Atlas
- A brief description of what you are looking to do
- How you think this will help
- Why this matters to you
13 results found
-
Manage Database by Kubernetes Operator
In our use case, we are dynamically creating feature environments based on the branch name. During this process, we are also creating MongoDB databases in the same Atlas cluster (different envs are sharing the same cluster but with different database names). For now, we need to do it with a separate CI step that should create this DB and delete it when the env is deleted.
It would be great to have the possibility to manage DB create/delete process with Kubernetes CRD. For now it's the only 1 missing feature that is blocking us to start using Operator.
5 votesThis is something we're considering for the future.
But the biggest problem we need to solve is that it's incredibly easy to create/delete/update databases within a deployment via many other interfaces. But if this happens, the Operator's source of truth (the custom resources) won't contain the changes, and the Operator would overwrite the changes using it's source of truth.
-
Use kubernetes Atlas operator to create and manage only users
We are already managing the Atlas project using terraform and do not want to have two separate tools managing the same resource. We would still like to generate atlas users automatically through the operator. If there was a method of just providing the project ID as a string like we are able to in the terraform resource definition, this would not be a problem.
5 votesWe're planning an investigation into this in Q2 (May-July for us). It's something more than one customer has asked for and allowing an Atlas ID to be used rather than a reference for another K8s resource feels like a reasonable approach.
No commitments for now till we've investigated in a little more detail.
-
Reconciliation resources flag improvement
atlas-reconciliation-policy=skip flag provide ability do not sync k8s object status with atlas. But in case when we need to make reference on objects in Atlas (Aka project) and do not create or modify this resource - this flag not very useful nothing. In this case i would be good if
- or operator sync object from API to k8s resources ( grub id by name so k8s resource will be valid )
- or at least provide opportunity to set id ( for new resources with atlas-reconciliation-policy=skip set by default )
Aka:- Follow the pattern of reading data from the…
5 votes -
Autogenerate a secret for AtlasDatabaseUser if none provided
I'd like for the controller to make a random password at the point of submitting the request, the password itself is not relevant to me as it'll be included in the returned connection string to be consumed by the applications themselves.
More details found on the Github Issue: https://github.com/mongodb/mongodb-atlas-kubernetes/issues/1370
3 votes -
fix example at https://www.mongodb.com/docs/atlas/operator/current/atlascustomrole-custom-resource/#basic-example
in the example at spec:
- "name" indentation is wrong
- also action should be specified as in API example connPoolStats should be actually "CONNPOOLSTATS" (with underscores, seems they disappear here when I post)
- at inheritedRoles there is no "role" as child field, should be databasea correct manifest would be:
apiVersion: atlas.mongodb.com/v1
kind: AtlasCustomRole
metadata:
name: tester
namespace: tester
labels: {}
annotations:
mongodb.com/atlas-resource-policy: keep
spec:
projectRef:
name: tester
namespace: tester
role:
name: tester
actions:
- name: CONNPOOLSTATS
resources:
- cluster: true
database: tester-database
collection: tester-collection
inheritedRoles:
- name: operator-role-1
database: tester-database2 votes -
Create an AtlasRole custom resource
Currently roles are configured as a list inside the AtlasProject resource. I propose managing these as individual AtlasRole custom resources instead.
Consider a typical microservice based application with several microservices deployed by individual Helm charts, each sharing the same project and cluster. In this scenario the app developer may want to create custom roles for each app / group of apps with access to certain collections, eg to stop them from being able to read collections with sensitive data
Currently it is possible for apps to create their own users, but those users must assume one or more existing roles.…
2 votes -
Reference a db user from namespace A to a project in a different namespace
We need to ability to reference db user in namespace A to an Atlas project in a different namespace.
2 votes -
Set project limits in AtlasProject CRD
It would be great to be able to set the project limits (API reference: https://www.mongodb.com/docs/atlas/reference/api-resources-spec/v2/#tag/Projects/operation/setProjectLimit) in the Atlas Kubernetes Operator through the AtlasProject CRD.
Specific use case for us is the ability to set "atlas.project.deployment.privateServiceConnectionsPerRegionGroup" to tweak the number of PSC connections. Now we have to circumvent the operator to be able to tweak this value.
1 vote -
Define the secret name for the connection string output from the operator
Hello all, instead of using a predictable naming convention for secrets:
$project-$cluster-$username
I'd like to be able to nominate the output secret name similar to how crossplane does this.
More details found at the Github Issue: https://github.com/mongodb/mongodb-atlas-kubernetes/issues/1369
1 vote -
AtlasProject CRD status fields
Hi, we have created the Atlas project via Terraform and used the mongodb operator for cluster and user management, but for now, we use sidecar to set the status.id field in AtlasProject CRD since we use skip-sync flag ( so operator will not change project config) and the operator will not sync this field for us.
If it is possible add a parameter to the spec where I could set this ID within resource creation ( so if we have a skip flag it will automatically set status.id ), so we can get rid of sidecar ( sidecar was implemented…1 vote -
Kubernetes AtlasDeployment snapshot persistence flag
When AtlasDeployment resources are deleted, the cluster is immediately terminated without the opportunity to enable snapshot persistence after deletion as there is in the Atlas UI. An option in the spec should be added to opt-into this behavior from a Kubernetes perspective.
1 vote -
Kubernetes AtlasProject backup compliance
Extending the AtlasProject Kubernetes resource to allow management of a backup compliance policy, including setting project-wide backup schedules and enabling backup persistence after termination by default, would be a major improvement over the current experience.
1 voteIt will be coming in a release in the next few weeks!
-
Kubernetes reconcilation interval
Kubernetes resources generally should be reconciled on a sensible schedule to prevent configuration drift between k8s and Atlas API.
AtlasProject is a critical one that can have widespread impact if API configuration drifts from Kubernetes and is not promptly corrected.
1 vote
- Don't see your idea?