# Backup Daemon > Source-anchored research for Backup Daemon. Run status: COMPLETED. Parent: [MongoDB Ops Manager and Cloud Manager](https://llms-explorer.com/tree/mongodb-ops-manager-and-cloud-manager/) · 5 facets · 19 facts · page: https://llms-explorer.com/tree/backup-daemon/ ## Structure and components - A July 2016 Ops Manager 1.8-to-2.0 upgrade account records the daemon becoming part of the Ops Manager RPM and recommends performing a backup after upgrading to verify daemon operation. This is historical upgrade evidence. (low confidence) — [source](https://mongomikeblog.wordpress.com/2016/07/15/upgrading-mongodb-ops-manager-from-1-8-x-to-2-0-x-release-using-the-rpm-upgrade-approach/) - The Ops Manager package installs the Backup Daemon component, but administrators must activate it to use it; the host service also has separate start, stop, restart, and status commands. (medium confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/tutorial/manage-backup-daemon.md) ## How it works - Administrators manage the daemon process through /etc/init.d/mongodb-mms-backup-daemon with start, stop, restart, and status commands. (low confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/tutorial/manage-backup-daemon.md) - The administration API creates an initial daemon configuration with PUT /daemon/configs/{MACHINE}, requiring the daemon process to be running first. Updating an existing configuration uses PUT /daemon/configs/{MACHINE}/{HEAD-ROOT-DIRECTORY}. (high confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/reference/api/admin/backup/daemonConfigs/create-one-backup-daemon-configuration.md) - Assignment labels restrict which backup jobs a daemon can process. Without labels, the daemon can process only jobs from projects that do not filter their jobs with labels. (high confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/reference/api/admin/backup/daemonConfigs/create-one-backup-daemon-configuration.md) - machine.headRootDirectory must end with a slash; omitting it causes a Java exception. Update requests must URL-encode directory slashes in the endpoint path. (high confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/reference/api/admin/backup/daemonConfigs/create-one-backup-daemon-configuration.md) - A March 2023 operator report for Ops Manager 6.0.11 records BackupDaemonPreFlightCheck repeatedly waiting for a system upgrade before starting. The forum response recommends support investigation and does not establish a root cause or a verified repair. (low confidence) — [source](https://www.mongodb.com/community/forums/t/backup-daemon-not-starting/219429) - The Ops Manager Backup Daemon performs backup maintenance, including groom jobs and queryable restores; it also creates incremental snapshots when using filesystem stores. (high confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/core/system-overview.md) - Administrators can dedicate a daemon to queryable restore work by disabling its other job types. The configuration API separately exposes backupJobsEnabled, garbageCollectionEnabled, and restoreQueryableJobsEnabled, with numWorkers controlling worker processes. (medium confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/tutorial/configure-backup-high-availability.md) - For initial daemon configuration, the documented API uses PUT /api/public/v1.0/admin/backup/daemon/configs/{MACHINE}; the daemon process must already be running. An already configured daemon should use the update endpoint. (low confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/reference/api/admin/backup/daemonConfigs/create-one-backup-daemon-configuration.md) ## Problems, failure modes and limitations - Ops Manager can run Backup Daemon maintenance on any configured application instance. The daemon runs grooming, incremental filesystem snapshot work, and queryable restores; client applications cannot communicate directly with it, and its state and job queues come from the Application Database. (low confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/core/system-overview.md) - Client applications cannot communicate directly with the Backup Daemon; its state and job queues come from the Ops Manager Application Database. (low confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/core/system-overview.md) ## Comparisons and alternatives - Stack Harbor, which says it operates client Ops Manager deployments, recommends sizing backup-daemon storage and IOPS for the total managed dataset and including the daemon in ongoing monitoring and upgrades. This is practitioner guidance rather than a MongoDB sizing formula. (low confidence) — [source](https://stackharbor.com/en/knowledge-base/mongodb-ops-manager-overview/) ## Facts and statements - Multiple Backup Daemons can increase concurrent job capacity and provide manual failover. Deploying multiple application hosts behind a load balancer does not remove the requirement to run at least one Backup Daemon. (high confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/core/system-overview.md) - Daemon configuration separates assignmentEnabled, backupJobsEnabled, garbageCollectionEnabled, restoreQueryableJobsEnabled, and resourceUsageEnabled. numWorkers controls worker processes for backup, restore, or grooming tasks. (high confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/reference/api/admin/backup/daemonConfigs/create-one-backup-daemon-configuration.md) - The MongoDB Enterprise Kubernetes Operator 1.14.0 release notes document a fix for a custom CA not being mounted into the Backup Daemon pod, which had prevented backups in hybrid-mode deployments using a custom CA. (low confidence) — [source](https://github.com/mongodb/mongodb-kubernetes/blob/master/RELEASE_NOTES_MEKO.md) - In its July 28, 2026 description of NDB 2.11 shard scale-out, Nutanix says NDB refreshes Ops Manager backup-daemon job references for the expanded topology before resuming protection and triggering a fresh baseline snapshot. This describes NDB integration behavior. (low confidence) (verified as of 2026-07-28) — [source](https://www.nutanix.com/blog/scaling-mongodb-sharded-clusters-with-nutanix-ndb) - NetBackup's MongoDB Ops Manager Administrator's Guide 11.1 requires clusters protected through its integration to be marked Third Party Managed in Ops Manager; its architecture retains Ops Manager in the protection workflow. (low confidence) (verified as of 2026-10-23) — [source](https://www.veritas.com/content/support/en_US/doc/167996996-170849162-1) - Multiple Backup Daemons increase concurrent job capacity and provide manual failover. After a daemon host fails, a Global Owner or global backup administrator can move its jobs to another daemon. (high confidence) — [source](https://www.mongodb.com/docs/ops-manager/current/core/system-overview.md)