Basically, on almost any ajax request (and this doesn't just apply to Carbon) it is a good idea to setup a retry system in case something goes wrong. In this case, that request can be retried and it will eventually succeed.
I would advise you to check the error and just retry the request
Instead of trying to see if the document is locked
Are you waiting for the promise to finish, or are you sending both of them at the same time?
I'm using vue, I write this code on the app.vue and I'm using the hot reload, any time that some change is detected the app is reloaded..
CarbonService.createDocument: function(carbon, rootPath, document, slug ){
return carbon.documents.createChild( "/"+rootPath, document, slug );
}
let carbon = CarbonService.init();
let something = {};
let slug = "users";
CarbonService.createDocument(carbon, "/", something, slug).then(
( [ persistedProject, response ] ) => {
console.log( something === persistedProject ); // true
console.log( something.id ); // document's URI
}
).catch( console.error );I run the script to create a document and I wait to see on the browser console the log of true and document URI, after that I'm going to my code again and change the slug variable to this:
CarbonService.createDocument: function(carbon, rootPath, document, slug ){
return carbon.documents.createChild( "/"+rootPath, document, slug );
}
let carbon = CarbonService.init();
let something = {};
let slug = "roles";
CarbonService.createDocument(carbon, "/", something, slug).then(
( [ persistedProject, response ] ) => {
console.log( something === persistedProject ); // true
console.log( something.id ); // document's URI
}
).catch( console.error );the app is reloaded and run the script again, and is there where I see the 409 error on the browser console.
There' s a bug in the SDK that is triggering this behavior
In your code, the requests are being sent to
//
Because of this:
"/"+rootPath
Can you open your developer tools and send us a screenshot of the requests being sent?
ammm I thought about other things that I did yesterday, and another thing that I did was delete a document on the workbench and after that, I begin to see the error logs, and I understand that is because the parent document was "locked". probably this was the cause of the error. but how I can know if the parent document is locked, what happens if some request is sent to create a child in the same parent document? Will I re-send the request until the document can be created?
So if you perform an action, that action finishes, and the document is still locked, there's an issue with the platform
Even then the lock expires after a few minutes
Right now there isn't a direct way to check if a document is locked other than trying something and receiving a 409
It is an internal mechanism after all
Normally you would program a retry mechanism with a configurable delay and a number of retries before failing
For now, you can remove any dead locks your platform may have by following these steps:
- Shut down your platform:
docker stop <container-name> - Delete the directory:
<carbon-directory>/data/zookeeper - Start your platform again:
docker start <container-name
Like I said, this shouldn't happen, so please let us know if a lock isn't being released and the steps you took to get there
Thanks for reaching out :smile:
Is that the right/shortest path?
We think the export/import process would not be compatible between legacy versions and the newest Alpha, right?
There's no migration path yet
Manually importing/exporting it is the only approach as of now
