Pavlovia: Server response 404 on loading Pavlovia 2020.2 libs (core, data, util etc.) for previously running experiment without changes

Online experiments
1

URL of experiment:
https://run.pavlovia.org/frank.papenmeier/expra_replication_glanzer_cunitz_1966/?group=TEST

Description of the problem:
This experiment has been running on Pavlovia without issues previously and we collected a number of participants already.

Yesterday, I received a message that the experiment ist stuck on the “initialising the experiment…” page. We did not perform any changes to the experiment (neither gitlab nor Pavlovia dashboard). So this might be related to the Pavlovia server maintenance performed recently?

The JavaScript Console reports the following errors:
Failed to load resource: the server responded with a status of 404 () data-2020.2.js:1
Failed to load resource: the server responded with a status of 404 () util-2020.2.js:1
Failed to load resource: the server responded with a status of 404 () visual-2020.2.js:1
Failed to load resource: the server responded with a status of 404 () sound-2020.2.js:1

So it looks like the symlinks on Pavlovia are broken?

Who can fix this? @apitiot @jon

All the best,

Frank.

2

Hello @frank.papenmeier,

I have just sorted out your experiment.
The issue had to do with the fact that your project has a html subfolder, with the resources inside, but your index.html and the associated .js files are outside of it. That confused the cloning procedure, which I re-run when we moved to the new server on the 3rd of January.
I believe we encountered the problem earlier, last year, and I manually solved it for you then. It regrettably re-occured on this occasion. I will take steps to mitigate that issues, moving forward.
My apologies for the interruption!
Best wishes,

Alain

3

Hello @apitiot,

no, we did not encounter the problem earlier and it worked without any issues last year. Last year, I first created my experiment on Pavlovia and added the html folder later on. This was a nice workaround to ensure that all resources located within the html folder were downloaded automatically. Although the experiment behaved this way prior to 3rd of January, it does not do so anymore. Following your current fix, I does not hang at the “initialising the experiment…” page anymore, which is great, but it does not automatically download the resources from within the html folder anymore. Any way to fix this too?

All the best,
Frank.

4

Hello again,

Then that makes perfect sense. It also explains why there was not record of the html directory in the experiment entry in the database.
I believe I have taken care of the matter. Would you mind giving it a try?
Cheers,

Alain

5

It works again!

Thanks for the fix,

Frank.

6

Excellent! Sorry again for the interruption.

Alain