Control of virtual full job execution

116 views
Skip to first unread message

Tom

unread,
Feb 2, 2024, 5:15:03 AM2/2/24
to bareos-users
Hello Bareos Community,

I'm reaching out to seek insights from Bareos experts on the best practices for configuring always incremental backups and handling consolidation jobs.

Context:

- Mode: Always Incremental.
- Data Size: 17 TB across 6 directories for one client.
- Setup: 12 pools (6 for incremental backups and 6 for consolidation backups), each with unique storage directories tailored for specific volumes and backup types.
- 6 volumes per backup type are used because there is a large data size.
- The volumes are stored on SSDs and not on tape.
- For the storage of volumes, Bareos has a 35TB SSD in RAID5 configuration.

Issue:

* My objective is to execute daily incremental backups concurrently for efficiency, while ensuring that virtual full jobs run sequentially and ideally, should not be run consecutively, but one per day. I do not want the virtual fulls to run simultaneously given the volume size.

Questions:

- Is it possible to control when the virtual full launched by the consolidation job are executed?
- Is there a more efficient strategy or configuration that could improve the management of these backup types, considering the volume of data and the desired scheduling?
- Is it relevant to have multiple volumes for the incrementals for the same client?

I've included an example of my configuration for one volume below for reference.

```
Client {
    Name = client
    Address = ip
    FDPort = 9102
    Password = <password>
    Maximum Concurrent Jobs = 20
    Passive = yes
}

JobDefs {
  Name = "DefaultAIJob"
  Type = Backup
  Level = Incremental
  Schedule = "AI-Sched"
  Messages = Standard
  Priority = 10
  Write Bootstrap = "/data/bareos/bootstrap/%c.bsr"
  AllowDuplicateJobs = no
  CancelLowerLevelDuplicates = yes
  CancelQueuedDuplicates = yes
  CancelRunningDuplicates = no
  Accurate = yes
  Allow Mixed Priority = yes
  Always Incremental = yes
  Always Incremental Job Retention = 20 days
  Always Incremental Keep Number = 20
  Always Incremental Max Full Age = 40 days
}

JobDefs {
  Name = "DefaultConsJob"
  Type = Backup
  Client = bareos-fd
  FileSet = "SelfTest"
  Storage = File
  Messages = Standard
  Pool = AI-Consolidated
  Priority = 25
  Write Bootstrap = "/data/bareos/bootstrap/%c.bsr"
  Incremental Backup Pool = AI-Incremental
  Full Backup Pool = AI-Consolidated
  AllowDuplicateJobs = no
  CancelLowerLevelDuplicates = yes
  CancelQueuedDuplicates = yes
  CancelRunningDuplicates = no
  Prune Volumes = yes
  Schedule = "Cons-Sched"
}

Job {
    Name = "Consolidate"
    Type = "Consolidate"
    Accurate = "yes"
    JobDefs = "DefaultConsJob"
}

Job {
    Name = "client-volume1-AI"
    Client = "client"
    JobDefs = "DefaultAIJob"
    FileSet = "client-volume1"
    Storage = "AI-Client-Dir1-File"
    Pool = "AI-Client-Dir1"
    Incremental Backup Pool = "AI-Client-Dir1"
    Full Backup Pool = "Cons-Client-Dir1"
    Virtual Full Backup Pool = "Cons-Client-Dir1"
}

Pool {
  Name = AI-Client-Dir1
  Pool Type = Backup
  Recycle = no
  Auto Prune = yes
  Volume Retention = 60 days
  Label Format = "AI-Client-Dir1-"
  Volume Use Duration = 1 days
  Storage = "AI-Client-Dir1-File"
  Next Pool = "Cons-Client-Dir1"
  Maximum Volume Jobs = 1
}

Pool {
  Name = Cons-Client-Dir1
  Pool Type = Backup
  Recycle = no
  Auto Prune = yes
  Volume Retention = 360 days
  Label Format = "Cons-Client-Dir1-"
  Volume Use Duration = 1 days
  Storage = "Cons-Client-Dir1-File1"
  Storage = "Cons-Client-Dir1-File2"
  Storage = "Cons-Client-Dir1-File3"
  Storage = "Cons-Client-Dir1-File4"
  Maximum Volume Jobs = 1
}
```

All storage configurations are set on a single device, with a maximum of one concurrent job allowed, and each has a distinct media type for the incremental and consolidated backups across 6 volumes.

If you need any other information, I can provide it to you.
Thank you in advance for your guidance and expertise.

Tom
Reply all
Reply to author
Forward
0 new messages