Warnings from database diagnosis

[birdbrain]

IMatch suggested I do a database diagnosis, as one had not been done for about 20 days.

I did so and it reported 10 warnings. Excerpt from log file is below.

Is there something I should know and or do?  (IMatch provides no direction about what caused this, potential damage to data, or actions to take).

Joe

PersonManager Stats:
    Files with faces: 26
    Persons: 9
    Faces: 44
    Confirmed: 44
    Unconfirmed: 0
    Unassigned: 0
    Warning: File [318] face [18] has a mismatch in face/anno sort order.
    Warning: File [319] face [20] has a mismatch in face/anno sort order.
    Warning: File [320] face [25] has a mismatch in face/anno sort order.
    Warning: File [322] face [31] has a mismatch in face/anno sort order.
    Warning: File [330] face [21] has a mismatch in face/anno sort order.
    Warning: File [331] face [27] has a mismatch in face/anno sort order.
    Warning: File [333] face [24] has a mismatch in face/anno sort order.
    Warning: File [339] face [37] has a mismatch in face/anno sort order.
    Warning: File [340] face [39] has a mismatch in face/anno sort order.
    Files: 26
    Faces: 44
    Entities: 44. Confirmed: 44
    Entity keywords: 44.
    Vector data size: 22KB
    Thumbnail data size: 363KB
    Annotations without face data: 14KB
Completed.

[Mario]

The sort order of some faces in your databases has been broken. Not important.
Run the diagnosis again. Has it been fixed?

Please also detail your workflow, how you work with face recognition, how you create and work with manual face annotations, if you manually reorder faces in your images etc.
It's hard and requires some effort to re-order faces in the 'wrong' way. All details matter, e.g. what you did in which order.
Do you files contain XMP face regions produced by other software (especially, Mylio)?

[birdbrain]

The sort order of some faces in your databases has been broken. Not important.
Run the diagnosis again. Has it been fixed?

I ran it again and this time only 2 files have warnings. Perhaps because I imported 2 more folders.

Do you files contain XMP face regions produced by other software (especially, Mylio)?

I only have a few XMP sidecars from my trial with Mylio and yes I kept them to help with face recog. After I import photos to IMatch, I immediately delete the sidecars. I have no idea what Mylio actually puts in the JPG itself.
I have very few files with XMP sidecars, so sounds like this should not be an overall concern based on your "Not important" comment.

Please also detail your workflow, how you work with face recognition, how you create and work with manual face annotations, if you manually reorder faces in your images etc.
It's hard and requires some effort to re-order faces in the 'wrong' way. All details matter, e.g. what you did in which order.

I appreciate the need to re-create steps, etc. to solve an issue, but this sounds benign. I am glad to do so for something critical, in fact about to post one now.

Thank you!

[Mario]

IMatch has probably imported the face annotations from the Mylio XMP?
Or did you run face recognition in IMatch only?
Did you change the face order? Removed existing face annotations? Moved them to another layer?

The mismatch can happen, but I don't know how you managed to do it. And the diagnosis finds and resolves it.
Although, IMatch imports existing face regions and orders them in the order in which they appear in the XMP record.
Hence my questions about what you did, how and in which order.

For IMatch, not only the fact that a file has face annotations, but also the way they are ordered is important: Ordering Persons in Images

Not sure if and how Mylio comes into play here. So far I was not impressed of how it creates and manages metadata.
Probably metadata is of little importance for the typical Mylio user.