Data Entry Weekly Period - 2026

Support - Assistance technique
1

Hello everyone,

I am using DHIS2 version 2.42.6 with a weekly dataset and have encountered an issue related to the Organisation Unit (OU) opening date.

If the week starts on Monday, Week 1 of 2026 should run from Monday, 29 December 2025, to Sunday, 4 January 2026.

However, when the OU opening date is set to 29 December 2025, DHIS2 does not allow me to capture data for Week 1 of 2026. Data capture only becomes possible when I change the OU opening date to 28 December 2025.

I would expect the system to allow data capture from 29 December 2025, since this is the first day of Week 1 of 2026 when using Monday as the first day of the week.

My questions are:

  1. Is this expected behaviour in DHIS2 2.42.6 when using a weekly dataset?

  2. Why does the OU opening date need to be set to 28 December 2025 instead of 29 December 2025?

  3. Could this behaviour be related to the existing issue DHIS2-20765?

Has anyone encountered similar behaviour or can explain how DHIS2 handles Organisation Unit opening dates in relation to weekly reporting periods?

Thank you for your help!

2

Hi Allan,

This behaviour in DHIS2 version 2.42.6 does not appear to be expected. If the dataset uses a Monday-starting weekly period, an OU opening date of 29 December 2025 should normally permit data entry for Week 1 of 2026 (29 December 2025 – 4 January 2026).

The fact that changing the opening date to 28 December resolves the issue suggests that the problem may lie in how DHIS2 evaluates OU opening dates against reporting-period boundaries. However, the exact cause has not been confirmed.

I suggest the following:

  1. Verify the period type: Confirm that the dataset uses the intended Monday-starting weekly period type.
  2. Check DHIS2-20765: Review the issue to see whether it describes a similar problem.
  3. Use a temporary workaround if necessary: Setting the opening date to 28 December may restore data entry, but it changes the OU’s recorded opening date. Avoid making this a permanent change unless it accurately reflects the OU’s operational history.
  4. Investigate further: If the issue persists, test on a newer compatible patch release or report it to the DHIS2 team, including the version, dataset period type, affected dates, and steps to reproduce the behaviour.

I would not recommend changing server or JVM timezone settings unless further investigation points to a timezone-related issue.

In summary: This may be a period-boundary validation issue, but further investigation is needed to establish the root cause and determine whether it is related to DHIS2-20765.

3

Hi Allan
My understanding is that a weekly dataset data would only be ‘available’ for capture after the end of the week. If your week starts on Monday 29 December, then you should only be capturing data on the Monday 5 January. I am not sure why you would want to capture data on Monday 29 December, as that is the beginning of the week and the data for the whole week is not available. I know that originally it was set up (DHIS1.4) to stop data entry before the reporting period ends.
Not sure if this helps
Norah

4

Thanks for these suggestions
For now we have opted to use a temporary work around and the issue is not blocking.
We have also decided to test on a newer patch in the future, We will see how it behaves and raise it should the need be
Thanks a lot!

5

Hi Norah,

The EPI weeks in 2026 start on 29 December 2025, with the first week covering 29 December 2025 to 4 January 2026. The issue is that the system does not allow the reporting period to be captured when the OU opening date is set to 29 December, even though this is the start of the reporting week.

The issue is not blocking data entry, as a workaround exists, so its priority is low.

Work Around: Set the Opening date of the OU to a date before the start of the Epi week

Thanks!