Ops Tools
237 results found
-
rename action for backup daemon "delete daemon" to "deactivate daemon"
when you press link "delete daemon" on Admin->Backup->Daemons page it "deletes" daemon from this page, but after mongodb-mms.service gets restarted, this daemon comes up again. this link does not delete daemon - it actually deactivate backup daemon. another part for deactivation of daemon is to modify settings for service mongodb-mms to start only mms application without daemon.
1 vote -
Allow MongoDBUser CRD to interact with a MongoDB resource in a seperate namespace
Allow MongoDBUser CRD to interact with a MongoDB resource in a separate namespace in k8s.
The goal is to have different teams deploy their resources (including applications, configmaps, secrets, PVCs, MongoDBUser CRDs etc) into their seperate namespaces, but have all of them share a single MongoDB replicaset in a common namespace such as
mongodb2 votes -
Include full date/time stamps to OpsManager's logfiles: mms0-startup.log and daemon-startup.log
Currently records in
mms0-startup.loganddaemon-startup.logonly have time stamp in the format:hour:min:sec,milliseconds.
eg.08:28:09,558 |-INFO in ch.qos.logback.classic.LoggerContext[default] - Could NOT find resource [logback-test.xml]It would be useful to include the full date/time stamps to each record as currently we can't say to which day do the records refer to.
3 votes -
Allow multiple authentication sources simultaneously
Currently Ops Manager authentication supports either the Application Database, LDAP, or SAML, but these methods cannot be combined. Ideally we would like to move to LDAP, but we are stuck with the local authentication method as we depend on a local admin user which is used when first deploying and configuring the Ops Manager ecosystem. We also do not want to depend solely on the availability of the LDAP servers regarding an admin user. The MongoDB cluster deployments do support multiple authentication methods at the same time (we have local admin and monitoring accounts while users are authenticating via LDAP),…
10 votes -
Profiler and Performance Advisor without Automation
Currently it is required to enable automation on a cluster in order use the Profiler/Performance Advisor.
https://docs.opsmanager.mongodb.com/current/tutorial/profile-database/
https://docs.opsmanager.mongodb.com/current/tutorial/performance-advisor/
As automation is sometimes not desired for a cluster, please add the ability to Profile Databases and use Performance Advisor when a cluster is monitored only.
4 votes -
Add addition authentication mechanisms to the Ops Manager Alerting webhook
Capability to use JWT or SSL certs would be great.
7 votes -
Ops Manager to be able to do archiving of Data to a different replica set than you Prod replica set.
To be able to set up Archiving in Ops Manager so you would be able to keep a certain amount of data in Prod and move the rest of to an Archiving replica set.
1 vote -
Include CPU usage in Sharded Cluster Monitoring Overview
In the sharded cluster monitoring overview it would be useful to also have the option to display the Process CPU usage to quickly identify shards which have a higher CPU load. Other hardware metrics might be useful as well.
4 votes -
Loadbalance functionality for Ops Manager
When Ops Manager is under heavy load from many active database services, f.e. to handle backup snapshots, the recommendation from the case support portal is to use a loadbalancer in front of Ops Manager.
It would be great, if Ops Manager could have this as an integrated functionality to forward connections from agents to additional Ops Manager instances.
Background: Our nTSE did confirm, that we could set a different Ops Manager URL in the agent config, but the transferred data from this agent is still sent to the initially configured Ops Manager instance.5 votes -
OpsManager Application DataBase Backup and Restore
Ops Manager Application needs a way to backup and restore AppDB without Ops Manager Downtime to recover from infrastructure failures or fatal errors.
43 votes -
2 votes
-
Do not start backups for Sharded clusters unless all requirements are met
With FCV > 4.2, Backups requires the backup module to be activated on at least one member per shard for sharded clusters but it also doesn't stop one from starting backups if the backup module is activated only on one member of the entire sharded cluster, for example, activating the backup module on a single member of the config server replica set and not on any member of the shards allow for one to start backups.
Currently, if backups are enabled and at least one of the Replica sets in the Cluster has a backup module, backups will start for…
1 vote -
Improvement of Restore Process in Ops Manager
Presently Ops manager restore process removes all the data from the target MongoDb deployment and then restores from the user selected Snapshot and subsequently applies the PIT restore.
It is noted that time taken by the restore operations increases as and when the DB size increases. The restore operation would also require the Applications to be shutdown and this would lead to increase in Application downtime during the restore process.In order to minimize the downtime we need to reduce the restore time.
One suggestion to reduce the restore time is to perform automated restore from the latest available snapshots…
2 votes -
Present MongoDB clients metadata on Ops Manager UI
Before MongoDB version upgrades we should verify all drivers compatibility for the new version.
It is sometimes difficult to track all of the drivers used connecting to the deployments. Currently it requires parsing of multiple mongos/mongod logs for client connections or internal code investigations.
Suggestion
It may be beneficial to collect and present clients statistics (version, type, IP) data on Ops Manager.
Maybe this could be a feature that can be toggled on demand.
1 vote -
System Statistics without Automation
Currently it is required to enable automation on a cluster in order to monitor system level statistics such as CPU.
As automation is sometimes not desired for a cluster, please add the ability to gather system level metrics when a cluster is monitored only.
2 votes -
Ops Manager to support Credentials Cache to connect to APP DB
Ops Manager to support Credentials Cache to connect to APP DB
Its related to case opened : Case: 00881812Also change alerting ( Email Subject )
Currently we get alert with email subject as below even if issue is resolved
System Alert - Backing Database Connection is UnhealthyWe Should change alert email subject to Say "Healthy " instead of "Unhealthy " when issue is resolved.
1 vote -
Ability to migrate from Ops Manager to Cloud Manager
Clearly users of Ops Manager use it because they have a requirement, keep MongoDB clusters within their own infrastructure.
Well, we have two solutions for that Ops Manager and also Cloud Manager. I personally think if we have automation in the product to move from the Ops Manager control plane to the Cloud Manager control plane it would make it easier for users to take advantage of MongoDB's backup infrastructure and keep updates coming without the hassle of managing it all by themselves.
2 votes -
Ops Manager: add LDAP timeout options
Our LDAP server is quite slow, so we sometimes get technical errors when trying to log in our Ops Manager instance.
We would need add an option to set LDAP server response timeout on Ops Manager side. It could be
mms.ldap.timeoutMs.(I'm refering to Ops Manager authentication, not to MongoDB deployement authentication)
3 votes -
Ops Manager and attempts to stepDown
When the majority commit point is not reached, Ops manager should refrain from any config change process like stepDown of a particular primary.
Sometimes the number of attempts for stepDown can go upto 200 times and the duratio ncould be hours dependign on the data size.
More details are present in the below ticket:
https://support.mongodb.com/case/008851731 vote -
Show time it took to complete a snapshot
In order to better troubleshoot snapshot performance issues, or maintain an understanding of how snapshots are performing and whether one is approaching a threshold where they wont be able to keep up, its a good idea to expose how long a snapshot took in the View All Snapshots page. This would also aid in quickly identifying which snapshots are full vs incremental.
46 votes
- Don't see your idea?