-
Notifications
You must be signed in to change notification settings - Fork 15
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
docs: Add NFS feature description #918
base: main
Are you sure you want to change the base?
Changes from 2 commits
c976af5
323d882
f625f40
95ebfe5
08694c2
ad3eaf5
b8da0e4
5e74609
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,2 +1,61 @@ | ||
# NFS | ||
|
||
## Overview | ||
|
||
The Cloud Manager module provides a Network File System (NFS) feature which allows you to create ReadWriteMany (RWX) volumes. RWX volumes can be used by multiple workloads. | ||
|
||
When you create an NFS instance in Kyma, you depend on the cloud provider of your Kyma cluster. The cloud provider in use determines the exact implementation. | ||
|
||
Cloud Manager module supports the NFS feature of two cloud providers: | ||
|
||
* Amazon Web Services [Amazon Elastic File System](https://aws.amazon.com/efs/) | ||
* Google Cloud's [Filestore](https://cloud.google.com/filestore?hl=en) | ||
|
||
You can configure Cloud Manager's NFS instances using a dedicated NFS custom resource corresponding with the cloud provider for your Kyma cluster, namely AwsNfsVolume CR or GcpNfsVolume CR. | ||
|
||
## Prerequisites | ||
|
||
To instantiate NFS, an IpRange CR must exist in the Kyma cluster. IpRange defines network address space reserved for your cloud provider's NFS resources. If you don't create the IpRange CR manually, Cloud Manager creates a default IpRange CR with the default address space and Classless Inter-Domain Routing (CIDR) selected. For more information, see [IpRange Custom Resoucre](./resources/04-10-iprange.md). | ||
|
||
## Lifecycle | ||
|
||
AwsNfsVolume and GcpNfsVolume are namespace-level CRs. Once you create a GcpNfsVolume or AwsNfsVolume resource, the following are also created automatically: | ||
|
||
* IpRange CR | ||
* IpRange is a cluster-level CR. | ||
* Only one IpRange CR can exist per cluster. | ||
* If you don't want the default IpRange to be used, create one manually. | ||
* Persistent Volume (PV) | ||
* PV is a cluster-level piece of storage. | ||
* The PV uses a unique name based on the NFS instance status ID. | ||
* The PV references the instantiated NFS in the respective cloud provider. | ||
* Persistent Volume Claim (PVC) | ||
* PVC is a namespace-level resource created in the same namespace as the IpRange CR. | ||
* By default, the PVC's name is the same as the name of your NFS instance unless you specify a different name. | ||
* The PVC references the automatically created PV. | ||
|
||
## Cloud Providers' Specifics | ||
|
||
### Amazon Web Services Kyma Cluster | ||
|
||
Kyma clusters provisioned on Amazon Web Services use Amazon Elastic File System. See the following list of CRs available in the Cloud Manager module dedicated to Amazon Elastic File System: | ||
|
||
* AwsNfsVolume CR | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. All of these are CRs. Do we need to mention CR here? If we do, why not mention it in the rest of the list. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I will use "CR" across the whole list. Thanks! |
||
* AwsNfsVolumeBackup | ||
* AwsNfsBackupSchedule | ||
* AwsNfsVolumeRestore | ||
|
||
### Google Cloud Kyma Clusters | ||
|
||
Kyma clusters provisioned on Google Cloud use Filestore as the underlying NFS. See the following list of all CRs available in the Cloud Manager module dedicated to Google Cloud Filestore: | ||
|
||
* GcpNfsVolume CR | ||
* GcpNfsVolumeBackup | ||
* GcpNfsBackupSchedule | ||
* GcpNfsVolumeRestore | ||
|
||
## Related information | ||
|
||
* [Cloud Manager Resources: NFS](./resources/README.md#nfs). | ||
* [Tutorials](./tutorials/README.md). | ||
* Pricing (link TBD) |
Original file line number | Diff line number | Diff line change | ||||
---|---|---|---|---|---|---|
@@ -1 +1,38 @@ | ||||||
# Redis | ||||||
|
||||||
## Overview | ||||||
|
||||||
The Cloud Manager module provides managed Redis which allows you to use caching functionality directly from your cloud provider. | ||||||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more.
Suggested change
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. When you provision a Redis instance, you can choose between the Standard and Premium Tier.
|
||||||
|
||||||
When you create a Redis instance in Kyma, you depend on the cloud provider of your Kyma cluster. The cloud provider in use determines the exact implementation. | ||||||
|
||||||
Cloud Manager module supports the Redis feature of three cloud providers: | ||||||
|
||||||
* Amazon Web Services' [Amazon ElastiCashe for Redis OSS](https://aws.amazon.com/elasticache/redis/) | ||||||
* Google Cloud's [Memorystore](https://cloud.google.com/memorystore?hl=en) | ||||||
* Microsoft Azure's [Azure Cache for Redis](https://azure.microsoft.com/en-us/products/cache) | ||||||
|
||||||
You can configure Cloud Manager's Redis instances using a dedicated Redis instance custom resource corresponding with the cloud provider for your Kyma cluster, namely AwsRedisInstance CR, GcpRedisInstance CR, or AzureRedisInstance CR. | ||||||
|
||||||
## Prerequisites | ||||||
|
||||||
To instantiate Redis, an IpRange CR must exist in the Kyma cluster. IpRange defines network address space reserved for your cloud provider's NFS resources. If you don't create the IpRange CR manually, Cloud Manager creates a default IpRange CR with the default address space and Classless Inter-Domain Routing (CIDR) selected. For more information, see [IpRange Custom Resoucre](./resources/04-10-iprange.md). | ||||||
|
||||||
## Lifecycle | ||||||
|
||||||
AwsRedisInstance, GcpRedisInstance, and AzureRedisInstance are namespace-level CRs. Once you create an AwsRedisInstance, GcpRedisInstance, or AzureRedisInstance resource, the following are also created automatically: | ||||||
|
||||||
* IpRange CR | ||||||
* IpRange is a cluster-level CR. | ||||||
* Only one IpRange CR can exist per cluster. | ||||||
* If you don't want the default IpRange to be used, create one manually. | ||||||
* Secret CR | ||||||
* The Secret is a namespace-level CR. | ||||||
* The Secret's name is the same as the name of the respective Redis instance CR. | ||||||
* The Secret holds values and information used to access the Redis instance. | ||||||
|
||||||
## Related Information | ||||||
|
||||||
* [Cloud Manager Resources: NFS](./resources/README.md#redis). | ||||||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more.
Suggested change
|
||||||
* [Tutorials](./tutorials/README.md). | ||||||
* Pricing (link TBD) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@ngrkajac please provide the sentence about the standard setup and the benefit of the NFS feature over the standard approach
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I will wait for Nikola's comment here as the product owner. But what you mentioned above is clear and on the point. NFS is not the feature here, it is just the protocol. ReadWriteMany is the feature, and as you mentioned RWX volumes can be used by multiple workloads. While regular persistent disk volumes can't be shared and can be used by only a single workload.