A Terraform provider for managing Apache Pulsar Entities.
-
From terraform registry
This provider has been published in the Terraform Registry:
terraform { required_providers { pulsar = { version = "0.1.0" source = "registry.terraform.io/streamnative/pulsar" } } }
-
From source code
- Clone this repository and cd into the directory
- Run
make build
, it will generate a binary file namedterraform-provider-pulsar
- Copy this
terraform-provider-pulsar
binary file to your terraform plugin directory based on your OS:
Operating System | User plugins directory |
---|---|
Windows(amd64) | %APPDATA%\terraform.d\plugins\registry.terraform.io\streamnative\pulsar\0.1.0\windows_amd64\ |
Linux(amd64) | ~/.terraform.d/plugins/registry.terraform.io/streamnative/pulsar/0.1.0/linux_amd64/ |
MacOS(amd64) | ~/.terraform.d/plugins/registry.terraform.io/streamnative/pulsar/0.1.0/darwin_amd64/ |
- Run
make build-dev
, it will build the binary and copy it to the plugin directory automatically.
Example provider with apache pulsar cluster, running locally with authentication disabled.
provider "pulsar" {
web_service_url = "http://localhost:8080"
token = "my_auth_token"
}
Property | Description | Required |
---|---|---|
web_service_url |
URL of your Apache Pulsar Cluster | Yes |
token |
Authentication Token for your Apache Pulsar Cluster, which is required only if your cluster has authentication enabled | No |
tls_trust_certs_file_path |
Path to a custom trusted TLS certificate file | No |
tls_allow_insecure_connection |
Boolean flag to accept untrusted TLS certificates | No |
api_version |
Used to request Apache Pulsar API service, default by 0, which represents use default version | No |
audience |
The OAuth 2.0 resource server identifier for the Pulsar cluster | No |
client_id |
The OAuth 2.0 client identifier | No |
issuer_url |
The OAuth 2.0 URL of the authentication provider which allows the Pulsar client to obtain an access token | No |
scope |
The OAuth 2.0 scope to request | No |
key_file_path |
The path of the private key file | No |
A resource for managing Apache Pulsar Clusters, can update various properties for a given cluster.
provider "pulsar" {
web_service_url = "http://localhost:8080"
}
resource "pulsar_cluster" "my_cluster" {
cluster = "eternals"
cluster_data {
web_service_url = "http://localhost:8080"
broker_service_url = "http://localhost:6050"
peer_clusters = ["skrulls", "krees"]
}
}
Property | Description | Required |
---|---|---|
cluster |
Name of the Cluster that you want to create | Yes |
cluster_data |
A Map of required fields for the cluster | Yes |
web_service_url |
Required in cluster data, pointing to your broker web service | Yes |
web_service_url_tls |
Pointing to your broker web service via tls | No |
broker_service_url |
Required in cluster data for broker discovery | Yes |
broker_service_url_tls |
Required in cluster data for broker discovery via tls | No |
peer_clusters |
Required in cluster data for adding peer clusters | Yes |
A resource for managing Pulsar Tenants, can update admin roles and allowed clusters for a tenant.
provider "pulsar" {
web_service_url = "http://localhost:8080"
}
resource "pulsar_tenant" "my_tenant" {
tenant = "thanos"
allowed_clusters = ["pulsar-cluster-1"]
admin_roles = ["godmode"]
}
Property | Description | Required |
---|---|---|
tenant |
Name of the Tenant that you want to create | Yes |
allowed_clusters |
An Array of clusters, accessible by this tenant | No |
admin_roles |
Admin Roles to be assumed by this Tenant | No |
A resource for creating and managing Apache Pulsar Namespaces, can update various properties for a given namespace.
provider "pulsar" {
web_service_url = "http://localhost:8080"
}
resource "pulsar_cluster" "test_cluster" {
cluster = "skrulls"
cluster_data {
web_service_url = "http://localhost:8080"
broker_service_url = "http://localhost:6050"
peer_clusters = ["standalone"]
}
}
resource "pulsar_tenant" "test_tenant" {
tenant = "thanos"
allowed_clusters = [pulsar_cluster.test_cluster.cluster, "standalone"]
}
resource "pulsar_namespace" "test" {
tenant = pulsar_tenant.test_tenant.tenant
namespace = "eternals"
enable_deduplication = true
// If defined partially, plan would show difference
// however, none of the mising optionals would be changed
namespace_config {
anti_affinity = "anti-aff"
max_consumers_per_subscription = "50"
max_consumers_per_topic = "50"
max_producers_per_topic = "50"
message_ttl_seconds = "86400"
replication_clusters = ["standalone"]
}
dispatch_rate {
dispatch_msg_throttling_rate = 50
rate_period_seconds = 50
dispatch_byte_throttling_rate = 2048
}
retention_policies {
retention_minutes = "1600"
retention_size_in_mb = "10000"
}
backlog_quota {
limit_bytes = "10000000000"
limit_seconds = "-1"
policy = "consumer_backlog_eviction"
type = "destination_storage"
}
persistence_policies {
bookkeeper_ensemble = 1 // Number of bookies to use for a topic, default: 0
bookkeeper_write_quorum = 1 // How many writes to make of each entry, default: 0
bookkeeper_ack_quorum = 1 // Number of acks (guaranteed copies) to wait for each entry, default: 0
managed_ledger_max_mark_delete_rate = 0.0 // Throttling rate of mark-delete operation (0 means no throttle), default: 0.0
}
permission_grant {
role = "some-role"
actions = ["produce", "consume", "functions"]
}
}
Property | Description | Required |
---|---|---|
tenant |
Name of the Tenant managing this namespace | Yes |
namespace |
name of the namespace | Yes |
enable_deduplication |
Message deduplication state on a namespace | No |
namespace_config |
Configuration for your namespaces like max allowed producers to produce messages | No |
dispatch_rate |
Apache Pulsar throttling config | No |
retention_policies |
Data retention policies | No |
backlog_quota |
Backlog Quota for all topics | No |
persistence_policies |
Persistence policies for all topics under a given namespace | No |
permission_grant |
Permission grants on a namespace. This block can be repeated for each grant you'd like to add | No |
namespace_config nested schema
Property | Description | Required |
---|---|---|
anti_affinity |
Anti-affinity group name | No |
is_allow_auto_update_schema |
Is schema auto-update allowed | No |
max_consumers_per_subscription |
Sets the max consumers per subscription | No |
max_consumers_per_topic |
Sets the max consumers per topic | No |
max_producers_per_topic |
Sets the max producers per topic | No |
message_ttl_seconds |
Sets the message TTL in seconds | No |
replication_clusters |
List of replication clusters for the namespace | No |
schema_compatibility_strategy |
Set schema compatibility strategy | No |
schema_validation_enforce |
Enable or disable schema validation | No |
The schema_compatibility_strategy
can take the following values:
- AutoUpdateDisabled
- Backward
- Forward
- Full
- AlwaysCompatible
- BackwardTransitive
- ForwardTransitive
- FullTransitive
A resource for creating and managing Apache Pulsar Topics, can update partitions for a given partition topic.
provider "pulsar" {
web_service_url = "http://localhost:8080"
}
resource "pulsar_topic" "sample-topic-1" {
tenant = "public"
namespace = "default"
topic_type = "persistent"
topic_name = "partition-topic"
partitions = 4 # partitions > 0 means this is a partition topic
permission_grant {
role = "some-role"
actions = ["produce", "consume", "functions"]
}
retention_policies {
retention_time_minutes = 1600
retention_size_mb = 20000
}
}
resource "pulsar_topic" "sample-topic-2" {
tenant = "public"
namespace = "default"
topic_type = "persistent"
topic_name = "non-partition-topic"
partitions = 0 # partitions = 0 means this is a non-partition topic
permission_grant {
role = "some-role"
actions = ["produce", "consume", "functions"]
}
retention_policies {
retention_time_minutes = 1600
retention_size_mb = 20000
}
}
Property | Description | Required |
---|---|---|
tenant |
Name of the Tenant managing this topic | Yes |
namespace |
Name of the Namespace for this topic | Yes |
topic_type |
Topic persistence (persistent , non-persistent ) |
Yes |
topic_name |
Name of the topic | Yes |
partitions |
Number of partitions (0 for non-partitioned topic, > 1 for partitioned topic) |
Yes |
permission_grant |
Permission grants on a topic. This block can be repeated for each grant you'd like to add. Permission grants are also inherited from the topic's namespace. | No |
retention_policies |
Data retention policies | No |
A resource for creating and managing Apache Pulsar Sources.
provider "pulsar" {
web_service_url = "http://localhost:8080"
api_version = "3"
}
resource "pulsar_source" "source-1" {
provider = pulsar
name = "source-1"
tenant = "public"
namespace = "default"
archive = "testdata/pulsar-io/pulsar-io-file-2.8.1.nar"
destination_topic_name = "source-1-topic"
processing_guarantees = "EFFECTIVELY_ONCE"
configs = "{\"inputDirectory\":\"opt\"}"
cpu = 2
disk_mb = 20480
ram_mb = 2048
}
Property | Description | Required |
---|---|---|
name |
The source's name | True |
tenant |
The source's tenant | True |
namespace |
The source's namespace | True |
destination_topic_name |
The Pulsar topic to which data is sent | True |
archive |
The path to the NAR archive for the Source. It also supports url-path [http/https/file (file protocol assumes that file already exists on worker host)] from which worker can download the package | True |
classname |
The source's class name if archive is file-url-path (file://) | False |
configs |
User defined configs key/values (JSON string) | False |
deserialization_classname |
The SerDe classname for the source | False |
processing_guarantees |
Define the message delivery semantics, default to ATLEAST_ONCE (ATLEAST_ONCE, ATMOST_ONCE, EFFECTIVELY_ONCE) | False |
parallelism |
The source's parallelism factor | False |
cpu |
The CPU that needs to be allocated per source instance (applicable only to Docker runtime) | False |
ram_mb |
The RAM that need to be allocated per source instance (applicable only to the process and Docker runtimes) | False |
disk_mb |
The disk that need to be allocated per source instance (applicable only to Docker runtime) | False |
runtime_flags |
User defined configs key/values (JSON string) | False |
A resource for creating and managing Apache Pulsar Sinks.
provider "pulsar" {
web_service_url = "http://localhost:8080"
api_version = "3"
}
resource "pulsar_sink" "sample-sink-1" {
provider = pulsar
name = "sample-sink-1"
tenant = "public"
namespace = "default"
inputs = ["sink-1-topic"]
subscription_position = "Latest"
cleanup_subscription = false
parallelism = 1
auto_ack = true
processing_guarantees = "EFFECTIVELY_ONCE"
archive = "testdata/pulsar-io/pulsar-io-jdbc-postgres-2.8.1.nar"
configs = "{\"jdbcUrl\":\"jdbc:clickhouse://localhost:8123/pulsar_clickhouse_jdbc_sink\",\"password\":\"password\",\"tableName\":\"pulsar_clickhouse_jdbc_sink\",\"userName\":\"clickhouse\"}"
}
Property | Description | Required |
---|---|---|
tenant |
The sink's tenant | True |
namespace |
The sink's namespace | True |
name |
The sink's name | True |
inputs |
The sink's input topics | False |
topics_pattern |
TopicsPattern to consume from list of topics under a namespace that match the pattern | False |
input_specs |
The map of input topics specs | False |
configs |
User defined configs key/values (JSON string) | False |
archive |
Path to the archive file for the sink. It also supports url-path [http/https/file (file protocol assumes that file already exists on worker host)] from which worker can download the package | True |
subscription_name |
Pulsar source subscription name if user wants a specific subscription-name for input-topic consumer | False |
subscription_position |
Pulsar source subscription position if user wants to consume messages from the specified location (Latest, Earliest) | False |
cleanup_subscription |
Whether the subscriptions the functions created/used should be deleted when the functions was deleted | True |
processing_guarantees |
Define the message delivery semantics, default to ATLEAST_ONCE (ATLEAST_ONCE, ATMOST_ONCE, EFFECTIVELY_ONCE) | False |
retain_ordering |
Sink consumes and sinks messages in order | False |
auto_ack |
Whether or not the framework will automatically acknowledge messages | True |
timeout_ms |
The message timeout in milliseconds | False |
parallelism |
The sink's parallelism factor | False |
cpu |
The CPU that needs to be allocated per sink instance (applicable only to Docker runtime) | False |
ram_mb |
The RAM that need to be allocated per sink instance (applicable only to the process and Docker runtimes) | False |
disk_mb |
The disk that need to be allocated per sink instance (applicable only to Docker runtime) | False |
custom_schema_inputs |
The map of input topics to Schema types or class names (as a JSON string) | False |
custom_serde_inputs |
The map of input topics to SerDe class names (as a JSON string) | False |
custom_runtime_options |
A string that encodes options to customize the runtime | False |
All resources could be imported using the standard terraform way.
terraform import pulsar_cluster.standalone standalone
- Change directory to the project </path/to/provider/terraform-provider-pulsar>
- In order to test the provider, you can run
make test
- In order to run the full suite of Acceptance tests, run
make testacc
Note: Acceptance tests create real resources, and often cost money to run.
Terraform is the work of thousands of contributors. We appreciate your help!
To contribute, please read the contribution guidelines: Contributing to Terraform - Apache Pulsar Provider
Issues on GitHub are intended to be related to bugs or feature requests with provider codebase. See https://www.terraform.io/docs/extend/community/index.html for a list of community resources to ask questions about Terraform.