining titles instead of gracefully moving on. This is especially painful on damaged or aging discs where only one or two items are bad.

Proposed feature:
Add an option (in Preferences or as a per-rip checkbox) something like:
  • “On unrecoverable read error for a title: [ ] Abort entire process [x] Skip the problematic title and continue with remaining selected titles”
When this is enabled:

1. MakeMKV performs its normal retries on a title.
2. If it still can’t read the title reliably, it logs the failure clearly, skips that title entirely (no partial/corrupted MKV is created), and moves on to the next selected title.
3. At the very end of processing the disc (or the batch), it shows a clear summary/report, for example:
Rip Summary
Successfully ripped: 7 titles
Skipped due to errors: 2 titles
Title 03 (Main movie) – Unrecoverable read error / bad sectors after 10 retries
Title 07 (Deleted scenes) – Hash check failed / read errors

This way I don’t lose hours waiting on one bad file, I still get all the good content off the disc in one pass, and I have a clear record of exactly what was missed (so I know whether to clean the disc, try another drive, or source a replacement for just those items).

Why this is different from previous “skip bad sectors” requests:
I completely understand (and agree with) the existing stance that MakeMKV should not create corrupted/partial video files by skipping sectors inside a stream. My request is not about patching around bad sectors inside a title. It’s about treating an entire title as unrecoverable and simply moving on to the other titles on the disc — exactly the behavior most people want when only one episode or one extra on a disc is damaged.

There’s already a registry hack (io_IgnoreReadErrors) that some people use, but an official, visible option with proper logging and a final report would be far more user-friendly and discoverable.

Has anyone else hit this “one bad title kills hours of progress” situation? Any current workarounds besides manually selecting titles one-by-one or constantly ejecting/re-inserting the disc?

Thanks for considering this — it would be a huge quality-of-life improvement for anyone ripping large collections or older/used discs.

Best regards,
Nicholas
Woodstock
Posts: 10936
Joined: Sun Jul 24, 2011 11:21 pm

Re: Feature Request: Auto-skip unreadable/problematic titles and continue with the rest + end-of-rip report

Post by Woodstock »

One problem with this is that sometimes, the retries aren't in MakeMKV's hands. The operating system controls retries most of the time, and it's only after the OS gives up that the retry settings in MakeMKV become a factor.

This is most notable if you try to abort the rip when the retries start. MakeMKV acknowledges your attempt to restart, but it can't do anything until the OS returns from the read request. I've had it take 20 minutes to abort some rips, but be instantaneous on others. I used to set my retry count high (30 or so) back when I had older DVD drives, but gave up on that back when I was ripping Hercules (one bad disk out of 20+ in multiple sets).

Maybe add to your request finding a way to get "inside" the OS call, so the attempt to stop this particular call to skip gets handled immediately, instead of waiting for the OS return to process the skip request.
Post Reply