Last modified time field updated with no evidence of how

[Jean_Thoensen]

We have a "Last modified time" field that updates when either of two other fields are changed. We discovered a group of records all modified on the same date and time (05/11/2023 at 12:50 PM) but when we opened each record to see the revision history, there is no trace of how the records were touched. How could that happen?

There are no automations that run on a schedule.

There is an automation that runs when certain triggers are met. We looked at the run histories but none of them ran at that time.

Any ideas?

[kuovonne]

What are the field types?

If they are linked record fields, they could have change if the primary field values of the linked records changed. 

If they are text fields, could someone have very quickly typed and then undo the typing? If it is done fast enough, sometimes it doesn't show up in record history, but could have still triggered the last modified time.

[Jean_Thoensen]

Thanks for your reply, @kuovonne. One field is a single select and the other is a linked record.

The single select field is the job that a person is scheduled to do. The linked record is the location where they will be working.

The strangest thing is that all of these records have the same last modified timestamp of 05/11/2023 at 12:50 PM. All 9 records have assigned workers to unique work locations. Given the small size of the team and our work flow, it is impossible for one person to have assigned 9 people in only one minute. That's why we're so puzzled.

[kuovonne]

Could the value of the primary field in the linked table have changed? What does its record history say? Is the primary field in the linked table a formula field or an editable field? If it is a formula, could the formula have changed?

[Jean_Thoensen]

The primary key of the record is a formula. The formula itself didn't change.

The primary key of the linked record is also a formula, but its formula didn't change, either.

Nothing in the revision history. And now that it has been more than a week since the anomaly was observed, it is impossible to tell anything from the revision history because it groups by week only. Utter useless.

Thanks for continuing to try to figure this out.

[stephensargent]

Any update on this? I'm experiencing the same problem.