I have copy pasted my project (copying groups of elements as this was easy with the copy paste functionality) into a new project, seeing a quick fix wasn't going to happen and I was stuck like a rhino jumping through a hula hoop.

In some cases, the app could not be updated. Usually when the model was changed. In the most of the cases is enough to uninstall the app from device and reinstall it.  If this error was show on the web emulator, so you need to clean your cache and try again.

We also have this Error message "Failed to execute canary query (probably due to zombie transaction)"; Is there already a sollution for this problem?

Hello,

I got the same 3 errors as Lars in my personal cloud while developing an mobile app.

"1 - Upgrade failed - rolling back to previous application version.
Script error for "DummyApp.entities"
https://requirejs.org/docs/errors.html#scripterror

2 - unable to begin transaction: cannot start a transaction within a transaction

3 - Failed to execute canary query (probably due to zombie transaction)"

Does someone have a fix for this? This only happens in personal areas?

Hi

I had the same problem, but not in my personal environment, in our development environment. Did somebody found out why and how to fix it?

Regards

Anneliese

Have you add a plugin in between those versions?

Any solution for this problem?

I had these errors pop up on my mobile app as well.  Along with one occurrence of a  'Cache manifest file is corrupt or invalid' error.  I also had the error:  

     'Upgrade failed - rolling back to previous application version.
      Unable to upgrade attribute 'EndDate' data type to 'Date':
      Unable to convert an existing value to Date.' 

which made no sense because the EndDate field had been created months before and had never been changed.  

I tried just reinstalling the app, and the errors where still there.  The only way I could get the errors to finally go away was just to uninstall the app, restart my mobile device, and then reinstall the app.  Which kind of makes it seem like there is some corruption happening in the storage of the local device itself.  I could be totally wrong thinking that, but I wanted to share what I observed just in case it will eventually help someone ultimately find the true culprit.

Lars De Pauw wrote:

If you have platform version before 10.0.804 this can happen when you make changes to local storage. This is due because, in those versions, first the data model is updated and then the physical tables in the devices and when the second step fails this means that the data model will be inconsistent with the physical tables present in the device which causes the error you are having.

For the 10.0.804.0, the process order was reversed. So if an error occurs during the change of the physical tables, the meta-model remains unchanged.

To you fix you need to force a table to be re-created on all devices. To achieve this you need to:

• In Service Studio, open the eSpace where the table is defined
• Select the table and copy it
• Delete the table
• Paste the previously copied table

This will cause a new key to be generated for the pasted table, which will create a new table on all the devices.

Attention that all local data in the table will be deleted on all devices.

Please let me know if this helps.

Hi All,

Old post, but I ran into the same error. In my application the error caused by a local entity with a identifier of <entity identifier>. The <entity> wasn't referenced in my module. 

Referencing the <entity> in module solved the issue.

Regards,

Matthieu