After releasing of version 2.023.21, me and my partners (more than 20-30 people) started complaining about submission duplications. The duplication rate increased drastically after the release. I am using the humanitarian server. We didn’t see any patterns for duplication, it looks random but there is no way to avoid it. Not only uuid is duplicating, the id is doesn’t duplicating
example form duplication on form: aGy93uHnGfNrKoqAn3GiRW
Hi, we mainly use Enketo, but some field teams use Kobo Collect. It looks both teams are facing a similar issue. Yes Kal_Lam, we are still facing the same issue, it all started after the release and still continue (7/20/2023 09:19 a.m)
I can provide you with more codes; Those are under the same form I shared the information with you. If needed, i can share examples from other forms that are used by different partners.
As far as I know, the documentation/forum answers for Kobo/ODK says that _uuid and also _id can be regarded as unique keys and users count on this, like we count on unique primary key in a database. (cc: @jnm)
_id is absolutely unique on Kobo (within a particular server, e.g. _id will never be duplicated on kf.kobotoolbox.org). It is literally the primary key of a relational database table.
_uuidshould be unique according to the ODK specifications, but sometimes is not. We’ve prioritized accepting data from clients even if it contains duplicate UUIDs. That is changing, and we will soon reject any submission that duplicates a UUID, even if the rest of the submission content differs. However, we’ll also have to deal with de-duplicating UUIDs among existing submissions—and this is not a trivial process.
Hello @Kal_Lam and @jnm , I am also facing this issue. We’ve deployed a survey and there are currently two forms with different data from the same UUID. Can you please advise on how to fix this? I’ve linked the kobo sychronous excel to PowerBI and so I dont know how to edit the backend to change this (but would prefer to do it there rather than in PBI).