Dear all,
I have a problem with the CPDB...I have >20.000 participants in my database. I'm trying just to empty the database. but I`m (my browser/LS/?)  struggling with the size/speed.
In order to empty the CPDB I always need to set the display-limit to 10.000 per page, wait 5 minutes for LS to finish this task, then select all, wait again for 5 minutes, delete all selected, wait 5 minutes and start again to delete the other 10.000 participants.
Do you have any tipps/tricks on how to use the CPDB with many participants a bit more comfortable?

Thanks!

I`m using LS Cloud 3.27.19
 

To be honest, you are probably one of the few users that uses this feature and then one of the few of the few who use it with such an amount of participants.

I have my doubts if it has ever been tested with so many participants. And my guess would be, that this has a lot to do with your server. 10.000 per page is quite a lot and also might cause your browser to choke.

And there is a good chance, that the way this was programmed in Limesurvey isn't helping either, but I am no expert on that and can not confirm this.

On the other hand: How often does it happen that you have to delete 10.000 participants or more at the same time? Sounds like a strange use case to me.

Let's see if someone with more insights on how CPDB is programmed can shed some light. But I don't think that has been going a lot of effort into CPDB in the last couple of years, but I could be wrong there.

I'm new to LS, so what is everyone using for larger non-anonymized surveys if not the CPDB?

EDIT: I import a large csv with my participants and even I spend a lot of time cleaning the data, after importing I got many errors/problems/duplicates. So I delete all, clean my data and import again.

I have a client using CPDB, have the same issue.

I'm new to LS, so what is everyone using for larger non-anonymized surveys if not the CPDB?

I don't know. I guess most users don't have your use case and/or your database size, I would guess.

EDIT: I import a large csv with my participants and even I spend a lot of time cleaning the data, after importing I got many errors/problems/duplicates. So I delete all, clean my data and import again.

OK, this would explain the approach.

To be honest, I have never really used the CPDB, so I have very little knowledge about it.

You don't need to use the CPDB to add participants to a single survey. Do it directly in the survey participants screen.

Thanks! I expected I always need to import to CPDB and then assign participants to a survey.

In general: Both tools are really really uncomfortable to use, especially if you want to use both. Somehow they have the same/similar function, but the usability is really bad, not only if you have a large amount of participants.
They work very differently, especially the csv-import:
- one needs attributes named exactly "atrribute_1", "attribute_2", etc, the other just uses whats written in the csv.
- one always removes duplicates, the other has an option
- and so on...

But it looks like I just need some patience. Seems to be working after a lot of trial and error

Well, this thingy called Limesurvey is open source. You can contribute to it. E.g. if you have coding skills, you can make pull requests for improvements, but even if you are not a skilled programmer (like me) you can help with bug reports, feature requests, suggestions, help with translations, etc.

Limesurvey has grown over years. Some features have been implemented by someone who needed it. When that person doesn't continue to contribute, the function might not be touched for years, until someone programmer needs something specific, etc.

I think the CPDB has not seen much love for years. Now with the add participants feature to a survey, I don't see many usability issues. But that might be the bias of someone who has seen and used it for years.

So feel free to contribute with suggestions on how to improve the process of uploading participants into a survey. For CPDB I don't know if it worth it and if there are any plans to work on that any time soon. This would be a question for the developers.

To be honest...I did not know that LS is open source. I just started a new job and they told me they paid for a LS-licence, so I did not think about LS as an open-source project.

Thumbs up!

Even so I'm doing a lot of programming in my statistics, my knowledge about real programming-langues is zero Thanks to LS I started to learn some CSS, but it's a lot of trial&error.

To be honest...I did not know that LS is open source.

No problem. I just thought I mention it, because it allows others to participate.

I just started a new job and they told me they paid for a LS-license, so I did not think about LS as an open-source project.

Hmmm, hile not every Open Source software is free of charge, there is no real "Limesurvey License" to pay for. So if someone is charging your company a license fee to use Limesurvey, there might be something wrong.

However, there are basically two situations where you pay for Limesurvey (not really, you pay for services around Limesurvey, but it could give the impression one is paying for Limesurvey):

1. SaaS: Someone else is providing Limesurvey as a service and is responsible for hosting and administering Limesurvey. E.g. this is what the Limesurvey GmbH offers. If one is hosting Limesurvey on their own servers, this is irrelevant.

2. ComfortUpdate: If you are using Limesurvey CE (Community Edition), the download and running it on your own server is totally free of charge. However, there is a feature called "ComfortUpdate", developed by Limesurvey GmbH, which makes, as the name says, updating Limesurvey more "comfortable". This is also a paid feature but does not impact other functionalities of Limesurvey. One can still update Limesurvey manually.

Interesting!

We pay for " www.limesurvey.org " to host our survey.

When you search for "Limesurvey" in Google/Germany, the first page is limesurvey.org.
If you look at the price overview, it makes the impression that it is rather shareware and you don't actually get any functionality without paying - for example "25 responses/month"...

So you paying for the SaaS service (my point 1). But in this case you are not paying for the software Limesurvey itself, but rather for someone providing you with a running LS installation, hosting, updating, administration, etc. I guess part of the money also goes into developing Limesurvey. Somehow the developers need to eat. Sometimes.

If you run your own Limesurvey installation you need a server/infrastructure, someone to take care of the server, its security, you need to update (server and Limesurvey), you need to administer, etc. So while the software is free, you still have costs of course.

There are advantages and disadvantages to both approaches.

You'll find the information about open source on exactly the same website you cited, limesurey.org on the bottom right, where the link to this forum is...

And yes, the Limesurvey GmbH SaaS service also gives you a free hosted account to test and play around. It is not very usable for production due to the 25 responses/month which is too low for most real projects, but it allows you to test every feature that Limesurvey offers, there is no limitation in features, which I think is great to get familiar with Limesurvey.

OpenSource also doesn't always mean "free of charge" (though with Limesurvey it does). There are other advantages to open source than "no license costs".