Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions sdk/storage/azure-storage-blob/migrationGuides/V8_V12.md
Original file line number Diff line number Diff line change
Expand Up @@ -306,3 +306,10 @@ final BlockBlobClient blobClient = new BlobClientBuilder()
OutputStream blobOutputStream = blobClient.getBlobOutputStream();
// insert your method of choice to write to an OutputStream
```

## Miscellaneous
- Examples for how to approximate v8 StorageEvent behavior can be found [here][storage event sample].
- Examples for how to approximate v8 LocationMode behavior can be found [here][location mode sample].

[storage event sample]: https://github.com/Azure/azure-sdk-for-java/blob/main/sdk/storage/azure-storage-blob/src/samples/java/com/azure/storage/blob/StorageEventExample.java
[location mode sample]: https://github.com/Azure/azure-sdk-for-java/blob/main/sdk/storage/azure-storage-blob/src/samples/java/com/azure/storage/blob/LocationModeExample.java
Original file line number Diff line number Diff line change
@@ -0,0 +1,107 @@
// Copyright (c) Microsoft Corporation. All rights reserved.
// Licensed under the MIT License.

package com.azure.storage.blob;
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: missing license header


import com.azure.storage.common.policy.RequestRetryOptions;

import java.time.Duration;


/**
* This example shows how to approximate the LocationMode behavior from the track 1 SDK. It is a general translation to
* achieve roughly the same results, but it is not an identical implementation. It may be modified to suit the use case.
* For more information on redundant storage, see here:
* https://docs.microsoft.com/en-us/azure/storage/common/storage-redundancy
*
* In a sense, the track 2 SDK is always primary-only or primary-then-secondary. However, by passing the secondary
* endpoint as the primary and the primary as the secondary, the behavior of secondary-only or secondary-then-primary
* can be achieved. To avoid confusion of terms in this example, 'preferred' and 'fallback' will refer to the location
* that is tried by the client first and then second respectively, whereas 'primary' and 'secondary' will refer to the
* Storage concept of primary read-write storage and back-up/redundant/read-only storage respectively.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should this link to docs explaining back-up/redundant/read-only

*
* The general pattern is to create a BlobClient and pass the preferred location to the builder as the endpoint. To
* configure a fallback location, set it as the `secondaryEndpoint` on {@link RequestRetryOptions} and pass the
* configured options to {@link BlobClientBuilder#retryOptions(RequestRetryOptions)}. Switching LocationMode requires
* using a different client that is configured for the new request behavior. In this case, concurrency control should
* be carefully considered to prevent race conditions.
*
* Requests will always go first to the preferred location passed as the endpoint. If a request must be retried, it will
* and the error indicates it may be helped by checking the fallback, a request will immediately be reissued to the
* fallback. If that also fails and still a retry may be helpful, the client will wait for a backoff period specified by
* the retry options before retrying the initial location again.
*
* The client does not internally track the LocationMode or read it from an object that is passed because of how that
* might cause race conditions if it is shared between clients.
*
* Each of the clients constructed in this sample will have behavior according to the variable name. This sample does
* not demonstrate meaningful independent behavior, so running it will do nothing, but these clients can be copied and
* used as a component in other code.
*
* This example can be combined with the StorageEventExample to approximate the Circuit Breaker RAGRS sample here:
* https://github.com/Azure-Samples/storage-dotnet-circuit-breaker-ha-ra-grs/blob/master/storage-dotnet-circuit-breaker-ha-ra-grs/Program.cs
* In this case, in the StorageEvent callback, rather than switching the LocationMode on the DefaultRequestOptions, the
* client should be swapped out to the client with the appropriate LocationMode and the request reissued, alternating
* between a primary only and secondary only client.
*
* The main areas of divergence from the original LocationMode behavior are:
* - There is no LocationMode type
* - The v12 analogue of LocationMode is configured at client build time and is static for a given client; a new client
* must be used if different location behavior is desired.
* - Changing LocationMode entails changing the client being used to issue requests
*/
public class LocationModeExample {

public static void main(String[] args) {
String primaryEndpoint = "<primary-endpoint>";
String secondaryEndpoint = "<secondary-endpoint>";

BlobClient primaryOnlyClient;
BlobClient secondaryOnlyClient;
BlobClient primaryThenSecondaryClient;
BlobClient secondaryThenPrimaryClient;

BlobClientBuilder builder = new BlobClientBuilder()
.containerName("<container-name>")
.blobName("<blob-name>");

/*
This could be refactored into a helper methods, but it is written out explicitly here for clarity and ease of
comparison.
Null in all cases indicates accepting the default value.
A distinct set of options must be created for each client to prevent overwriting the options held by another
client.
*/
// Create a primary only client by passing the primary endpoint as the preferred and passing no fallback.
RequestRetryOptions primaryOnlyRetryOptions = new RequestRetryOptions(null, null, (Duration) null, null, null,
null);
primaryOnlyClient = builder
.endpoint(primaryEndpoint)
.retryOptions(primaryOnlyRetryOptions)
.buildClient();

// Create a secondary only client by passing the secondary as the preferred and passing no fallback.
RequestRetryOptions secondaryOnlyRetryOptions = new RequestRetryOptions(null, null, (Duration) null, null, null,
null);
secondaryOnlyClient = builder
.endpoint(secondaryEndpoint)
.retryOptions(secondaryOnlyRetryOptions)
.buildClient();

// Create a primary then secondary by passing a primary as the preferred and secondary as a fallback.
RequestRetryOptions primaryThenSecondaryRetryOptions = new RequestRetryOptions(null, null, (Duration) null,
null, null, secondaryEndpoint);
primaryThenSecondaryClient = builder
.endpoint(primaryEndpoint)
.retryOptions(primaryThenSecondaryRetryOptions)
.buildClient();

// Create a secondary then primary by passing a secondary as the preferred and a primary as a fallback.
RequestRetryOptions secondaryThenPrimaryRetryOptions = new RequestRetryOptions(null, null, (Duration) null,
null, null, primaryEndpoint);
secondaryThenPrimaryClient = builder
.endpoint(secondaryEndpoint)
.retryOptions(secondaryThenPrimaryRetryOptions)
.buildClient();
}
}