The AEM Orchestrator requires that there be an application.properties placed in the same root directory as the application JAR file.
There are several properties that do not have default values (generally because they are defined at stack creation), which need to be present in the application.properties. Here is a list of these properties with example values:
aws.cloudformation.stackName.author = example-aem-author-stack
aws.cloudformation.stackName.authorDispatcher = example-aem-author-dispatcher-stack
aws.cloudformation.stackName.publish = example-aem-publish-stack
aws.cloudformation.stackName.publishDispatcher = example-aem-publish-dispatcher-stack
aws.cloudformation.stackName.messaging = example-aem-messaging-stack
aws.sqs.queueName = example-aem-asg-event-queueThe aem-aws-stack-builder will generate these names for you, they just need to be added to this Orchestrator application.properties file.
If one of these required properties is not defined in the application.properties file, then you will see an error in the orchestrator.log file when the Orchestrator is started. Here is an example:
ERROR o.s.boot.SpringApplication - Application startup failed
...
Caused by: org.springframework.beans.InvalidPropertyException: Invalid property 'aws.sqs.queueName' of bean class [com.shinesolutions.aemorchestrator.model.EnvironmentValues]: Queue name cannot be empty or null
Here is a complete list of all configurable properties:
| Property Name | Default Value | Example Value |
|---|---|---|
| aws.region | ap-southeast-2 | |
| aws.sqs.queueName | example-aem-asg-event-queue | |
| aws.cloudformation.stackName.messaging | example-aem-messaging-stack | |
| aws.cloudformation.stackName.publishDispatcher | example-aem-publish-dispatcher-stack | |
| aws.cloudformation.stackName.publish | example-aem-publish-stack | |
| aws.cloudformation.stackName.authorDispatcher | example-aem-author-dispatcher-stack | |
| aws.cloudformation.stackName.author | example-aem-author-stack | |
| aws.cloudformation.autoScaleGroup.logicalId.publishDispatcher | PublishDispatcherAutoScalingGroup | |
| aws.cloudformation.autoScaleGroup.logicalId.publish | PublishAutoScalingGroup | |
| aws.cloudformation.autoScaleGroup.logicalId.authorDispatcher | AuthorDispatcherAutoScalingGroup | |
| aws.cloudformation.loadBalancer.logicalId.author | AuthorLoadBalancer | |
| aws.cloudformation.sns.logicalId.eventTopic | AEMASGEventTopic | |
| aws.client.connection.timeout | 30000 | |
| aws.client.max.errorRetry | 10 | |
| aws.client.useProxy | false | true |
| aws.client.proxy.host | yourdomain.proxy.com | |
| aws.client.proxy.port | 8080 | |
| aws.client.protocol | https | |
| aws.snapshot.tags | Component,StackPrefix | |
| aws.device.name | /dev/sdb | |
| aem.credentials.s3.use | false | true |
| aem.credentials.s3.file.uri | s3://bucket/file.ext | |
| aem.credentials.replicator.username | admin | |
| aem.credentials.replicator.password | admin | |
| aem.credentials.orchestrator.username | admin | |
| aem.credentials.orchestrator.password | admin | |
| aem.protocol.publishDispatcher | http | |
| aem.protocol.publish | http | |
| aem.protocol.authorDispatcher | http | |
| aem.protocol.author | http | |
| aem.port.publishDispatcher | 80 | |
| aem.port.publish | 4503 | |
| aem.port.authorDispatcher | 80 | |
| aem.port.author | 80 | |
| aem.client.api.debug | false | |
| aem.client.api.connection.timeout | 30000 | |
| aem.reverseReplication.enable | false | |
| aem.reverseReplication.transportUri.postfix | /bin/receive?sling:authRequestLogin=1 | |
| aem.relaxed.ssl.enable | true | false |
| endpoints.health.enabled | true | false |
| endpoints.info.enabled | true | false |
| startup.waitForAuthorElb.maxAttempts | 100 | |
| startup.waitForAuthorElb.backOffPeriod | 5000 | |
| http.client.relaxed.ssl.enable | true | false |
| You can also view the base application.properties file. |
By default the region is derived from the AWS host, so this property is only needed if planning to run the Orchestrator from a non AWS hosted instance. Helpful when integration testing from a local instance pointing to an AWS stack.
This is the main SQS queue that the Orchestrator will use to be notified about events on the stack. It's a required property and the Orchestrator cannot work without this being set.
The stack names are generated by the aem-aws-stack-builder. Each tier (Author, Author Dispatcher, Publish etc) has a different stack name. Generally the naming pattern is:
<stack prefix>-aem-<tier name>-stack
For example a publish tier stack with a prefix of 'example' will have a name:
example-aem-publish-stack
If you are unsure, look for the 'aws:cloudformation:stack-name' tag on instances within the tier. The tag will contain the stack name. These are required properties, as the stack names are needed to perform Cloud Formation resource lookups at startup.
The logical ids only need to be set if they have been changed from their default values given by the aem-aws-stack-builder. The logical ids are used in conjunction with the stack names (see property above) to get the AWS physical resource ids of the auto scale groups, load balancer and SNS topic at Orchestrator start up. You can view the logical id of a resource within the stack by looking at the 'aws:cloudformation:logical-id' tag.
The timeout in milliseconds when making client calls to the AWS API.
The max number of retry attempts when making client calls to the AWS API.
If the Orchestrator sits behind a proxy, then set this to true
If the proxy is enabled (aws.client.useProxy=true), then these properties will define the host and port for the proxy
By default the AWS client uses https, but this can be set to http if needed
This is a comma delimetered list of AWS tag names that are transferred from instance to the snapshot when the snapshot is taken. For example if the instance has 10 tags, and you want the Name and StackPrefix tags to be applied to the snapshot, then add them to this property e.g.:
aws.snapshot.tags=Name,StackPrefix
The AWS device name used when getting the Volume ID for creating the snapshot
Setting this to true will tell the Orchestrator to read the AEM credentials from an S3 bucket (see aem.credentials.s3.file.uri below for specifing the location), instead of the application.properties file.
The AWS S3 bucket location of an unencrypted JSON file containing AEM credentials. Here is an example of the file format:
{
"orchestrator": "<orchestrator-password>",
"replicator": "<replicator-password>"
}
The Orchestrator has two roles. Below is a table outlining what actions each role performs:
| Role | Actions |
|---|---|
| replicator | Create Flush/Replication Agent |
| orchestrator | Pause Replication Agent |
| orchestrator | Restart Replication Agent |
| orchestrator | Delete Flush/Replication Agent |
The AEM credentials (username/password) for the replicator and orchestrator roles. See table above for details
The protocol to use when performing AEM actions at each tier (Author, Author Dispatcher, Publish, Publish Dispatcher). Will be http or https.
The port to use when connecting to AEM at each tier (Author, Author Dispatcher, Publish, Publish Dispatcher).
If set to true, the Orchestrator will output debug information for all AEM http calls via the swagger-aem-4j API.
Connection timeout duration for all AEM http calls via the swagger-aem-4j API.
If set to true, the Orchestrator will create (and delete) a Reverse Replication Agent between the Publish and Author Primary in addition to the regular Replication Agent.
Only required if reverse replication is enabled and a different transport URI is required for reverse replication. Note the postfix only applies to the end of the Publish AEM url. For example:
<publish url><postfix>
The default postfix is:
/bin/receive?sling:authRequestLogin=1
Here is an example of a Publish URL:
http://localhost:4503
Enable if you want self-certified SSL certificates to be accepted by AEM when setting up replication agents.
The Orchestrator has a very basic JSON formatted health check page, which can be accessed via:
http://<orchestrator domain>:8080/health
By default the health check page is enabled, but this can be turned off by setting this property to false
The Orchestrator has a very basic JSON formatted info page, which can be accessed via:
http://<orchestrator domain>:8080/info
It generally contains information about environment properties loaded at startup. Can be useful for debugging.
Before the Orchestrator starts receiving messages on the SQS queue, it first waits for the Author Elastic Load Balancer (ELB) to be in a healthy state. This helps to avoid continuous errors in the logs at start up. This property defines the max attempts to check the Author ELB is healthy before terminating the Orchestrator.
Same as above, but defines how long to wait in seconds between checking the Author ELB is in a healthy state.
When Orchestrator performs health checks to AEM instances (via https) it will ignore host name verification if this is enabled.