Azure Cosmos DB
Have feedback for Azure Cosmos DB product? Submit your idea here or upvote other ideas. All feedback is monitored and reviewed by the Azure Cosmos DB team.
This site is for feature suggestions only. For technical questions or issues, please submit them to StackOverflow,where we and the community can better help you.
Please use the following categories when submitting your idea.
Gremlin API: Graph features and capabilities using Gremlin API.
MongoDB API: Features and capabilities using MongoDB API.
Table API: Features and capabilities using Table API.
Etcd API: Features related to using etcd as a configuration store for Kubernetes.
Built-in Notebooks: Features related to built-in Notebooks in Azure Cosmos DB.
SDK: Features related to Azure Cosmos DB SDKs for SQL API.
Change Feed: Features related to Change Feed.
Management: All management features, backup/restore,monitoring, ARM, PowerShell and CLI.
Portal: All features for Azure Portal and Cosmos DB Explorer.
Emulator: Features related to the Azure Cosmos Emulator.
Other: Features not related to any other category.
-
[CosmosDB] Be able to manage Backup and Restore with Portal and APIs
Please provide the ability to manage backup and restore on our own.
Restore through support is not clear and inconvenient.
It turned out restore is only available through paid support plan.
This document https://docs.microsoft.com/en-us/azure/cosmos-db/online-backup-and-restore#restoring-a-database-from-an-online-backup does not say it in any word.
The situation made our customer extremely unhappy.
730 votesThis feature is now available in Preview.
You can read more here.
https://docs.microsoft.com/en-us/azure/cosmos-db/continuous-backup-restore-introductionThanks.
-
Add ability to rename databases and collections
It should be possible to rename databases and collections.
This feature is universal in all if not the vast majority of database products.
Lack of ability to rename anything makes it very hard for us to manage releases if we are adding new data. Currently, we have to create a copy of the existing database (which is a feat in itself, since this again is not core functionality), make the change there, and then re-point the website to the updated database. All this time, we are incurring extra costs and inconvenience.
Renaming things also allows us to correct typos, which…
728 votesThe ability to rename databases and collections is not currently something that is in our road map.
Currently the way to accomplish this task in Cosmos DB is to create a new collection with the changed name then use bulk exec or change feed to populate it.
Thanks for your request and input on this. We may consider this for a future planning cycle.
thanks.
-
Improve scaling down experience: remove redundant physical partitions
During initial import we increased RU to 200K. Cosmos Db SQL API created 20 physical partitions with 10K throughput each.
During regular day we just need 30K Ru. That would mean each physical partition would have only 1,5K RU allocated and a lot of redundant partitions with possible expensive cross-partition calls (total data size is less than 30GB)Is there way to decrease amount of physical partitions to reasonable value?
In a nutshell it is not fair pricing, when you are limited to 1,5K RU and a lot of cross-partition calls you have to pay extra294 votesWork for this feature is now being planned. Will update when work on this feature starts and when it becomes generally available.
Thank you.
-
Set Account Level throughput on Cosmos DB Table API with ARM template
Make it possible to set "Account level throughput" on a Table API account through ARM template or Powershell.
According to the official Twitter handle of Azure Cosmos DB, it is only possible to set the "Account level throughput" on a Table API database account through the Azure Portal. You cannot do this with an ARM template.
https://twitter.com/AzureCosmosDB/status/117507143322931200176 votesChanging this back to under review.
We took a look at this earlier this Spring but as we got deeper we realized this causes issues with conforming to our resource provider schema.
Within our RP, each API type has a database-level resource and one or more container-level resources. Each of these has a child throughput resource which is how you provision either shared throughput or dedicated throughput.
Table API is different in that there is no database-level resource. It’s simply the account and multiple tables. databaseAccount has no notion of throughput in our RP. We are not able to associate the throughput resource to the databaseAccount resource as this would not conform to the account resource definition.
All that said we are looking for a way to support this. Please bear with us as we work out options. Will provide further updates as this progresses.
Thanks.
-
Make autoscale work like other services' autoscale and pricing
Autopilot in general cost 50% more per 100 RU than the dedicated throughput and is cost effective if there is significant variation from peek load usage. This is confusing as customer that auto pilot could be costing more in some sceanrio like if you only lower loads on weekends etc. And morever this is different from all the other services available in Azure where the Auto Scale Up/down does not have a different pricing then the regular mode. this cost should not vary with provisioned throughput ...
22 votesUpdate. We will move this to unplanned as we may revisit this and offer an autoscale similar to what is described.
Will change the title to better reflect the ask however.
-
Faster resource deletion
CosmosDB resources take a very long time to delete. Make this faster please!
19 votesWe have reviewed this and have added to to our committed work items for this semester.
One thing to point out however is that resource deletion for Cosmos will never be super fast. The reason is that we need to ensure that all connections and caches are drained before we delete the resource. This can some minutes for this to happen.
We will update here when we release this work.
Thank you.
-
Default Firewall Config for Restored Cosmos DB Account
Default FW config for a restored cosmos db account has “Allow All Networks Enabled”. With this option enabled if a user has the connection string then the account can be accessed from ALL networks ie outside of company proxies. I would like to see the default option for a restored account to be “Selected networks” with nothing else in the FW config. This will at the very least restrict any off-network access to the account.
18 votes -
We need to view the last backup date & time and a flag to indicate if it has been backed up successfully.
We need to view the last backup date & time and a flag to indicate if it has been backed up successfully. This is for resource governance and data availability purposes. Management needs to view governance dashboards and verify backups are successfully completing as described in the documentation.
15 votesThis is currently planned with point in time restore feature we are currently working on. When this is released as GA customers will have full control via Azure Portal.
Thanks.
-
Reorganization/Optimization for data in physical partition/index pages
Request feature for Reorganization/Optimization for data in physical partition/index pages.
There are 2 environments (staging/production) which have almost same data/volume.
But when a query runs against specific partition key, staging is faster but production is much slower.
Hence requesting there should be Reorganization/Optimization feature to be stable performance among all environments like RDBMS has it.12 votesThis work is currently planned. Will update as this progresses.
-
Update documents to add notes time to be effective after consistency level change
Request to update documents adding notes time to be effective after consistency level change.
This request is because we faced a lot of timeout on the application by taking long time for process on app after consistency level change. We surely confirmed the change is done on Azure portal and the application restarted.
But the issue persisted until application restarted again.So that should be clearly described on the doc.
4 votes -
RBAC - Firewall & Replicate Data Globally
Hi MS,
We are using custom RBAC Roles in Azure.
I do not wish to provide :- Firewall and
- Replicate data global
to my clients to change.
Currently the clients are able to add any ip they like despite of having privateendpoint on the cosmos db account.
4 votes -
Availability to deploy custom TYPE with ARM template for Cassandra API
Please consider to add availability to deploy Cosmos DB with custom TYPE for Cassandra API TABLEs using ARM template.
4 votesThank you for your suggestion.
Are you referring to custom data types here?
https://cassandra.apache.org/doc/latest/cql/types.html#custom-types
Or are you referring to user-defined types here?
https://cassandra.apache.org/doc/latest/cql/types.html#udts
Thanks.
-
Backup & Restore of cosmosDB to the existing collection itself.
Hi Team,
Recently in past, I have raised a request to restore a collection with some specified time stamp but the restored data should over write the existing collection itself.
Azure Support advised that this is not supported in Azure Cosmos db and this is by design. I would highly recommend to support this feature for following reason:
1. Azure Support team are supporting back & restore processing with the help of technical support.
2. Restoring it to other place and migrating to new location or existing location kills a lot of time in copying data from one place to…4 votesTo do this we need to be able to rename resources which currently today we do not support.
We will eventually look to support this but it is not on our current plan.
Will mark as unplanned now and update when this changes.
Thanks.
-
Custom role operation for Cosmos DB
Add operations to read, write and delete of firewall rule for Cosmos DB.
I could find some other services like SQL, MariaDB and PostgreSQL are having those kinds of operations,
so I was wondering why Cosmos DB doesn’t.4 votesThanks for your suggestion. This is not currently on our road map. Will keep on backlog and revisit in future planning cycles.
Thanks.
-
Request for Incremental backup
Request for Incremental backup;
Currently only full backup is supported but would like incremental backup .
I think it is possible technically as Cosmos DB has _ts. That feature would help users to reduce additional charge.3 votesWe don’t offer incremental backup and have no plans to build it. We do however now offer continuous backup with point in time restore.
Maybe this is what you are looking for?
https://docs.microsoft.com/en-us/azure/cosmos-db/continuous-backup-restore-introduction -
Allow Cosmos DB Offer Type to be Migrated using ARM template
We use ARM template to automate throughput scaling for our Tables. Currently, if we want to swap from manual scaling to autoscaling, we get a failure message saying we must use the migration API.
This means that we cannot use ARM template to change throughputs quickly throughout our many existing instances of Tables, and must either make the change manually or write a separate script to make this change.
3 votesThanks for your suggestion.
The throughput migration API call is a PUT operation which is not supported by ARM. It is this reason why we cannot offer this via ARM template and it must be called using PS, CLI or Azure Management SDK.
The other thing to note as well is even if this were supported, the throughput value is determined by the system, it cannot be set when doing the migration. This also works against use of ARM templates.
Thanks.
-
Notify only when ip rule changes are in effect.
As of today, when I add IP rule in cosmosdb account firewall so that I get access to container documents, it goes down for 15 minutes as mentioned in public documentation. However, the notification for firewall rule update shows success in few seconds only. It would be better if notification shows success after 15 minutes when actually IP rule is in effect.
1 vote -
Allow removal of Free-tier status on database in production use.
Was inforrmed by support that our production database, which has hit it's limit of 5 shared throughput databases for a free-tier enabled account, can not be updated to a non -free tier. Subsequenly they suggested manually copying the data (all 80+ containers) over to a new account using DataFactory.
Note that the DB was not intentionally setup as a free-tier account, but since this was a realtively new option on the Azure Portal, and the default is to enable the free-tier option our Cloud Ops specialist erroneously left it enabled.
The limit of 5 shared databses also isn't immediatley obvious…1 vote -
Notify only when ip rule changes are in effect.
As of today, when I add IP rule in cosmosdb account firewall so that I get access to container documents, it goes down for 15 minutes as mentioned in public documentation. However, the notification for firewall rule update shows success in few seconds only. It would be better if notification shows success after 15 minutes when actually IP rule is in effect.
0 votes
- Don't see your idea?