Currently, getting updates to custom user data on the frontend requires periodically refreshing the token. A potential improvement would be for MongoDB Realm/SDKs to detect these changes automatically and streamline the token refresh so application code wasn't required to handle this case.

Logs can only be filtered by type, user and timestamps. Would be very useful to filter by function name

I would like to have a tutorial/tool explaining/enabling how to work with different environments (DEV, TEST, PROD...), so that I can set an automated pipeline for CD.

Similar to AWS SAM (https://docs.aws.amazon.com/serverless-application-model/latest/developerguide/serverless-sam-cli-using-invoke.html) or Google Cloud Functions Framework (https://cloud.google.com/functions/docs/functions-framework).

I.e. the ability to run functions locally. It's expensive and slow to have to run and test code in a project remotely all the time even for small changes.

Hi, I am an existing Realm/Realm Cloud user and am struggling to understand what the MongoDB Realm offering is and how it compares with the existing Realm/Realm Cloud solution.

Is there any documentation that explains what the difference is between the two solutions and what if any migration is required.

It's not clear whether the latest cocoa Realm database v5.1.0 is still compatible with Realm Cloud nor is it clear whether there is any plan to continue supporting the current Realm Cloud solution. A nice clear description would be much appreciated.

Regards
Duncan

Basic password reset email customization would be greatly appreciated when using the UserPassword auth provider. Short of fully customized emails, logo upload and basic verbiage editing would be an improvement.

Currently it is only possible to link a Github account for deployment.
It would be great if we can use Bitbucket as well. I have my code in a private bitbucket repository and would love to have the ability to deploy from it.

We need the following enhancements
1. Need more than 80 characters
2. Need ability to turn it on and off by level (Log4J)
3. Need ability to turn to higher level of logging on a single call

When deploying from a Github commit, can you make the Realm files only update what is being committed? It seems like Realm is updating all files, which is taking forever when you have a bunch of functions, along with Hosting files, which in my case is a good bit.

If I go directly into the UI and update a function from there, the Deploy happens fairly quick. But, as soon as I commit from Github, that function gets updated by what is in my Github Repo, which in this case, would overwrire what I just did in the UI.

Currently deployent is only available to AWS but Atlas allows deployment in GCP.

We need to have deployment close to GCP for better performance consideration.

It will be great if Tests could be defined within the UI.

For example having the option to write a jest test that is executed before deploying any changes to the Hosted app would be awesome.

Need to be able to specify custom read preferences (e.g. tags) for Realm data sources, as I may want to segregate the load from my Realm apps with other operational load.

Would be nice to have this feature as a checkbox on each collection or default for the entire database and it should be in utc or a checkbox for utc.

Currently, you only support Google and Facebook for social logins. Can Twitter also be added?

It would be nice to use webhook path to pass payload to the webhook for example:
/api/mypath/{deviceId}/data

Instead of
/api/mypath/data?deviceId={deviceId}

I want to port my application but I can't port all my old users.

I've moved from Firebase because of the lack of support for React-Native and I've just discovered that I need to use Firebase to receive Push Notifications from my App. I'm aware of the complexity of adding this layer to SDKs, but you're already much ahead of Firebase in terms of SDK maintenance and support, being able to make a full app's stack would be wonderful for many, I think.

We would like to push secrets via the Admin API or CLI.

It would be great to add security to requests by being able to respect the relevant Atlas project peering and IP Whitelisting to reach Functions.