NDMP trying to use wrong device - unable to write to tape

[Kalle Kadakas]

Greetings! I hope that there is someone wiser out there who could help a man out.

I am trying to set up BareOS with NDMP to communicate with NetApp Storage which has a HP MSL G3 Series tape library attached to it. NDMP tests are successful, the device is recognized and I am able to label the LTO tapes.

However, whenever I try to make an actual backup of a volume, the process fails:

06-Feb 10:00 bareos-dir JobId 274: Start NDMP Backup JobId 274, Job=ndmp-backup-job.2020-02-06_10.00.53_37

06-Feb 10:01 bareos-dir JobId 274: NDMP Devices for storage storage-backup0:(HP      MSL G3 Series   7.00)

06-Feb 10:01 bareos-dir JobId 274:  nrst1l

06-Feb 10:01 bareos-dir JobId 274: successfully reserved NDMP Tape Device nrst1l for job 274

06-Feb 10:01 bareos-dir JobId 274: Using Data  host storage-backup0-1.big.local

06-Feb 10:01 bareos-dir JobId 274: Using Tape  host:device:address  storage-backup0-1.big.local:nrst1l:@1

06-Feb 10:01 bareos-dir JobId 274: Using Robot host:device(ident)  storage-backup0-1.big.local:mc0(HP      MSL G3 Series   7.00)

06-Feb 10:01 bareos-dir JobId 274: Using Tape record size 64512

06-Feb 10:01 bareos-dir JobId 274: Async request NDMP4_LOG_MESSAGE

06-Feb 10:01 bareos-dir JobId 274: LOG_MESSAGE: 'Session identifier: 33784'

06-Feb 10:01 bareos-dir JobId 274: ndmca_connect_tape_agent: 0 0x7f9810006a00

06-Feb 10:01 bareos-dir JobId 274: Async request NDMP4_LOG_MESSAGE

06-Feb 10:01 bareos-dir JobId 274: LOG_MESSAGE: 'Session identifier: 33799'

06-Feb 10:01 bareos-dir JobId 274: Error finding volume for append

06-Feb 10:01 bareos-dir JobId 274: removed reservation of NDMP Tape Device nrst1l for job 274

06-Feb 10:01 bareos-dir JobId 274: Error: Bareos bareos-dir 18.2.5 (30Jan19):

  Build OS:               Linux-4.4.92-6.18-default redhat CentOS Linux release 7.6.1810 (Core) 

  JobId:                  274

  Job:                    ndmp-backup-job.2020-02-06_10.00.53_37

  Backup Level:           Full

  Client:                 "storage-backup0-1" 

  FileSet:                "Trident" 2020-02-04 14:19:39

  Pool:                   "ndmp-tape" (From Job resource)

  Catalog:                "MyCatalog" (From Client resource)

  Storage:                "storage-backup0" (From Job resource)

  Scheduled time:         06-Feb-2020 10:00:38

  Start time:             06-Feb-2020 10:00:55

  End time:               06-Feb-2020 10:01:14

  Elapsed time:           19 secs

  Priority:               10

  NDMP Files Written:     0

  NDMP Bytes Written:     0 (0 B)

  Rate:                   0.0 KB/s

  Volume name(s):         

  Volume Session Id:      0

  Volume Session Time:    0

  Last Volume Bytes:      0 (0 B)

  Bareos binary info:     bareos.org build: Get official binaries and vendor support on bareos.com

  Termination:            *** Backup Error ***

The error would initially suggest that the issue is with finding available volumes/tapes to append the backup to, but that is not the case:

get ndmp_vol_list...

 Slot |   Volume Name    |   Status  |  Media Type    |         Pool             |

------+------------------+-----------+----------------+--------------------------|

    1*|         CLNU30L1 |         ? |              ? |                        ? |

... ... ... ...

   20 |         000064L6 |    Append |      NDMP-Tape |                     Full |

What I rather noticed was that BareOS is trying to use the nrst1l device to write the backup which is a LTO-4/5 native density device but the tape itself is LTO-6. It should instead be using nrst1h or nrst1a devices, which are meant for LTO-6. At least this is the only real issue that I am able to find.

  tape Hewlett-Packard LTO-6

    device     nrst1l

      attr       0x4

      set        WORLD_WIDE_NAME=WWN[5:001:438004:d36cd1]

      set        ELECTRICAL_NAME=0d.1

      set        DENSITY=LTO-4/5 Native Density  

      set        ALIAS 0=st1

      set        ALIAS 1=st21

    device     nrst1m

      attr       0x4

      set        WORLD_WIDE_NAME=WWN[5:001:438004:d36cd1]

      set        ELECTRICAL_NAME=0d.1

      set        DENSITY=LTO-4/5 Compressed      

      set        ALIAS 0=st1

      set        ALIAS 1=st21

    device     nrst1h

      attr       0x4

      set        WORLD_WIDE_NAME=WWN[5:001:438004:d36cd1]

      set        ELECTRICAL_NAME=0d.1

      set        DENSITY=LTO-6 2.5TB             

      set        ALIAS 0=st1

      set        ALIAS 1=st21

    device     nrst1a

      attr       0x4

      set        WORLD_WIDE_NAME=WWN[5:001:438004:d36cd1]

      set        ELECTRICAL_NAME=0d.1

      set        DENSITY=LTO-6 6.25TB Compressed 

      set        ALIAS 0=st1

      set        ALIAS 1=st21

I did manually define the device in Storage configuration, but that does not seem to have any effect on this even after reloading/rebooting. Here is the full configuration:

Job {

  Name          = "ndmp-backup-job"

  type          = backup

  protocol      = NDMP_NATIVE

  level         = full

  client        = storage-backup0-1

  storage       = storage-backup0

  backup format = dump

  messages      = Standard

  Pool          = ndmp-tape

  FileSet       = "Trident"

  Save File History = no

}

Client {

  Name = storage-backup0-1

  Address = storage-backup0-1.my.domain

  Port = 10000

  Protocol = NDMPv4

  Auth Type = MD5

  Username = "ndmpadmin"

  Password = "my_password"

  Maximum Concurrent Jobs = 1

  NDMP Log Level = 7

}

Fileset {

  Name = "Trident"

  Include {

    Options {

      meta = "DEBUG=Y"

      meta = "VERBOSE=Y"

      meta = "TYPE=dump"

      meta = "FILESYSTEM=/backup"

      meta = "LEVEL=0"

#      meta = "HIST=F"

#      meta = "DIRECT=Y"

#      meta = "RECURSIVE=Y"

#      meta = "BUTYPE=DUMP"

    }

    File = /backup/trident_qtree_pool_trident_SOLKGJVXRJ_dest

  }

}

Storage {

  Name = storage-backup0    

  Address = storage-backup0-1.my.domain                # N.B. Use a fully qualified name here (do not use "localhost" here).

  Port = 10000

  Protocol = NDMPv4

  Auth Type = MD5

  Username = "ndmpadmin"

  Password = "my_password"

  Maximum Concurrent Jobs = 1

  Autochanger = yes

  MediaType = NDMP-Tape

  Device = nrst1a

#  Device = st21

  NDMP Changer Device = mc0

}

Pool {

  Name = ndmp-tape

  Pool Type = Backup

  Recycle = yes                       # Bareos can automatically recycle Volumes

  AutoPrune = yes                     # Prune expired volumes

  Volume Retention = 1 min         # How long should the Full Backups be kept? (#06)

#  Scratch Pool = Scratch

#  Maximum Volume Bytes = 50G          # Limit Volume size to something reasonable

#  Maximum Volumes = 100               # Limit number of Volumes in Pool

#  Label Format = "Full-"              # Volumes will be labeled "Full-<volume-id>"

}

If anyone has any thoughts on this, any help is much appreciated!

Best regards,

Kalle