docs: First iteration of active-active guide steps (#876)

* Align projection and RES samples to make them fit better with RES as an iterative step from the gRPC Projection sample * filtering in RES optional replication only for vip --------- Co-authored-by: Patrik Nordwall <[email protected]> Co-authored-by: Renato Cavalcanti <[email protected]>
akka · May 5, 2023 · 865d3ab · 865d3ab
1 parent 6288ce6
commit 865d3ab
Show file tree

Hide file tree

Showing 43 changed files with 876 additions and 1,150 deletions.
diff --git a/.github/workflows/checks.yml b/.github/workflows/checks.yml
@@ -149,7 +149,7 @@ jobs:
       - name: Compile Scala Projection gRPC sample Shopping Cart
         run: |-
           cd samples/grpc/shopping-cart-service-scala
-          sbt compile -Dakka-projection.version=`cat ~/.version`
+          sbt test -Dakka-projection.version=`cat ~/.version`
 
       - name: Compile Scala Projection gRPC sample Analytics
         run: |-
@@ -159,7 +159,7 @@ jobs:
       - name: Compile Java Projection gRPC sample Shopping Cart
         run: |-
           cd samples/grpc/shopping-cart-service-java
-          mvn compile -nsu -Dakka-projection.version=`cat ~/.version`
+          mvn test -nsu -Dakka-projection.version=`cat ~/.version`
 
       - name: Compile Java Projection gRPC sample Analytics
         run: |-

diff --git a/...istributed-cluster-docs/src/main/paradox/guide/1-event-sourced-shopping-cart.md b/...istributed-cluster-docs/src/main/paradox/guide/1-event-sourced-shopping-cart.md
@@ -12,7 +12,7 @@ The [Event Sourcing with Akka video](https://akka.io/blog/news/2020/01/07/akka-e
 
 ### Commands
 
-Commands are the "external" API of an entity. Entity state can only be changed by commands. The results of commands are emitted as events. A command can request state changes, but different events might be generated depending on the current state of the entity. A command can also be rejected if it has invalid input or can’t be handled by current state of the entity.
+Commands are the public API of an entity that other parts of the system use to interact with it. Entity state can only be changed by commands. The results of commands are emitted as events. A command can request state changes, and different events might be generated depending on the current state of the entity. A command can also be rejected if it has invalid input or can’t be handled by the current state of the entity.
 
 Scala
 :  @@snip [ShoppingCart.scala](/samples/grpc/shopping-cart-service-scala/src/main/scala/shopping/cart/ShoppingCart.scala) { #commands #events }
@@ -22,8 +22,8 @@ Java
 
 ### State
 
-Items added to the Cart are added to a `Map`. The contents of the `Map` comprise the Cart’s state along with a customer id and customer category for the customer
-owning the cart, if set, and a checkout timestamp if the cart was checked out:
+Items added to the Cart are added to a `Map`. The contents of the `Map` comprise the Cart’s state along with an optional checkout timestamp, which
+is set when the cart is checked out:
 
 Scala
 :  @@snip [ShoppingCart.scala](/samples/grpc/shopping-cart-service-scala/src/main/scala/shopping/cart/ShoppingCart.scala) { #state }
@@ -34,15 +34,26 @@ Java
 
 ### Command handler
 
-The Cart entity will receive commands that request changes to Cart state. We will implement a command handler to process these commands and emit a reply. Our business logic allows only items to be added which are not in the cart yet and require a positive quantity.
+The Cart entity will receive commands that request changes to Cart state. We will implement a command handler to process these commands and emit a reply,
+the handler logic selected is different depending on if the cart is checked out already, replying with an error, or if the cart is still open for
+adding and removing items.
 
+The command handler for an open cart looks like this:
 
 Scala
 :  @@snip [ShoppingCart.scala](/samples/grpc/shopping-cart-service-scala/src/main/scala/shopping/cart/ShoppingCart.scala) { #commandHandler }
 
 Java
 :  @@snip [ShoppingCart.java](/samples/grpc/shopping-cart-service-java/src/main/java/shopping/cart/ShoppingCart.java) { #commandHandler }
 
+@@@ div { .group-java }
+
+The actual logic for handling the commands is implemented in methods on the `ShoppingCart` class, for example the `onAddItem` method: 
+
+Java
+:  @@snip [ShoppingCart.java](/samples/grpc/shopping-cart-service-java/src/main/java/shopping/cart/ShoppingCart.java) { #onAddItem }
+
+@@@
 
 ### Event handler
 
@@ -69,8 +80,8 @@ Java
 
 ## Serialization
 
-The state, commands and events of the entity must be serializable because they are written to the datastore or sent between nodes within the Akka cluster. The template project includes built-in CBOR serialization using the @extref[Akka Serialization Jackson module](akka:serialization-jackson.html). This section describes how serialization is implemented. You do not need to do anything specific to take advantage of CBOR, but this section explains how it is included.
-The state, commands and events are marked as CborSerializable which is configured to use the built-in CBOR serialization. The template project includes this marker interface CborSerializable:
+The state, commands and events of the entity must be serializable because they are written to the datastore or sent between nodes within the Akka cluster. The sample project includes built-in CBOR serialization using the @extref[Akka Serialization Jackson module](akka:serialization-jackson.html). This section describes how serialization is implemented. You do not need to do anything specific to take advantage of CBOR, but this section explains how it is included.
+The state, commands and events are marked as CborSerializable which is configured to use the built-in CBOR serialization. The sample project includes this marker interface CborSerializable:
 
 Scala
 :  @@snip [CborSerializable.scala](/samples/grpc/shopping-cart-service-scala/src/main/scala/shopping/cart/CborSerializable.scala) { }

diff --git a/akka-distributed-cluster-docs/src/main/paradox/guide/2-service-to-service.md b/akka-distributed-cluster-docs/src/main/paradox/guide/2-service-to-service.md
@@ -40,13 +40,14 @@ Java
 
 Events can be transformed by application specific code on the producer side. The purpose is to be able to have a
 different public representation from the internal representation (stored in journal). The transformation functions
-are registered when creating the `EventProducer` service. Here is an example of one of those transformation functions:
+are registered when creating the `EventProducer` service. Here is an example of one of those transformation functions
+accessing the projection envelope to include the shopping cart id in the public message type passed to consumers:
 
 Scala
-:  @@snip [PublishEvents.scala](/samples/grpc/shopping-cart-service-scala/src/main/scala/shopping/cart/PublishEvents.scala) { #transformItemAdded }
+:  @@snip [PublishEvents.scala](/samples/grpc/shopping-cart-service-scala/src/main/scala/shopping/cart/PublishEvents.scala) { #transformItemUpdated }
 
 Java
-:  @@snip [PublishEvents.java](/samples/grpc/shopping-cart-service-java/src/main/java/shopping/cart/PublishEvents.java) { #transformItemAdded }
+:  @@snip [PublishEvents.java](/samples/grpc/shopping-cart-service-java/src/main/java/shopping/cart/PublishEvents.java) { #transformItemUpdated }
 
 To omit an event the transformation function can return @scala[`None`]@java[`Optional.empty()`].
 
@@ -61,11 +62,11 @@ Java
 
 ## Consume events
 
-The consumer is defined in a separate service, the shopping analytics service. It is a separate project with its own lifecycle,
-it is started, stopped, deployed separately, and has its own separate database from the shopping cart service. It may run in the
-same data center or cloud region as the shopping cart, but it could also run in a completely different location.
+The consumer is defined in a separate @java[maven]@scala[sbt] project in the shopping analytics service.
 
-FIXME a bit messy this section
+The analytics service runs in a separate Akka cluster which is deployed and scaled separately from the shopping cart service.
+When running it will have its own separate database from the shopping cart service. It may run in the same region as 
+the shopping cart, but it could also run in a completely different location.
 
 On the consumer side the `Projection` is a @extref[SourceProvider for eventsBySlices](akka-projection:eventsourced.html#sourceprovider-for-eventsbyslices)
 that is using `eventsBySlices` from the GrpcReadJournal. We use @extref[ShardedDaemonProcess](akka:typed/cluster-sharded-daemon-process.html) to distribute the instances of the Projection across the cluster.
@@ -86,9 +87,8 @@ The gRPC connection to the producer is defined in the @extref[consumer configura
 
 The @extref:[R2dbcProjection](akka-persistence-r2dbc:projection.html) has support for storing the offset in a relational database using R2DBC.
 
-The event handler for this sample is just logging the events rather than for example actually building its own read side:
-
-FIXME should we make it more realistic?
+The event handler for this sample is just logging the events rather than for example actually building its own read side
+in a database for querying:
 
 Scala
 :  @@snip [ShoppingCartEventConsumer.scala](/samples/grpc/shopping-analytics-service-scala/src/main/scala/shopping/analytics/ShoppingCartEventConsumer.scala) { #eventHandler }
@@ -179,7 +179,184 @@ And the consuming analytics service:
 * Java: https://github.com/akka/akka-projection/tree/main/samples/grpc/shopping-analytics-service-java
 * Scala: https://github.com/akka/akka-projection/tree/main/samples/grpc/shopping-analytics-service-scala
 
-FIXME running locally instructions here as well
+## Running the sample code locally
+
+With a copy of each of the two sample projects for the language of your liking you can run the two services locally on
+your own workstation. Docker, a JDK and @java[maven]@scala[sbt] is all that needs to be installed.
+
+### The Shopping Cart
+
+@@@ div { .group-scala }
+
+1. Start a local PostgresSQL server on default port 5432. The `docker-compose.yml` included in the shopping-cart project starts everything required for running locally.
+
+    ```shell
+    docker-compose up -d
+
+    # creates the tables needed for Akka Persistence
+    # as well as the offset store table for Akka Projection
+    docker exec -i postgres_db psql -U postgres -t < ddl-scripts/create_tables.sql
+    ```
+
+2. Start a first node:
+
+    ```shell
+    sbt -Dconfig.resource=local1.conf run
+    ```
+
+3. (Optional) Start another node with different ports:
+
+    ```shell
+    sbt -Dconfig.resource=local2.conf run
+    ```
+
+4. (Optional) More can be started:
+
+    ```shell
+    sbt -Dconfig.resource=local3.conf run
+    ```
+
+5. Check for service readiness
+
+    ```shell
+    curl http://localhost:9101/ready
+    ```
+
+6. Try it with [grpcurl](https://github.com/fullstorydev/grpcurl). Add at least a total quantity of 10 to the cart, smaller carts are excluded by the event filter.
+
+    ```shell
+    # add item to cart
+    grpcurl -d '{"cartId":"cart1", "itemId":"socks", "quantity":3}' -plaintext 127.0.0.1:8101 shoppingcart.ShoppingCartService.AddItem
+    
+    # get cart
+    grpcurl -d '{"cartId":"cart1"}' -plaintext 127.0.0.1:8101 shoppingcart.ShoppingCartService.GetCart
+    
+    # update quantity of item
+    grpcurl -d '{"cartId":"cart1", "itemId":"socks", "quantity":5}' -plaintext 127.0.0.1:8101 shoppingcart.ShoppingCartService.AddItem
+    
+    # check out cart
+    grpcurl -d '{"cartId":"cart1"}' -plaintext 127.0.0.1:8101 shoppingcart.ShoppingCartService.Checkout
+   
+    ```
+
+   or same `grpcurl` commands to port 8102 to reach node 2.
+
+@@@
+
+@@@ div { .group-java }
+
+1. Start a local PostgresSQL server on default port 5432. The included `docker-compose.yml` starts everything required for running locally.
+
+    ```shell
+    docker-compose up -d
+
+    # creates the tables needed for Akka Persistence
+    # as well as the offset store table for Akka Projection
+    docker exec -i postgres_db psql -U postgres -t < ddl-scripts/create_tables.sql
+    ```
+
+2. Make sure you have compiled the project
+
+    ```shell
+    mvn compile 
+    ```
+
+3. Start a first node:
+
+    ```shell
+    mvn compile exec:exec -DAPP_CONFIG=local1.conf
+    ```
+
+4. (Optional) Start another node with different ports:
+
+    ```shell
+    mvn compile exec:exec -DAPP_CONFIG=local2.conf
+    ```
+
+5. (Optional) More can be started:
+
+    ```shell
+    mvn compile exec:exec -DAPP_CONFIG=local3.conf
+    ```
+
+6. Check for service readiness
+
+    ```shell
+    curl http://localhost:9101/ready
+    ```
+
+7. Try it with [grpcurl](https://github.com/fullstorydev/grpcurl):
+
+    ```shell
+    # add item to cart
+    grpcurl -d '{"cartId":"cart1", "itemId":"socks", "quantity":3}' -plaintext 127.0.0.1:8101 shoppingcart.ShoppingCartService.AddItem
+    
+    # get cart
+    grpcurl -d '{"cartId":"cart1"}' -plaintext 127.0.0.1:8101 shoppingcart.ShoppingCartService.GetCart
+    
+    # update quantity of item
+    grpcurl -d '{"cartId":"cart1", "itemId":"socks", "quantity":5}' -plaintext 127.0.0.1:8101 shoppingcart.ShoppingCartService.AddItem
+    
+    # check out cart
+    grpcurl -d '{"cartId":"cart1"}' -plaintext 127.0.0.1:8101 shoppingcart.ShoppingCartService.Checkout
+    ```
+
+   or same `grpcurl` commands to port 8102 to reach node 2.
+
+@@@
+
+### The analytics service
+
+@@@ div { .group-scala }
+
+
+1. Start a local PostgresSQL server on default port 5432. The included `docker-compose.yml` starts everything required for running locally. Note that for convenience this service and the shopping cart service is sharing the same database, in an actual service consuming events the consuming services are expected to have their own separate databases.
+
+    ```shell
+    docker-compose up -d
+
+    # creates the tables needed for Akka Persistence
+    # as well as the offset store table for Akka Projection
+    docker exec -i postgres_db psql -U postgres -t < ddl-scripts/create_tables.sql
+    ```
+
+2. Start a first node:
+
+    ```shell 
+    sbt -Dconfig.resource=local1.conf run
+    ```
+
+3. Start `shopping-cart-service` and add item to cart
+
+4. Add at least a total quantity of 10 to the cart, smaller carts are excluded by the event filter.
+
+5. Notice the log output in the terminal of the `shopping-analytics-service`
+
+@@@
+
+@@@ div { .group-java }
+
+1. Start a local PostgresSQL server on default port 5432. The included `docker-compose.yml` starts everything required for running locally. Note that for convenience this service and the shopping cart service is sharing the same database, in an actual service consuming events the consuming services are expected to have their own separate databases.
+
+    ```shell
+    docker-compose up -d 
+
+    # creates the tables needed for Akka Persistence
+    # as well as the offset store table for Akka Projection
+    docker exec -i postgres_db psql -U postgres -t < ddl-scripts/create_tables.sql
+    ```
+
+2. Start a first node:
+
+    ```shell
+    mvn compile exec:exec -DAPP_CONFIG=local1.conf
+    ```
+
+3. Start `shopping-cart-service` and add item to cart
+
+4. Notice the log output in the terminal of the `shopping-analytics-service`
+
+@@@
 
 ## What's next?