Hello,

I apologize if this has been answered, but after googling I’m having

trouble finding something like this particular case. We have an ML6000

autochanger with two LTO-6 drives. For a long time we were just using one

pool, but we had two SSD disks (for spooling) and two tape drives, so we

ordered some new tapes and set up a second pool. It seemed like it could

be a good thing, and it worked for over a month... then it suddenly and

mysteriously stopped working when the new pool's tapes came back around

again. I'll give some configuration below, but here is basically how it

works:

Pool1 - tapes 000101-000140

Pool4 - tapes 000401-000420

None of these are newly labeled. They have all been through successful

backup rotations. It was all working.

Then a large backup had the status:

is waiting for an appendable Volume

In the Job definition:

  JobDefs = "Pool4Job"

which is assigned to Pool4.

+------------+-------+-------+------+-----------+---------------------+

| volumename | pool  | gb    | slot | volstatus | lastwritten         |

+------------+-------+-------+------+-----------+---------------------+

| 000401     | Pool4 | 2,439 |    3 | Full      | 2016-08-23 20:36:42 |

| 000402     | Pool4 | 2,460 |    5 | Purged    | 2016-07-07 01:21:20 |

| 000403     | Pool4 | 2,927 |    8 | Purged    | 2016-07-09 04:51:54 |

| 000404     | Pool4 | 2,502 |    9 | Purged    | 2016-07-10 22:23:54 |

| 000405     | Pool4 | 2,551 |   10 | Purged    | 2016-07-06 01:08:40 |

There are several volumes in Pool4 in the autochanger, and they have

already been written to during the last rotation. However:

Device tape: "Tapedrv-1" (/dev/nst1) is not open.

    Device is BLOCKED waiting to create a volume for:

       Pool:        Pool4

       Media type:  LTO-6

    Drive 0 is not loaded.

I also see a message:

Please use the "label" command to create a new Volume for:

    Storage:      "Tapedrv-1" (/dev/nst1)

    Pool:         Pool4

    Media type:   LTO-6

bconsole seems to agree that there are Pool4 tapes in the autochanger with

status Purged, so I can’t figure out why bacula won’t use them anymore.

After I wrote all of that, as a drastic measure I deleted and relabeled

those tapes. They worked fine, and jobs assigned to Pool4Job were writing

to them. Then the next batch came back into rotation, and it happened all

over again. I tried mounting the volumes, but bacula wouldn't do it.

At this point, I'm not sure what to look at. It's possible that we're

missing something very simple and obvious. Any suggestions?

Thanks for any help or advice that you have,

Mark

Also, Here are selections from configuration files in case that’s useful:

JobDefs {

  Name = "Pool4Job"

  Type = Backup

  Level = Incremental

  Client = (hostname)-fd

  FileSet = "Full Set"

  Schedule = "WeeklyCycle"

  Storage = "autochanger"

  Messages = Standard

  Pool = Pool4

  SpoolAttributes = yes

  SpoolData = yes

  Priority = 10

  Write Bootstrap = "/bacula/working/%c.bsr"

}

Pool {

  Name = Pool1

  Pool Type = Backup

  Recycle = no                       # Bacula can automatically recycle

Volumes

  AutoPrune = no                     # Prune expired volumes

  Volume Retention = 40 days

  Maximum Volumes = 100               # Limit number of Volumes in Pool

}

Pool {

  Name = Pool4

  Pool Type = Backup

  Recycle = no                       # Bacula can automatically recycle

Volumes

  AutoPrune = no                     # Prune expired volumes

  Volume Retention = 40 days

  Maximum Volumes = 100               # Limit number of Volumes in Pool

}

Autochanger {

  Name = "changer3"

  Device = Tapedrv-1,Tapedrv-2

  Changer Command = "/bacula/etc/mtx-changer %c %o %S %a %d"

  Changer Device = /dev/sg5

}

Device {

  Name = Tapedrv-1

  Drive Index = 0

  Media Type = LTO-6

  Archive Device = /dev/nst0

  AutomaticMount = yes;

  AlwaysOpen = yes;

  RemovableMedia = yes;

  RandomAccess = no;

  AutoChanger = yes

  Spool Directory = /bspool1

  Maximum Spool Size = 700G

  Maximum File Size = 5G

  Alert Command = "sh -c 'smartctl -H -l error %c'"

}

Device {

  Name = Tapedrv-2

  Drive Index = 1

  Media Type = LTO-6

  Archive Device = /dev/nst1

  AutomaticMount = yes;

  AlwaysOpen = yes;

  RemovableMedia = yes;

  RandomAccess = no;

  AutoChanger = yes

  Spool Directory = /bspool2

  Maximum Spool Size = 700G

  Maximum File Size = 5G

  Alert Command = "sh -c 'smartctl -H -l error %c'"

}

------------------------------------------------------------------------------

_______________________________________________

Bacula-users mailing list

Bacula-users AT lists.sourceforge DOT net

https://lists.sourceforge.net/lists/listinfo/bacula-users