pbm commands

pbm CLI is the command line utility to control the backup system. This page describes pbm commands available in Percona Backup for MongoDB.

For how to get started with Percona Backup for MongoDB, see Initial setup.

pbm backup

Creates a backup snapshot and saves it in the remote backup storage.

The command has the following syntax:

$ pbm backup [<flags>]

For more information about using pbm backup, see Starting a backup

The command accepts the following flags:

Flag	Description
-t, --type	The type of backup. Supported values: physical, logical (default), incremental, external. When not specified, Percona Backup for MongoDB makes a logical backup.
--base	For incremental backups only. Set the backup as the base and start tracking the incremental backup history to calculate and save the difference in data blocks for subsequent incremental backups.
--compression	Create a backup with compression. Supported compression methods: gzip, snappy, lz4, s2, pgzip, zstd. Default: s2 The none value means no compression is done during backup.
--compression-level	Configure the compression level from 0 to 10. The default value depends on the compression method used.
-o, --out=text	Shows the output format as either plain text or a JSON object. Supported values: text, json
--wait	Wait for the backup to finish. The flag blocks the shell session.
-l, --list-files	For external backups only. Shows the list of fines per node to copy.
--ns="database.collection"	Makes a logical backup of the specified namespace - the database and collection(s). To back up all collections in the database, specify the value in the --ns="database.*" format. In version 2.0.0, only a single namespace is supported for the backup.

JSON output

{
  "name": "<backup_name>",
  "storage": "<my-backup-dir>"
}

pbm backup-finish

Closes the backupCursor and finishes the external backup. Must be run after running pbm backup -t external. To learn more, refer to API for snapshot-based physical backups.

The command has the following syntax:

$ pbm backup-finish [backup-name] 

pbm cancel-backup

Cancels a running backup. The backup is marked as canceled in the backup list.

The command accepts the following flags:

Flag	Description
-o, --out=text	Shows the output format as either plain text or a JSON object. Supported values: text, json

JSON output

{
  "msg": "Backup cancellation has started"
}

pbm cleanup

Deletes outdated backup snapshots and point-in-time recovery oplog slices.

The command has the following syntax:

pbm delete-pitr [<flags>]

The command accepts the following flags:

Flag	Description
--older-than=TIMESTAMP	Deletes backups older than date / time specified in the format: - %Y-%M-%DT%H:%M:%S (e.g. 2020-04-20T13:13:20), - %Y-%M-%D (e.g. 2020-04-20), - XXd (e.g. 30d). Only days are supported
-w, --wait	Wait for the cleanup to finish. The flag blocks the shell session
-y, --yes	Cleans up the data storage without asking for a user’s confirmation
--dry-run	Checks for the old data to be deleted without deleting it. Allows to verify what data to delete

pbm config

Sets, changes or lists Percona Backup for MongoDB configuration.

The command has the following syntax:

$ pbm config [<flags>] [<key>]

The command accepts the following flags:

Flag	Description
--force-resync	Resync backup list with the current storage
--list	List current settings
--file=FILE	Upload the config information from a YAML file
--set=SET	Set a new config option value. Specify the option in the <key.name=value> format.
-o, --out=text	Shows the output format as either plain text or a JSON object. Supported values: text, json

PBM configuration output

{
  "pitr": {
    "enabled": false,
    "oplogSpanMin": 0
  },
  "storage": {
    "type": "filesystem",
    "s3": {
      "region": "",
      "endpointUrl": "",
      "bucket": ""
    },
    "azure": {},
    "filesystem": {
      "path": "<my-backup-dir>"
    }
  },
  "restore": {
    "batchSize": 500,
    "numInsertionWorkers": 10
  },
  "backup": {}
}

Setting a config value

[
  {
    "key": "pitr.enabled",
    "value": "true"
  }
]

pbm delete-backup

Deletes the specified backup snapshot or all backup snapshots that are older than the specified time. The command deletes backups that are not running regardless of the remote backup storage being used.

The following is the command syntax:

$ pbm delete-backup [<flags>] [<name>]

The command accepts the following flags:

Flag	Description
--older-than=TIMESTAMP	Deletes backups older than date / time specified in the format: - %Y-%M-%DT%H:%M:%S (e.g. 2020-04-20T13:13:20) or - %Y-%M-%D (e.g. 2020-04-20)
--force	Forcibly deletes backups without asking for user’s confirmation
--yes	Deletes backups without asking for user’s confirmation

pbm delete-pitr

Deletes oplog slices produced for Point-in-Time Recovery.

The command has the following syntax:

pbm delete-pitr [<flags>]

The command accepts the following flags:

Flag	Description
-a, --all	Deletes all oplog
--older-than=TIMESTAMP	Deletes oplog slices older than date / time specified in the format: - %Y-%M-%DT%H:%M:%S (e.g. 2020-04-20T13:13:20) or - %Y-%M-%D (e.g. 2020-04-20) When you specify a timestamp, Percona Backup for MongoDB rounds it down to align with the completion time of the closest backup snapshot and deletes oplog slices that precede this time. Thus, extra slices remain. This is done to ensure oplog continuity. To illustrate, the PITR time range is 2021-08-11T11:16:21 - 2021-08-12T08:55:25 and backup snapshots are: 2021-08-12T08:49:46Z 13.49MB [restore_to_time: 2021-08-12T08:50:06] 2021-08-11T11:36:17Z 7.37MB [restore_to_time: 2021-08-11T11:36:38] Say you specify the timestamp 2021-08-11T19:16:21. The closest backup is 2021-08-11T11:36:17Z 7.37KB [restore_to_time: 2021-08-11T11:36:38]. PBM rounds down the timestamp to 2021-08-11T11:36:38 and deletes all slices that precede this time. As a result, your PITR time range is 2021-08-11T11:36:38 - 2021-08-12T09:00:25. NOTE: Percona Backup for MongoDB doesn’t delete the oplog slices that follow the most recent backup. This is done to ensure point in time recovery from that backup snapshot. For example, if the snapshot is 2021-07-20T07:05:23Z [restore_to_time: 2021-07-21T07:05:44] and you specify the timestamp 2021-07-20T07:05:45, Percona Backup for MongoDB deletes only slices that were made before 2021-07-20T07:05:23Z.
--force	Forcibly deletes oplog slices without asking a user’s confirmation
-o, --out=json	Shows the output as either the plain text (default) or a JSON object. Supported values: text, json.
--yes	Deletes backups without asking for user’s confirmation

pbm describe-backup

Provides the detailed information about a backup:

• backup name
• type
• status
• namespaces - what was backed up during a selective backup
• size
• error message for failed backup
• last write timestamp
• last write time - human-readable indication of the last write
• last transition time - the timestamp when a backup changed its status
• cluster information: the replica set name, the backup status on this replica set, whether it is used as a config server replica set, last write timestamp
• replica set info: name, backup status, last write timestamp and last transition time, mongod security options, if encryption is configured.
• for snapshot-based backups, provides the list of files being copied

The command has the following syntax:

$ pbm describe-backup [<backup-name>] [<flags>] 

Flag	Description
-o, --out=text	Shows the status as either plain text or a JSON object. Supported values: text, json
-l, --list-files	Shows the list of files being copied for snapshot-based backups

JSON output

{
  "name": "<backup_name>",
  "opid": "<string>",
  "type": "logical",
  "last_write_ts": Timestamp,
  "last_transition_ts": Timestamp,
  "last_write_time": "2022-09-30T14:02:49Z",
  "last_transition_time": "2022-09-30T14:02:54Z",
  "namespaces": [
    "flight.booking"
  ],
  "mongodb_version": "<version>",
  "pbm_version": "<version>",
  "status": "done",
  "size": 470805945,
  "size_h": "449.0 MiB",
  "replsets": [
    {
      "name": "<name>",
      "status": "done",
      "last_write_ts": Timestamp,
      "last_transition_ts": Timestamp,
      "last_write_time": "2022-09-30T14:02:49Z",
      "last_transition_time": "2022-09-30T14:02:53Z"
    }
  ]
}

pbm describe-restore

Shows the detailed information about the restore.

The command has the following syntax:

$ pbm describe-restore [<restore-timestamp>] [<flags>] 

The command accepts the following flags:

Flag	Description
-c, --config=CONFIG	Only for physical restores. Points Percona Backup for MongoDB to a configuration file so it can read the restore status from the remote storage. For example, pbm describe-restore -c /etc/pbm/conf.yaml <restore-name>.
-o, --out=TEXT	Shows the output as either the plain text (default) or a JSON object. Supported values: text, json.

Selective restore status

{
 "name": "<restore_name>",
 "opid": "string",
 "backup": "<backup_name>",
 "type": "logical",
 "status": "done",
 "ts_to_restore": Timestamp,
 "time_to_restore": "Time",
 "namespaces": [
    "<database.*>"
 ]
 "replsets": [
   {
     "name": "rs1",
     "status": "done",
     "last_transition_ts": Timestamp,
     "last_transition_time": "Time"
   },
   {
    "name": "rs0",
     "last_transition_ts": Timestamp,
     "last_transition_time": "Time"
   },
   {
     "name": "cfg",
     "status": "done",
     "last_transition_ts": Timestamp,
     "last_transition_time": "Time"
   }
 ],
}

Physical restore status

{
 "name": "<restore_name>",
 "opid": "string",
 "backup": "<backup_name>",
 "type": "physical",
 "status": "done",
 "last_transition_ts": Timestamp,
 "last_transition_time": "Time",
 "replsets": [
   {
     "name": "rs1",
     "status": "done",
     "last_transition_ts": Timestamp,
     "last_transition_time": "Timestamp",
     "nodes": [
       {
         "name": "IP:port",
         "status": "done",
         "last_transition_ts": Timestamp,
         "last_transition_time": "Timestamp"
       }
     ]
   }
 ],
}

pbm help

Returns the help information about pbm commands.

pbm list

Provides the list of backups. In versions 1.3.4 and earlier, the command lists all backups and their states. Backup states are the following:

• In progress - A backup is running
• Canceled - A backup was canceled
• Error - A backup was finished with an error
• No status means a backup is complete

As of version 1.4.0, only successfully completed backups are listed. To view currently information about a running or a failed backup, run pbm status.

When Point-in-Time Recovery is enabled, the pbm list also provides the list of valid time ranges for recovery and point-in-time recovery status.

The command has the following syntax:

$ pbm list [<flags>]

The command accepts the following flags:

Flag	Description
--restore	Shows last N restores. Starting with version 2.0, the output shows restore names instead of backup names, as multiple restores can be done from a single backup.
--size=0	Shows last N backups. It also provides the information whether the restore is a selective one.
-o, --out=text	Shows the output format as either plain text or a JSON object. Supported values: text, json
--unbacked	Shows Point-in-Time Recovery oplog slices that were saved without the base backup snapshot. Available starting with version 1.8.0.
--replset-remapping	Maps the replica set names for the data restore / oplog replay. The value format is to_name_1=from_name_1,to_name_2=from_name_2

List of backups

{
  "snapshots": [
    {
      "name": "<backup_name>",
      "status": "done",
      "completeTS": Timestamp,
      "pbmVersion": "1.6.0"
    }
  ],
  "pitr": {
    "on": false,
    "ranges": [
      {
        "range": {
          "start": Timestamp,
          "end": Timestamp
        }
      },
      {
        "range": {
          "start": Timestamp,
          "end": Timestamp
        },
      {
        "range": {
          "start": Timestamp,
          "end": Timestamp (no base snapshot)
        }
      }
    ]
  }
}

Restore history

Full restore

 {
    "start": Timestamp,
    "status": "done",
    "type": "snapshot",
    "snapshot": "<backup_name>",
    "name": "<restore_name>"
  }

Selective restore

  {
    "start": Timestamp,
    "status": "done",
    "type": "snapshot",
    "snapshot": "<backup_name>",
    "name": "<restore_name>",
    "namespaces": [
      "<database.collection>"
    ]
  }

Point-in-time restore

  {
    "start": Timestamp,
    "status": "done",
    "type": "pitr",
    "snapshot": "<backup_name>",
    "point-in-time": Timestamp,
    "name": "<restore_name>"
  }

Selective point-in-time restore

{
    "start": Timestamp,
    "status": "done",
    "type": "pitr",
    "snapshot": "<backup_name>",
    "point-in-time": Timestamp,
    "name": "<restore_name>",
    "namespaces": [
      "<database.collection>"
    ]
  }
]

pbm logs

Shows log information from all pbm-agent processes.

The command has the following syntax:

$ pbm logs [<flags>]

The command accepts the following flags:

Flag	Description
-t, --tail=20	Shows last N entries. By default, the output shows last 20 entries. 0 means to show all log messages.
-e, --event=EVENT	Shows logs filtered by a specified event. Supported events: - backup - restore - resyncBcpList - pitr - pitrestore - delete
-o, --out=text	Shows log information as text (default) or in JSON format. Supported values: text, json
-n, --node=NODE	Shows logs for a specified node or a replica set. Specify the node in the format replset[/host:port]
-f, --follow	Follow log output. Allow to view the logs dynamically
-s, --severity=I	Shows logs filtered by severity level. Supported levels are (from low to high): D - Debug, I - Info (default), W - Warning, E - Error, F - Fatal. The output includes both the specified severity level and all higher ones
--timezone=TIMEZONE	Timezone of the log output. Supported values: UTC (default), local or the timezone in the IANA timezone format (e.g. America/New_York)
-i, --opid=OPID	Show logs for an operation in progress. The operation is identified by the OpID
-x, --extra	Show extra data in the text format

Find the usage examples in Viewing backup logs.

Logs output

[
  {
    "t": "",
    "s": 3,
    "rs": "rs0",
    "node": "example.mongodb.com:27017",
    "e": "",
    "eobj": "",
    "ep": {
      "T": 0,
      "I": 0
    },
    "msg": "listening for the commands"
  },
  ....
]

pbm oplog-replay

Allows to replay the oplog on top of any backup: logical, physical, storage level snapshot (like EBS-snapshot) and restore it to a specific point in time.

To learn more about the usage, refer to Point-in-Time Recovery oplog replay.

The command has the following syntax:

$ pbm oplog-replay [<flags>]

The command accepts the following flags:

Flag	Description
start=timestamp	The start time for the oplog replay.
end=timestamp	The end time for the oplog replay.
--replset-remapping	Maps the replica set names for the oplog replay. The value format is to_name_1=from_name_1,to_name_2=from_name_2.

pbm restore

Restores database from a specified backup / to a specified point in time. Depending on the backup type, makes either logical, physical, or a snapshot-based restore.

The command has the following syntax:

$ pbm restore [<flags>] [<backup_name>]

For more information about using pbm restore, see Restoring a backup.

The command accepts the following flags:

Flag	Description
--external	Indicates the backup as the one made outside PBM (for example, snapshot-based)
--time=TIME	Restores the database to the specified point in time. Available for logical restores and if Point-in-time recovery is enabled.
-w	Wait for the restore to finish. The flag blocks the shell session.
-o, --out=text	Shows the output format as either plain text or a JSON object. Supported values: text, json
--base-snapshot	Restores the database from a specified backup to the specified point in time. Without this flag, the most recent backup preceding the timestamp is used for point in recovery. Available in Percona Backup for MongoDB starting from version 1.6.0. Starting with version 2.2.0, this flag is mandatory for making a point-in-time recovery from physical backups. Without it, PBM looks for a logical backup to restore from.
--replset-remapping	Maps the replica set names for the data restore / oplog replay. The value format is to_name_1=from_name_1,to_name_2=from_name_2
--ns="database.collection"	Restores the specified namespace(s) - databases and collections. To restore all collections in the database, specify the values as --ns="database.*". The --ns flag accepts several namespaces as the comma-separated list. For example, ns=”db1.*,db2.coll2,db3.coll1,db3.collX”
-c, --config	The path to the mongod.conf file

Restore output

{
   "name": "<restore_name>"
   "snapshot": "<backup_name>"
}

Point-in-time restore

{
  "name":"<restore_name>",
  "point-in-time":"<backup_name>"
}

pbm restore-finish

Instructs PBM to complete the snapshot-based physical restore. Must be run after running pbm restore --external. To learn more, refer to API for snapshot-based physical backups.

The command has the following syntax:

$ pbm restore-finish <restore_name> [flags]

The command accepts the following flags:

Flag	Description
-c	The path to the PBM configuration file. Required to complete the restore.

pbm status

Shows the status of Percona Backup for MongoDB. The output provides the following information:

• pbm-agent processes version and state
• Currently running backups or restores
• Backups stored in the remote storage
• Point-in-Time Recovery status
• Valid time ranges for point-in-time recovery and the data size

The command accepts the following flags:

Flag	Description
-o, --out=text	Shows the status as either plain text or a JSON object. Supported values: text, json
-s, --sections=SECTIONS	Shows the status for the specified section. You can pass several flags to view the status for multiple sections. Supported values: cluster, pitr, running, backups.
--replset-remapping	Maps the replica set names for the data restore / oplog replay. The value format is to_name_1=from_name_1,to_name_2=from_name_2

Status information

{
  "backups": {
    "type": "FS",
    "path": "<my-backup-dir>",
    "snapshot": [
       ...
      {
        "name": "<backup_name>",
        "size": 3143396168,
        "status": "done",
        "completeTS": Timestamp,
        "pbmVersion": "1.6.0"
      },
    ],
    "pitrChunks": {
      "pitrChunks": [
         ...
        {
          "range": {
            "start": Timestamp,
            "end": Timestamp
          }
        },
        {
          "range": {
            "start": Timestamp,
            "end": Timestamp (no base snapshot) !!! no backup found
          }
        },
      ],
      "size": 677901884
    }
  },
  "cluster": [
    {
      "rs": "<replSet_name>",
      "nodes": [
        {
          "host": "<replSet_name>/example.mongodb:27017",
          "agent": "<version>",
          "ok": true
        }
      ]
    }
  ],
  "pitr": {
    "conf": true,
    "run": false,
    "error": "Timestamp.000+0000 E [<replSet_name>/example.mongodb:27017] [pitr] <error_message>"
  },
  "running": {
      "type": "backup",
      "name": "<backup_name>",
      "startTS": Timestamp,
      "status": "oplog backup",
      "opID": "6113b631ea9ba5b815fee7c6"
    }
}

pbm version

Shows the version of Percona Backup for MongoDB.

The command accepts the following flags:

Flag	Description
--short	Shows only version info
--commit	Shows only git commit info
-o, --out=text	Shows the output as either plain text or a JSON object. Supported values: text, json

Version information

{
  "Version": "1.6.0",
  "Platform": "linux/amd64",
  "GitCommit": "f9b9948bb8201ba1a6400f6558496934a0685efd",
  "GitBranch": "main",
  "BuildTime": "2021-07-28_15:24_UTC",
  "GoVersion": "go1.16.6"
}

Contact Us

For free technical help, visit the Percona Community Forum.

To report bugs or submit feature requests, open a JIRA ticket.

For paid support and managed or consulting services , contact Percona Sales.

---

Last update: August 22, 2023
Created: August 22, 2023