No information in status message

Duplicati tells me that there are errors:
image

When I click more information, it shows this:
image

Can this page more explanatory? For example which files are affected.

Try clicking on the lines with dates and times - even though the “hand” icon doesn’t show it should expand to show details like below.
image

Is there an easy way to export the 4112 errors that my backup gave to a text file to analyze them in more detail?

A simple method would be copy/paste.

A complex method would be a sqlite tool and some SQL to export the log table contents.

If the errors are repeatable you could add --log-file and --log-file-log-level=profiling to the job and run it again.

A copy paste does only give the unexpaned items, and the do not contain any information.

I am not sure if the errors would repeat. But if a backuptool gives errors, I better check them or save them somewhere. It would be very handy if the GUI has a button: save all mesages/errors.

That’s a great idea!

Because the logs (as stored in the database) can be quite long how would you expect such a button to work - save ALL entries for the log being viewed or just the ones currently on the screen?

Maybe several options:

  1. Entries
  2. Entries of a certain job
  3. Entries of last run

With two selectors:

  1. the log level (Errors, Warnings, etc)
  2. Maximum number of messages (but I have the feeling one will never use this option)

What? That’s totally the option I would use pretty much all the time! :smiley:

Before I go create something on GitHub I should probably check with @kenkendk or @kees-z to make sure there isn’t already something like a command line option to export logs to a text file. I know there’s create-report but that’s not really targeted at end users…

The problem is that the number of errors can be really large and take up several MB. For that reason, they are compacted to 10, showing ... for the rest. You can change that with --full-results, which will expand the results.

Underlying, they problem is: “how much logging do we save?”. In principle we save it all, but in practice this becomes unmanageable. More annoying, when we throw something away, we only discover later that we wanted it.

I am open to changing this, as I see that it is not optimal, but I am not sure what the best strategy is.