For users running DBaaS MongoDB products powered by Ops Manager automation, it would be useful if the API provided a way to replicate the cluster import behavior where MongoDB users are automatically discovered. This would help in scenarios where an existing Ops Manager deployment fails, and hundreds of deployments need to be moved to a different Ops Manager without knowing what MongoDB credentials the end user of that deployment had created. Having this API would allow the DBaaS provider's own scripting/automation to orchestrate moving these deployments to a new Ops Manager stack with reduced downtime.

Allow changing SNMP port from default (162) for customers who run SNMP on non-standard ports.

Currently the port number of the SNMP trap receiver is hardcoded to 162.

As a DBA using Ops Manager, I want to get an alert when a collection is a certain size or near no storage so that I can do an investigation to see if there is abnormal data growth. As of right now, I can only monitor cluster data size.

As a DBA using Ops Manager, I want to restore a single database within a MongoDB cluster and not store the all the databases in a MongoDB cluster so that my restore only impacts a part of my cluster. As of right now, it restores the entire cluster with all database in it.

What is the problem that needs to be solved? Ops Manager Automation does not take into account compound indexes (https://docs.mongodb.com/manual/core/index-compound/), e.g. if we have { a: 1, b: 1 } index already exist Ops Manager will still create { a: 1 } index for a Shard Key.

Why is it a problem? (the pain) This creates unnecessary indexes with performance impact on the MongoDB Server process.

Hi all

We would like to enable auditing for activities in Ops Manger itself, getting the information directly using the default audit logging options (filesystem / syslog).

Currently we are getting the detailed information in OpsManager, we get some information in the audit.log log when config changes are made via Ops Manager, but not in detail what exactly had been modified in the log files.

To establish a holistic, company wide auditing spanning over all levels (OS / DB / Application) taking use of standard auditing tools this would be highly beneficial.

Regards, Roger

Right now informational alerts can't go to PagerDuty, OpsGenie, or VictorOps. Ideally all alert notifications should go to the same endpoint, even if those alerts are "informational" and no acknowledgement is required.

When multiple clusters are added into an Ops Manager project, the only way to create a user account is to add it to all of the clusters in the project and to then add server IP Address restrictions to prevent the user from logging into servers other than the intended cluster. This would still create the user account on all of the clusters in the Ops Manager project.

Users would need a way to create a user account on a specific cluster rather than all of the clusters of the project.

Please provide a feature in OM so we can search globally a given HOSTNAME to know about its related Organization/Replica Set

Up until Ops Manager version 4.0+, we are/were able to view "Properties" of all servers/Replica sets by clicking on the ellipsis option on the each server. But will the latest versions of Ops Manager, that option is no longer available. We are often looking to quickly identify the allocated resources such as CPU/RAM on servers to help troubleshoot and deduct if we need to add more resources, so this has been a helpful feature we have used. We would very much like to see this feature option to be revived in the current versions of Ops Manager 4.2 and 4.4.4.

Our standard replica set deployments require a number of items to be set in the edit/modify area of each replica set...such as:
sslMode
directoryPerDB
directoryForIndexes
javascriptEnabled
redactClientLogData

Since we create hundreds of replica sets the amount of time spent repeatedly setting these attributes, which are considered "default" in our shop, adds up. Having the ability to set a template of defaults per project would save a great deal of time.

When Ops Manager is in "Local Mode", a system administrator must manually populate the mongodb-releases directory.

Please consider adding a GUI for uploading the releases over HTTP(s) from the Ops Manager admin panel. This will alleviate server access restrictions/issues, as well as provide a quick method of populating the releases from a workstation in an internet restricted environment.

We need to deploy ops manager with LDAP settings via yaml file where one of the setting "mms.ldap.bindPassword=" requires clear text password.
Is there any secure option to this?

This could be useful for testing SMS messages and twilio integrations.

WE have a 4 node replica set and when we removed 2 of the nodes at one time the election process didn't know how to elect a new primary. We would like the election process to accommodate an even number of replica set nodes.

Ability to enter multiple values for "KMIP Server Host" under KMIP Server Configuration in the Backup Configuration page of the Ops Manager Config. This would be similar to entering multiple values for 'kmipServerName' for deployments. This would help in terms of HA for KMIP; use a secondary KMIP server if the primary were to be unavailable.

Disable the dropdown to add Agent functions on hosts that don't have the MongoDB Agent installed and were only imported for Backup/Monitoring.
The elipsis (...) selections to add Monitoring & Backup Agent functions are available on hosts without the MongoDB Agent. If selected the function appears on the host as grey and standby. This is misleading. The options should not be available on hosts without the MongoDB Agent.

On Servers Page, Activate/Deactivate Monitoring/Backup module should be activated. So that multiple monitoring module should be activated for hot stand by.