Dear,

    We have a simple bacula configuration.

    And autochanger with two tape device (LTO3 and LTO5), for two kind of different backups. Both work when they run independently. But If  a job on LTO5 is running the other jobs related to LTO3 are waiting until this jobs has finished. This is  a pain in the ass, because the LTO5 run very long unique/full backups (along days) then I found that a couple of incremental backups on LT03 running on the same day. The jobs run on different clients and event sometimes in different catalogs.

       These are the relevant configuration:

bacula-dir.conf

     Director {                            # define myself
       Name = bacula-dir
       DIRport = 9101                # where we listen for UA connections
       QueryFile = "/etc/bacula/scripts/query.sql"
       WorkingDirectory = "/var/lib/bacula"
       PidDirectory = "/var/run/bacula"
       Maximum Concurrent Jobs = 10
       Password = "YYYYYYYYYYYYYY"         # Console password
       Messages = Daemon
       DirAddress = "xxxxxx"
       .....

bacula-sd.conf

 Storage {                             # definition of myself
     Name = bacula-sd
     SDPort = 9103                  # Director's port     
     WorkingDirectory = "/var/lib/bacula"
     Pid Directory = "/var/run/bacula"
     Maximum Concurrent Jobs = 20
     SDAddress = "xxxxx"
     .....

   Autochanger {
      Name = TSM3500
       Device = ULT3580-TD3
      Device = ULT3580-TD5
      Changer Device = /dev/sg7
      Changer Command = "/etc/bacula/scripts/mtx-changer %c %o %S %a %d"
                    # %c = changer device
                    # %o = command (unload|load|loaded|list|slots)
                    # %S = slot index (1-based)
                    # %a = archive device (i.e., /dev/sd* name for tape drive)
                    # %d = drive index (0-based)
}

Device {
    # The TSM3100's first tape drive
    Name = ULT3580-TD3
    Archive Device = /dev/nst0
    Device Type = Tape
    Media Type = LTO-3
    Autochanger = Yes
    # Changer Device = <inherited from Changer>
    Alert Command = "sh -c '/usr/sbin/tapeinfo -f /dev/sg5 | /bin/sed -n /TapeAlert/p"
    Drive Index = 1
    RemovableMedia = yes
    Random Access = no
    Maximum Block Size = 262144
    Maximum Network Buffer Size = 262144
    Maximum Spool Size = 40gb
    Maximum Job Spool Size = 30gb
    Spool Directory = /backup/spool
    AutomaticMount = Yes;
}

Device {
    # The TSM3100's second tape drive
    Name = ULT3580-TD5
    Archive Device = /dev/nst1
    Device Type = Tape
    Media Type = LTO-5
    Autochanger = Yes
    # Changer Device = <inherited from Changer>
    Alert Command = "sh -c '/usr/sbin/tapeinfo -f /dev/sg6 | /bin/sed -n /TapeAlert/p"
    Drive Index = 0
    RemovableMedia = yes
    Random Access = no
    Maximum Block Size = 262144
    Maximum Network Buffer Size = 262144
    Maximum Spool Size = 40gb
    Maximum Job Spool Size = 30gb
    Spool Directory = /backup/spool
    AutomaticMount = Yes;
}

And the storage definitions:

Storage {
  Name = TSM3500-LTO3
  Address = "xxxxxx"
  SDPort = 9103
  Password = "YYYYYYYYYYYYYYYYYYY"
  Device = ULT3580-TD3
  Media Type = LTO-3
  Autochanger = yes
  Maximum Concurrent Jobs = 5
}

Storage {
  Name = TSM3500-LTO5
  Address = "xxxxx"
  SDPort = 9103
  Password = "YYYYYYYYYYYYYYYYYYY"
  Device = ULT3580-TD5
  Media Type = LTO-5
  Autochanger = yes
  Maximum Concurrent Jobs = 5
}

 I think that  "Maximum concurrent Jobs" variable is well defined (well maybe the storage definition level should be 1, I do not see a tape writing jobs in parallel, I checked this and neither work )...but I am not able to find the misconfiguration.

Any Idea ??

Cheers, I.

--
####################################
Iban Cabrillo Bartolome
Instituto de Fisica de Cantabria (IFCA)
Santander, Spain
Tel: +34942200969
####################################
Bertrand Russell:
"El problema con el mundo es que los estúpidos están seguros de todo y los inteligentes están llenos de dudas"