High-availability deployment for the on-premises gateway
You can achieve high availability (HA) for the BMC Helix Intelligent Integrations on-premises gateway by deploying it in an active-passive configuration. This configuration ensures continuous operation with minimal downtime by automatically switching to a standby instance if the active instance fails.
A high-availability deployment of the on-premises gateway uses object storage to back up and restore the on-premises gateway configuration. You can use one of the following options as the object storage:
- BMC Helix object storage — BMC Helix object storage — The storage associated with the current BMC Helix SaaS or BMC Helix IT Operations Management (ITOM) tenant. This storage type is available for both BMC Helix ITOM users and BMC Helix subscribers.
- Private object storage — A storage deployed and managed in your on-premises environment. You can use a standalone instance of any S3-compatible storage (for example, MinIO or Amazon Simple Storage Service ) available in your environment as the private storage. This storage type is available for both BMC Helix ITOM on-premises users and BMC Helix subscribers.
HA deployment with BMC Helix object storage
When you configure an on-premises gateway to use the BMC Helix object storage, the high-availability deployment uses the object storage associated with the current BMC Helix SaaS or BMC Helix ITOM tenant to back up and restore the on-premises gateway configuration.
In a high-availability deployment, the first on-premises gateway instance (primary instance) that you configure is assigned as Active. When you configure the subsequent gateway instances (secondary instances), they are assigned as Standby. When the primary instance is up, it updates the configuration backup file on the storage according to the data push interval that you specify (default is 5 minutes), as shown in the following figure:
The secondary instance polls the object storage. The timestamp of the last backup performed by the primary instance is checked against the system clock of the secondary instance. If the primary instance fails, the time gap between the timestamp on the object storage and the system clock exceeds the data push interval. In such a case, the secondary instance takes over and is assigned as Active. The existing primary instance becomes the secondary instance and is assigned as Standby, as shown in the following figure. The backup is restored on the newly assigned primary instance, and the data collection resumes.
HA deployment with private object storage
When you configure an on-premises gateway instance to use the private object storage, a bucket is created on the storage with the name that you specify. The bucket contains the Leader_backup.json file for storing the backup of the configuration file.
The first instance that you configure is assigned as Active in the metadata of the configuration backup file. When you update any subsequent gateway instances with the storage details, they are assigned as Standby. When the primary instance is up, the primary instance updates the configuration backup file on the storage according to the data push interval that you specify (default is 5 minutes), as shown in the following figure:
The secondary instance polls the backup file on the storage. The timestamp of the last backup performed by the primary instance is checked against the system clock of the secondary instance. If the primary instance fails, the time gap between the backup file’s timestamp and the system clock exceeds the data push interval. In such a case, the secondary instance takes over and is assigned as Active in the metadata of the backup file. The existing primary instance becomes the secondary instance and is assigned as Standby, as shown in the following figure. The backup file is restored on the newly assigned primary instance, and the data collection resumes.
