when creating a private link, the resulted dns for the cluster is in the form of "cluster-pl-.{somerandomstring}.mongodb.net"
the only difference from the public endpoint is the 'pl' addition in the first section.

in order to configure the vpn wildcards and rules,
it can be extremely helpful to have this separation more general, for example
cluster.{randomstring}.pl.mongodb.net -- private links
cluster.{randomstring}.mongodb.net -- public

that way we can use the wild card *.pl.mongodb.net to differentiate and route the traffic accordingly

allow multiple Atlas CIDR (VPC Region) in the same project and allow attaching each cluster to the desired vpc.

today in order to have multiple vpcs in atlas (for example dev &prod)
they need to be separated projects, that create a lot of duplicated work and settings

In addition to AWS, Azure and GC, give an option to host on DigitalOcean. DigitalOcean has a really cheap and well-made Kubernetes system that requires VPC because container IPs are always changing, and I would want them to connect to Atlas without having to just open Atlas to whitelist every IP.

When looking at the cluster page, I can see Disk Usage = 1.6GB, for example, but I don't know what the breakdown of that is. For instance, if I look at the same DB and add up the size of the collections themselves it's barely 1.5MB. So 1.6 Gig disk = 1.5 Meg data. I suspect this is oplog etc. but it is a pretty confusing potential red herring when diagnosing performance issues etc.

Being unable to lodge a support ticket without paying for a support plan - when we are paying for a premium service is seriously f'ed up

Please include eviction metrics as well in monitored metrics in atlas so that it will provide an overview how many evictions are happening in particular time and there by decide correct memory for the instance.

When adding or editing a custom role in Atlas - When I expand/collapse the Actions or Roles in dropdown, it also toggles the checkboxes. So, when I want to see the roles and actions available, it also checks/unchecks the checkbox, which makes editing the actions in roles difficult.

Please click on highlighted areas in Atlas, as shown in screenshot attached

Thanks.

There is a gap in the Cluster Tiers for clients that have high CPU requirements but lower disk space requirements. We run on a M30 Cluster where we max out the CPU when running aggregations, we use little of the available disk and do not want to move to a M40.

In the email for alerts of type "Users do not have two-factor authentication enabled" it would be helpful to see the user id of the “violator” (user which does not have two-factor authentication enabled), so that I don’t even have to click and log in to Atlas.

Maybe this could be make optional, if there are privacy/security concerns about sending around the user ID in cleartext.

The delete button with low contrast is the one that appears after you click the delete icon on for a document on the collections screen.
The button text is white on a pink background.
See attached file.
It is hard to see the button.

Expose oauth apps to let other services integrate with Atlas, this way other platform can integrate with atlas for provisioning a mongodb instance.

It would allow for a lot of opportunities for developers like automatic mongodb provisioning for a deployment platform

Right now if you deploy with platforms like Zeit or Heroku you have to create a mongodb database and whitelist the right iP manually

I'm trying to setup CI for my project using GitHub actions, while restricting Network Access. The problem is that GitHub's machines have weekly changing IPs: https://help.github.com/en/actions/reference/virtual-environments-for-github-hosted-runners#ip-addresses-of-runners-on-github-hosted-machines

Is there any way to automatically integrate that list of IPs in the Network Access IP Whitelist? This feature would benefit all users of GitHub Actions.

The Atlas activity feed is currently searchable by Type of event. It should have an option to search by cluster name to quickly narrow down any particular activities relating to a particular cluster

As we move into a more stable version of testing the 4.4 beta it would be useful to allow beta-testers to spin up a cluster with the beta version of MongoDB. Right now beta testers have to download MongoDB which means for some of the more exotic server features in 4.4 they have to build a sharded cluster or replica set. This would be much easier to do in Atlas.

Restore snapshots to different clusters.

I have a staging environment cluster whose data I'd like updated daily with anything new from my production cluster.

Currently I do a dump and restore, but this is slow because it gets piped externally.

Ideally I'd be able to restore a production snapshot to my staging cluster through a script running on a cron. Whether than is API or CLI doesn't matter.

Please look at the images, try to make similar UI, it would improve a lot the productivity

we should be able to plug in the already developed algorithms say from GitHub or any other repository

Stitch CLI error messages are obsolete.

I'm having difficult with connecting to a Stitch app and the 403 forbidden returns a message to check https://cloud.mongodb.com/v2#/account/publicApi but it redirects to /overview.