Upload good tutorials in YouTube about connecting mongodb atlas with available three options.

on MongoAtlas metrics dashboard, I can organize graphically different metrics as I want to display first "insert" metric for example, then "cpu" metric, etc ...
If I change this actual order, then I go to another screen (for example "network access") and finally I come back to this metric dashboard, my new metrics added (insert, cpu, ...) in first position are not in wanted first position (at the top of metrics list), but at the end of this metrics list.

It would be great to keep metrics context (displayed metrics and ordering) on Metrics dashboard.

Mongorestore allows for the number of concurrent insertion workers per collection to be increased using the [--numInsertionWorkersPerCollection](https://docs.mongodb.com/database-tools/mongorestore/#std-option-mongorestore.--numInsertionWorkersPerCollection) option.

However, this is not available for mongomirror -- while you can increase the number of collections copied in parallel ([--numParallelCollections](https://docs.atlas.mongodb.com/reference/mongomirror/#std-option-mongomirror.--numParallelCollections)), there's no parameter surfaced to allow more inserts to be occurring at the same time over multiple threads within each collection.

|t would be helpful to have a parameter like this to use to increase parallelization in order to increase the speed of migrations of large collections.

Please include the human-friendly cluster/host names in Monitoring Alert Notifications (e.g. email, PagerDuty, etc) in addition to, or instead of, the less clear internal cluster/hostnames (i.e. atlas-abcdef-shard-0).

Currently is not possible to enforce any kind of password policy via Atlas UI.
Would be nice to let user define an expiration time and password policy that user should meet.

Test environment are usually paused during the night and the weekends. Currently the only way to do that is to automatise a RestAPI call or call mongocli automatically (e.g. via crontab)
Would be nice to allow to define a time windows on which the cluster stays pause via UI

The BSON spec includes support for UUID values, also known as GUID values.

The format of UUID values are defined in rfc4122.txt. For example,

f81d4fae-7dec-11d0-a765-00a0c91e6bf6

Users would like to work with UUID/GUID values in Data Explorer as a fully supported data type, including search support in the RFC layout, ability to create new document fields of type UUID, support for UUID type _id fields and the ability to edit UUID values in existing documents.

The mongo shell documentation provides this example:

UUID("3b241101-e2bb-4255-8caf-4136c566a962")

Customers often want centralized permissioning for the database and LDAP is not generally supported well in the cloud. If a customer is using Azure, they want us to support Azure AD for database authentication. I believe that means using service principals

I access Mongo Cloud / Atlas very frequently. However, almost every day I need to re-login using my Google account. It doesn't seem to "remember" my session.
I would love if I can extend the default, and let me (and other members) stay logged in for a longer period of time.

When indexes are built from an application or a mongorestore there's no way to see if it's ongoing in the UI.

There should be an indication in the "Real-time" tab saying that an index on collection X is in progress. This would explain why performance is currently impacted.

It would be good to see index start and end marked on the "Metrics" visualizations so we can see the impact of index builds on IOPS, CPU, memory etc. This could help understand if we need to upscale the cluster.

Please consider adding 2FA support for hardware keys (Yubikey). Our company started enforcing this in our compliance policy for accessing production environments.

Similar to how AWS includes tags/labels in their billing dashboard, I'd like the ability for my cluster labels to be included as a column in my billing usage details (or even the exported CSV).

This will allow me to categorize my billing usage as granular as clusters in addition to the current method which is by project.

Currently there is no way to test how applications behave during maintenance apart from waiting for maintenance, the ability to force a rolling cluster restart as happens during maintenance would be great. The failover is a different process and affects primary mainly.

We have a need to rename our databases as we sync them over to Atlas. However, mongomirror doesn't support that. So that leaves us with mongodump and mongorestore but our databases are large enough that we can't take the outage window necessary to go down that path. When will mongomirror support renaming databases during the import process?

Currently the Query Profiler can plot queries that were logged up to 24 hours in the past.

It would be helpful to allow for visualization of a custom date range older than 24 hours ago, rather than only queries logged within the past 24 hours. This could help with RCAs for events that occurred more than 1 day ago, and also help teams who collaborate to investigate queries over a time period longer than 1 day.