With insertMany, we are able to insert a large number of documents within a single operation. However, when doing this with an upsert, we have to perform separate updateOne operations.

If we are able to provide an array of filter-updateOperation pairs, it would become possible to perform multiple upserts in one operation.

Currently, with the JSON Schema validation, we are able to limit the values for a field using enumerations. However, we need to have a way to limit the values to those entered in another collection.

We have multiple microservices using the same collection and we'd like to track how many calls come from each, and also understand which transactions are successful or not.
Ultimately this will allow us to understand the MDB usage by each of the microservices.
[setting verbose logging level to 1 and then using these logs to create a dashboard in Splunk doesn't give us enough granularity]

At Airbus, we would like to use geoIntersects feature with 2D Index.
We don't need 2DSphere Index, but we have to use it if we want to have geoIntersects avalaible.
As a consequence, we have to manage Spherical margins which introduces a useless and hard to maintain workaround in our product.
Could you please plan to implement geoIntersects and not only geoWithin with 2D Index ?
Thank you

I would like to be able to find all geoJSON points located within a collection of polygons. My personal goal is to find points located within earth's water bodies. The water bodies are located in their own collection while the geoJSON points are located in a separate collection within the same DB. Mongo's geoWithin function only lets me hard code the polygons, but I would like to reference a collection of polygons.

See https://stackoverflow.com/questions/63162823/mongo-geowithin-error-polygon-coordinates-must-be-an-array

It would be great to have a function where we can , using only atlas interface to filter a month and a year , and we cann see historical data about sizes.

I use Atlas because it's easy to set up, especially with backups. Without this feature, I have to build sync tools with the API to make sure my backups are secure. If someone got into our database and deleted the cluster - the backups are completely lost.

Changes done through Ops Manager portal are visible only in Alerts(Activity Feeds). Though these Alerts(Activity Feeds) can be derived through API, our auditors may not accept API calls and we need ops manger to log these change in respective deployment mongo audit.log

Thank you

Provide Kafka Topic as a write target for database auditing and database message logging.
https://docs.mongodb.com/manual/core/auditing/
Auditing is currently limited to a local and editable JSON/BSON file or the system console log.

The SYSLOG is not recommended by MongoDB. "The syslog message limit can result in the truncation of the audit messages. The auditing system will neither detect the truncation nor error upon its occurrence."

MongoDB can already suggest useful indexes. Why not take the next step and allow MongoDB to autonomously create and manage indexes. Ideally it will automatically maintain the indexes over time as the structure and usage of the database changes.

Recently we had a compact operation fail, and the mongod logs stated this was due to "eviction pressure" on the server. We have learned that this is a failsafe mechanism built into the compact command, but we would like to have a more verbose explanation of this failure mode in the logs to better understand why it failed and steps to remediate the issue.

The MongoDB engine should allow for hidden members in a replica set to be automatically elevated to a voting member. When quorum is lost and can't be met, the hidden replica is automatically reconfigured to a voting member to obtain quorum.
When the failed voting member comes back online and catches up the hidden replica is demoted back to a normal hidden member.

we've already seen a full text search and it will be awesome if you manage to implement partial version :)

From a user - A MongoDB in-memory emulator would help development against MongoDB in a completely isolated way. They deploy self contained applications for developers that don't have any external connections other than MongoDB. The customer found a community project, fongo, that does this, but it is not supported and trails development by a few versions. https://github.com/fakemongo/fongo. I would be happy to connect the users with relevant project managers