@andrewsjakkko02
What database software and what sessiontype (file or database) is used?

To reproduce the issue we all would have to use the same version/build of LimeSurvey.
The survey seems to be still running on 6.6.0+240729.

Did you export the responses into a file to see if the first response is in the file?
Perhaps there is a small bug in LimeSurvey hiding the first response in the GUI, but it is still saved in the database.
 

Ok maybe the issue was that I was trying to fill the surveys from the same device as the admin's one. Now that I tried a different device it seems that the answers are coming through. Anyway

> All three respondents submitted the survey, but there are only two records in the response table?

Yes, that's exactly what happened, and the missing response was always the first submitted.

Anyway I also tried starting filling the survey from another device, then closing it and reopening from the admin's device to finish filling and that also didn't work, so I'm guessing I just need to avoid using the admin's device to fill surveys (that won't be a problem during the real election), and advise voters that once they opened the survey on one device they should not switch to another device to finish filling it in.

Thank you again, I hope this will solve my issues, and that we will be able to use this system to have a smooth election at the end of the month.

Best,
Andrea

I don't really know where to find this information you're asking for, sorry. I tried exporting the responses but the first one simply isn't there. Anyway maybe we've solved the issue (as you may see in my previous reply, to Joffm), I will be running more tests with other people as well to see if that's actually the case. Thank you for your support.

Best,
Andrea

andrewsjakkko02 wrote:

Ok maybe the issue was that I was trying to fill the surveys from the same device as the admin's one. Now that I tried a different device it seems that the answers are coming through.
 

That could be the rootcause of your issue. When testing surveys from your logged in admin account alway use a different browser of the private mode of your browser to prevent that your admin session is used when filling out test surveys.
There are all kind of issues which can occur but hard to reproduce. The session handlng is a mess. You can save yourself a lot of time and pain when you not test surveys as survey admin. Sounds silly but guarantees peace of mind.

I second jelo's comment!
When testing always use the "Preview survey" button. It generates a test URL including "...newtest=Y...". That forces a new session to be created which is helpful for testing.
When publishing the survey use the link shown at "Share survey". It comes without "newtest=Y" which is better for the common user because the system will try to re-load an unfinished survey from the survey session data e.g. in case someone closes the survey accidentally.

Ok maybe the issue was that I was trying to fill the surveys from the same device as the admin's one.

I am pretty sure this is what causes the problem. This issue exists for quite some time, when you are testing while logged in as a survey admin.

@Denis: And the anonymity issue doesn't exist just since LS 6. This has been a "feature" of anonymous mode in Limesurvey for quite some time and with quite some controversial discussions, because there are some people in the Limesurvey team that see no issue at all with an anonymous mode that allows to pass on token information to the survey, which in my opinion makes the anonymous mode completely obsolete.

bugs.limesurvey.org/view.php?id=18253

Still not agreeing with this "anonymous" mode, as it is NOT anonymous. But hey, seems like my "anonymity fundalism" is not en vogue anymore as so many in the team did not see any issues with the way the "anonymous" mode works in Limesurvey. I can just say: Stay away from it, as it doesn't ensure any annonymity at all the way it is at the moment. But that is a different topic...