I have an annoying issue that I don't yet understand
I read the howtos around volumes and media type etc but so far couldn't
get it working:
There are 2 storages "File" and "HP-Autoloader" ... so disk- and
tape-based backups.
There are pools like "daily" "weekly" "monthly" that contain physical tapes.
"daily" and "weekly" have labels like "CMR123L6" (side question: I'd
like to configure that as well into the pools. What would be the
LabelFormat here?)
"monthly" has "ARC123L6"
and the file-based volumes get LabelFormat = "Vol-"
The physical tapes use
MediaType = "LTO-6"
the disk-based virtual tapes use
MediaType = "File"
Now where´s the problem?
It seems that the Job "BackupCatalog" (sometimes?) creates a new volume
like "Vol-252", but with MediaType "LTO-6", which is then NOT FOUND in
the tape autoloader. Sure, this one doesn't exist. "Vol-" volumes are
for storage "File" only.
So the whole queue of Jobs gets stuck because of this.
I read that Director and Storage Daemon somehow have to agree on Media
Types etc . I have:
## sd
Device {
Name = "FileStorage"
MediaType = "File"
ArchiveDevice = "/mnt/backup/bacula"
RemovableMedia = no
RandomAccess = yes
AutomaticMount = yes
LabelMedia = yes
AlwaysOpen = no
SpoolDirectory = "/mnt/amhold/bacula"
MaximumSpoolSize = 80000000000
}
Device {
Name = "HP-Ultrium"
MediaType = "LTO-6"
ArchiveDevice = "/dev/nst0"
RemovableMedia = yes
RandomAccess = no
LabelMedia = yes
AlwaysOpen = yes
Autochanger = yes
MaximumBlockSize = 262144
MaximumFileSize = 8589934592
SpoolDirectory = "/mnt/amhold/bacula"
MaximumSpoolSize = 1000000000000
}
## director
Storage {
Name = "File"
SdPort = 9103
Address = "samba"
Password = "xxx"
Device = "FileStorage"
MediaType = "File"
}
Storage {
Name = "HP-Autoloader"
SdPort = 9103
Address = "samba"
Password = "xxx"
Device = "HP-Autoloader"
MediaType = "LTO-6"
Autochanger = "HP-Autoloader"
MaximumConcurrentJobs = 1
}
# pools
Pool {
Name = "Daily"
Description = "daily backups"
PoolType = "Backup"
MaximumVolumes = 30
VolumeRetention = 864000
Storage = "HP-Autoloader"
}
Pool {
Name = "Default"
PoolType = "Backup"
MaximumVolumes = 100
MaximumVolumeBytes = 53687091200
VolumeRetention = 31536000
AutoPrune = yes
Recycle = yes
}
Pool {
Name = "File"
PoolType = "Backup"
LabelFormat = "Vol-"
MaximumVolumes = 101
MaximumVolumeBytes = 53687091200
VolumeRetention = 432000
NextPool = "Daily"
Storage = "File"
AutoPrune = yes
Recycle = yes
}
Pool {
Name = "Monthly"
Description = "monthly backups"
PoolType = "Backup"
UseVolumeOnce = no
Storage = "HP-Autoloader"
}
Pool {
Name = "Scratch"
PoolType = "Backup"
}
Pool {
Name = "Weekly"
Description = "weekly backups"
PoolType = "Backup"
UseVolumeOnce = no
VolumeRetention = 2592000
Storage = "HP-Autoloader"
}
# Job
Job {
Name = "BackupCatalog"
Level = "Full"
Storage = "HP-Autoloader"
Fileset = "Catalog"
Schedule = "WeeklyCycleAfterBackup"
JobDefs = "DefaultJob"
WriteBootstrap = "/opt/bacula/working/%n.bsr"
Runscript {
RunsOnClient = no
RunsWhen = "Before"
Command = "/opt/bacula/scripts/make_catalog_backup.pl MyCatalog"
}
Runscript {
RunsOnClient = no
RunsWhen = "After"
Command = "/opt/bacula/scripts/delete_catalog_backup"
}
Priority = 11
Pool = "Daily"
}
Anything missing? I get a bit lost here and don't want to
"overconfigure" things where there might be clever defaults or automatisms.
addon: for weekly and monthly the catalog should run to the according
tapes, not to a "daily" tape, right?
Somewhere I read the catalog should (additionally?) be written to File,
that would complicate it even more.
Thanks in advance for any help here. It seems that is the missing bit
for this config to be running autonomously for a week or so (which isn't
the case yet unfortunately ...)
Stefan
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
