Title: Problem with concurrent jobs, pools, volumes.... ;-)

All right, that's maybe a bit long and problem most is that i still
don't really understand how all configuration fit together, but maybe
you've some time to help...  ;-))

all my backups will go to disk (finally convinced my boss to buy two
big RAIDs :-)) and i would have single backup-file for each backup run


what i did:

- setup pools as in the example Kern gave in "Using Pools to Manage Volumes"
  with some changes: volume retention for full is only 2 month. Another thing
  i would like was to have uniqe name for the volumes created according to
  the job-name (is this a good idea?)

- created unique schedules for each client (also not sure if that's a good idea)
  as shown in the attached configs (i currently have 5 clients for testing, all
  are created like the one attached)

- create some jobs that should be run "almost" parallel, but will be started
  with some minutes delay (as you remember i had some trobles when running
  jobs in "real" parallel).


now my problems...

when starting some jobs manually the jobs won't run in parallel:

Running Jobs:
 JobId Level   Name                       Status
======================================================================
     7 Increme  sunserv-data.2005-05-02_08.09.35 is waiting on max Storage jobs
     6 Increme  sunserv-system.2005-05-02_08.09.32 is waiting execution
     5 Full    sonne-data.2005-05-02_08.09.03 is running
     4 Increme  sonne-system.2005-05-02_08.08.59 is waiting for higher priority jobs to finish
     3 Full    atpc4ujc-data.2005-05-02_08.06.14 is running
====

sonne-data, sunserv-data and sunserv-system will all use the same pool and same storage
but why are the "sunserv"-jobs waiting for sonne to complete?


last weekend, when i first tried this configuration, the problem was that almost all of
the backups didn't run because the jobs were waiting for volume to be labeled (but some
of them run and also created volumes automatically!)!!  funny, isn't it?


so, sorry if these is really long to read and maybe really dumb...

but... THANK's a lot for your help!!
Christian



here's part of my configuration:

bacula-dir.conf:

# Pool for Full-Backups
#
Pool {
  Name = Full-Pool
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Volume Retention = 2 months
  Accept Any Volume = yes
  Maximum Volume Jobs = 1
  Label Format = "Full-${Job}-${NumVols:p/4/0/r}"
}

# Pool for differential Backups
#
Pool {
  Name = Diff-Pool
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Volume Retention = 40 days
  Accept Any Volume = yes
  Maximum Volume Jobs = 1
  Label Format = "Diff-${Job}-${NumVols:p/4/0/r}"
}

# Pool for daily incremental Backups
#
Pool {
  Name = Incr-Pool
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Volume Retention = 20 days
  Accept Any Volume = yes
  Maximum Volume Jobs = 1
  Label Format = "Incr-${Job}-${NumVols:p/4/0/r}"
}


# Storage for Unix-Server (SUNs)
#
Storage {
  Name = UnixServer
  Address = atpcc7fc
  SDPort = 9103
  Password = "XXXX"
  Device = UnixServer-FileStore
  Media Type = File
  Maximum Concurrent Jobs = 10
}

# Storage for Windows-Server
#
Storage {
  Name = WinPC
  Address = atpcc7fc
  SDPort = 9103
  Password = "XXXX"
  Device = WinPC-FileStore
  Media Type = File
  Maximum Concurrent Jobs = 10
}


## i've unique schedules for each client that i'll backup,
##  here's one of them...
#
Schedule {
  Name = "Atpc4ujcSystem"
  Run = Full 1st sat at 03:15
  Run = Full 3rd sat at 03:15
  Run = Differential 2nd sat at 03:15
  Run = Differential 4th-5th sat at 03:15
  Run = Incremental sun-fri at 03:15
}
Schedule {
  Name = "Atpc4ujcWeb"
  Run = Full 1st sat at 03:20
  Run = Full 3rd sat at 03:20
  Run = Differential 2nd sat at 03:20
  Run = Differential 4th-5th sat at 03:20
  Run = Incremental sun-fri at 03:20
}
Schedule {
  Name = "Atpc4ujcData"
  Run = Full 1st sat at 03:25
  Run = Full 3rd sat at 03:25
  Run = Differential 2nd sat at 03:25
  Run = Differential 4th-5th sat at 03:25
  Run = Incremental sun-fri at 03:25
}

# Client (File Services) to backup
#

Client {
  Name = atpc4ujc-fd
  Address = atpc4ujc
  FDPort = 9102
  Catalog = DefaultCatalog
  Password = "XXXX"
  File Retention = 30 days
  Job Retention = 2 month
  AutoPrune = yes
  Maximum Concurrent Jobs = 5
}


## and here some jobs for this client
##
Job {
  Name = "atpc4ujc-system"
  Client = atpc4ujc-fd
  Type = Backup
  FileSet = "atpc4ujc-system"
  Schedule = "Atpc4ujcSystem"
  Storage = WinPC
  Pool = Default
  Full Backup Pool = Full-Pool
  Differential Backup Pool = Diff-Pool
  Incremental Backup Pool = Incr-Pool
  Messages = Standard
  Write Bootstrap = "/var/bacula/atpc4ujc-system.bsr"
  Priority = 8
}

Job {
  Name = "atpc4ujc-web"
  Client = atpc4ujc-fd
  Type = Backup
  FileSet = "atpc4ujc-web"
  Schedule = "Atpc4ujcWeb"
  Storage = WinPC
  Pool = Default
  Full Backup Pool = Full-Pool
  Differential Backup Pool = Diff-Pool
  Incremental Backup Pool = Incr-Pool
  Messages = Standard
  Write Bootstrap = "/var/bacula/atpc4ujc-web.bsr"
  Priority = 5
}



bacula-sd.conf:

Storage {
  Name = atpcc7fc-sd
  SDPort = 9103
  WorkingDirectory = "/var/bacula"
  Pid Directory = "/var/run"
  Maximum Concurrent Jobs = 10
}

Device {
  Name = UnixServer-FileStore
  Media Type = File
  Archive Device = /Backup2/UnixServer
  LabelMedia = yes
  Random Access = Yes
  AutomaticMount = yes
  RemovableMedia = no
  AlwaysOpen = no
}

Device {
  Name = WinPC-FileStore
  Media Type = File
  Archive Device = /Backup1/WinPC
  LabelMedia = yes
  Random Access = Yes
  AutomaticMount = yes
  RemovableMedia = no
  AlwaysOpen = no
}

Reply via email to