Acceldata
 Pulse
Search documentation… ⌘K

Last updated: Oct 29, 2026 07:53 UTC

Initialize the Pulse Manager Server

Before deploying Pulse services using the APIs or Admin Console, initialize the Pulse Manager Server and configure the required database services.

This page describes how to initialize the Manager Server, deploy database services, and verify their status in a Pulse Kubernetes environment.

Note

This page applies only to fresh installations. If a cluster is already configured in your Pulse Kubernetes environment, you can skip these steps.

Prerequisites

Before you begin, ensure that you have:

  • Configured the Manager Server ingress endpoint.
  • Identified the Pulse version to deploy.
  • Obtained the container registry URL and registry secret name.
  • Obtained the authorization credentials for the Manager Server APIs.
  • Made the Elasticsearch certificate file (cacerts) available, if databases are deployed on virtual machines (VMs) or bare-metal servers.

Step 1: Initialize the Manager Server

Initialize the Manager Server by calling the /init API with the Pulse version, container registry details, and data retention settings.

  1. Update the following parameters with values specific to your environment:

Parameter

Description

<Ingress API for Manager Server>

Ingress endpoint of the Manager Server.

<Pulse Version>

Version of Pulse to deploy.

registry_url

Container registry URL.

hdfs_retention_days

HDFS data retention period, in days.

mongo_retention_days

MongoDB data retention period, in days.

logstore_retention

Log retention setting.

metrics_retention

Metrics retention setting.

<registry secret name>

Kubernetes secret name used to access the container registry.

  1. Run the following command:
curl --request POST \
  --url http://<Ingress API for Manager Server>/api/v1/mserver/init \
  --header 'Authorization: Basic <base64-encoded-credentials>' \
  --header 'Content-Type: application/json' \
  --data '{
    "pulse_version": "<Pulse Version>",
    "registry_url": "191579300362.dkr.ecr.us-east-1.amazonaws.com/acceldata",
    "hdfs_retention_days": 31,
    "mongo_retention_days": 32,
    "logstore_retention": 32,
    "metrics_retention": 32,
    "registry_secret": "<registry secret name>"
 }'
  1. Replace <base64-encoded-credentials> with the appropriate authorization value.
  2. Verify that the API request completes successfully before proceeding.

Step 2: Upload Elasticsearch Certificates

Note

Complete this step only if the databases are deployed on VMs or bare-metal servers.

Upload the required certificates for the Elasticsearch deployment.

  1. Ensure that the cacerts file is available on the machine from which you run the API request.
  2. Run the following command:

curl --request PUT \
  --url http://<Ingress API for Manager Server>/api/v1/mserver/upload/certs/es \
  --header 'Authorization: Basic <base64-encoded-credentials>' \
  --header 'Content-Type: multipart/form-data' \
--form es-certs.p12=@cacerts
  1. Verify that the certificate upload completes successfully.

Step 3: Deploy Database Services

Note

Complete this step only if the databases are deployed on VMs or bare-metal servers.

Deploy the database services by calling the /deploy/db API.

  1. Run the following command:
curl --request PUT \
  --url http://<Ingress API for Manager Server>/api/v1/mserver/deploy/db \
--header 'Authorization: Basic <base64-encoded-credentials>'
  1. Verify that the API request completes successfully before proceeding.

Step 4: Verify Database Containers

Note

Complete this step only if the databases are deployed on VMs or bare-metal servers.

After deploying the database services, verify that the required containers are running.

  1. Log in to one of the database nodes.
  2. Run the following command:
docker ps
  1. Verify that the database containers appear in the output and that their status is Up.
    The following table lists the expected container types.

Container

Service

ad-db-router

MongoDB router

ad-db-config

MongoDB configuration server

ad-db-sh1rs1, ad-db-sh2rs1, ad-db-sh3rs1

MongoDB shard instances

ad-vminsert

VictoriaMetrics ingestion service

ad-vmselect

VictoriaMetrics query service

ad-vmstorage

VictoriaMetrics storage service

ad-redis

Redis

ad-pg-ui

PostgreSQL

ad-gauntlet

Gauntlet

ad-events

Event messaging service

ad-logstash

Logstash

ad-elastic

Elasticsearch

Important

The number of containers deployed on a database VM or bare-metal server depends on the total number of nodes configured in the Pulse Kubernetes environment.

MongoDB uses sharding and replication to distribute data across the cluster. For example, if three shards are configured with three replicas each, the cluster contains nine MongoDB data-bearing instances.

Step 5: Verify Database Status

After the database deployment is complete, verify the status of the database services by calling the /dbstatus API.

  1. Run the following command:
curl --request GET \
  --url http://<Ingress API for Manager Server>/api/v1/mserver/dbstatus \
--header 'Authorization: Basic <base64-encoded-credentials>'
  1. Review the API response to verify that all required database services are running and healthy.
  2. Resolve any reported issues before proceeding with Pulse service deployment.

Next Steps

After initializing the Manager Server and verifying the database status, proceed with deploying Pulse services using the APIs or Admin Console. For details, see Deploy Pulse on Kubernetes.