Issue with Formula Field Not Updating Automatically in Baserow (Docker Self-Host)

I’m using Baserow self-hosted on Docker, and I’ve encountered an issue with a formula field.

I have two fields in my table:

  • A DateTime field named “Time Review”
  • A Formula field with the formula: now() - field('Time Review')

This formula is meant to calculate the time elapsed since the “Time Review” date, displaying a real-time interval that updates automatically. Initially, this worked perfectly. However, in the last few days, the formula field has stopped updating continuously.


Result not updated automatically

Current Issue: To see the updated time interval, I now have to manually go into “Edit Field” for the formula and click “Save.” This manually refreshes the value, but I need it to update automatically without manual intervention.


When click Saved, the result update

Potential Cause? My database setup includes several hundred tables and rows, but I don’t believe this should be too much for Baserow to handle.

Question: Has anyone else faced this issue or found a solution to ensure the formula field with now() updates continuously in Baserow?

Thanks in advance for any advice!

Hey @kieuminhnhat, apologies for the delay with the response. We’ll investigate the issue: Formula field with NOW() function not updating automatically (#3153) · Issues · Baserow / baserow · GitLab :slightly_smiling_face:

1 Like

Thank you for your response.
It’s working good before. I’m not sure why is it happened now. It is same setup.
I updated to lastest version Baserow 1.28

I’m still waiting for this solution. It is stuck and not updated.

Hello @kieuminhnhat,

thank you for your patience! We haven’t started working on this yet as we’re currently focused on other tasks. To help us prioritize your request, we need some additional information to understand better and replicate the bug and assess its potential impact.

Could you confirm if the issue occurs with other tables as well? Additionally, if you check the logs, you should find entries for baserow.contrib.database.fields.tasks.run_periodic_fields_updates. Could you verify if there are any errors around the same time and share the relevant log excerpt with us?

Thank you for your help, we’ll look into it as soon as possible.

Yes I checked again it happen to all other tables.

May I ask to find this log I need to go to Audit log feature?

Yes I checked again it happen to all other tables.

Ok, it likely is something specific to your instance because they are working properly on our SaaS. Any additional information you might have on replicating it is greatly appreciated.

May I ask to find this log I need to go to Audit log feature?

No, if you run the all-in-image of Baserow with Docker, you should be able to see them with Docker Desktop or using the docker logs command.

I have the same problem, I have a formula field to check how much days are left before a call which didn’t update for 100’s of days if no changes where made in the row or if I didn’t manually click on save on the edit formula screen to trigger an update.

My formula: totext(date_diff(‘day’, todate_tz(concat(now(), ’ 03:00:00’), ‘YYYY-MM-DD HH:MI:SS’, ‘Europe/Brussels’), todate_tz(concat(datetime_format(field(‘Datum start’), ‘YYYY-MM-DD’), ’ 03:00:00’), ‘YYYY-MM-DD HH:MI:SS’, ‘Europe/Brussels’)))

I also run baserow self hosted using Docker.

I just running on Docker windows, it is working before.
But suddenly I don’t know why it is stopped working.
Until now tables still not updated this formula field.

We would love to help you solve the issue, but until we can replicate it ourselves, there’s not much we can do to help.

Have you used this guide to run Baserow: Install with Docker?

If you have Docker Desktop, you should see baserow in the list of the running containers. If you click on it, it should show Baserow logs. If you don’t have Docker Desktop, you should be able to see the logs from the terminal with docker logs baserow.

Every 10 minutes, you should see that a task named baserow.contrib.database.fields.tasks.run_periodic_fields_updates() starts and ends without errors, like in the image attached below. Could you please check that? If you see any errors, could you please share the logs?

Hi Davide,

Thank you very much for your reply,
This is screenshot when I search with your task name, it looks like it has no errors.

Hi Davide,

I do get errors for that task!
Anything I can do about it ?

Yes, could you please export the workspace and import it back into a new one to verify if it has the same problem? You should be able to see a similar error for other fields.

If so, please send me (either here or privately) your exported workspace so I can try to debug it in my dev environment. If sensible/private information exists, you can export it without data and change the field names.

Screenshot 2025-01-10 at 15.19.00

This is screenshot when I search with your task name, it looks like it has no errors.

@kieuminhnhat, I cannot see exactly when the task ends. You should see something like Task baserow.contrib.database.fields.tasks.run_periodic_fields_updates[625f67f7-7443-45a9-9814-9c5a48986203] succeeded in 1.5668367509788368s: None or an error like @Marnik 's screnshot.

I got this

Several times on the log, it have some kind of error before this task baserow.contrib.database.fields.tasks.run_periodic_fields_updates

Can you see the problem on these picture, let me know if I can export things. Please help me.

Thank you

I exported and imported the workspace into a new one.
The fields still won’t update and there are indeed similar errors for the new copied fields.

I want to send the export privately but can’t seem to find your email or a way to contact you privately.

Ok ,this guide should solve your issue: Baserow
The numbers there are very high, but consider a default of 64, that error can happen even in situation with less tables.

Please try to follow the guide and double that number to see if it fixes the problem.

You can send me your exported workspace to davide at baserow.io or as a private message here, by clicking on my name and then on “Message”

Upon reviewing the logs, I discovered that some of the fields showing update failures don’t actually exist in the database. Maybe it’s trying to update inexisting fields and that’s why others fail too ?