Ocassional problem: changes in row(s) not saved

I have been warned from one user of my database, that in some rows (it is more than just a few rows - more tens, let’s say 30-60 rows) some data were not saved/updated.
I cannot imagine what could be the reason that some data are missing (f.e. text fields updated with different text or single-select field option updated by another single-select option), but I better want to report it and ask IF YOU OR SOMEONE ELSE FROM THE COMMUNITY HAS THE SAME EXPERIENCE. ???

What I would say is that this user made a mistake - f.e. he updated the data field but did not move to another row so the row with the updated field wasn’t saved, instead of he maybe closed the laptop lid (so the laptop went to sleep mode and connection to the database was lost). Next time when he woke upt his laptop, he continued to work.
Not sure, just guessing…

Also, if a user opens his laptop lid (and makes it wake up), there is no evidence that the database connection is not established. So there are situations when someone works and fills data fields with new or updated data, they are simply not received by the database.

Could it be because of how web browser works (cache - so there is no indicator, that the connection is not established)? If yes, then is it possible to disable the web browser cache just for the database? Or Add some kind of indicator (small red or green light in a corner) which shows if client (user’s computer) is connected to the database correctly or not)? Because when a PC goes back from sleep mode and a browser with the baserow page is just in foreground (as it was, when the PC went to sleep mode), there is never this small pop-up display, until I do a page refresh:


My environment is:

  • baserow self-hosted as a docker container (on a home Synology NAS device)
  • the mentioned user with the problems described above is connected to baserow on the same local network via local IP (so there is no external connection or interned involved in our workflow)

Would you have any conclusion to these issues? Maybe think about it, if it would be possible to make some changes to avoid these critical situations. Let

I am opened to any testing phases, providing logs etc. if needed.

Hello @marcus, I apologize for the delay in responding. I have asked the development team to look into your issue.

OK, thanks.
I am monitoring the situation. If the same issue will occur again (it will be reported to me), I will let you know as soon as possible.

Hey @marcus,

I can’t think of any valid reason for this issue. We know that users have to refresh the page after waking up from standby, and we plan to fix this. However, every update request should either succeed and save the data or fail and display an error message in the UI.

Thank you for keeping an eye on the situation. If you have any suggestions on how we can reproduce this issue, please let us know. We’ll be happy to fix it as soon as possible.

hi @davide @olgatrykush
It happened again today - a problem with data not writing into cells. A member who was working with baserow, was connected on the same (local) network as usual and there is no evidence that his/her computer could be not connected (logged in) properly to baserow.

Together with it also “Table not found” error occured, I mean database did not respond properly, like there would be no such a table.

I described it here:

Also I downloaded the log from my baserow docker container immediately after it happened. How can I provide it to you (it has some 16 MB - csv or html format).

Any help with this will be very welcomed.

Hi @marcus,

can you send me the logs in private here or at davide at baserow.io ?

I’ll look into it, and I’ll let you know ASAP.

Done. You will find it as zipped attachment in you email.
Thank you very much fort your help!

Hey @marcus,

I’ve looked into the log files and I believe I have identified the issue. I will discuss it with the team first thing on Monday and will get back to you as soon as possible.

Great, thanks.
Looking forward for a solution or patch (or small update) eagerly because this bug is, as I believe, very critical.

Hi, please, do you already have any conclusion to solve the issue described in this thread?

Hi @marcus,

we investigated the issue, and we think this MR will solve the issue Draft: Don’t clear Django global apps cache.

Because it’s a very important change, we need to properly test it, but I can tell that the plan is to include it in the next release.

At the moment, the release is planned for the next week.

OK, thanks for the info. Update in next week sounds good :+1: