(resending, deleted quoted text from previous messages so my message will
pass the 40kb limit)
Myles,
1. Basically, I suspect rclone filled its cache and bacula stopped the
backup at that time. My guess is that if you were to run a backup of less
than 1GiB right now in bacula, it would succeed. Guaranteed not to fill the
rclone buffer. Maybe adjust rclone cache size or otherwise fiddle with that
portion of it? Definitely exclude the rclone cache (if written to disk)
from the bacula backup!
2. I think there may be a mistake in how your bacula-sd.conf defines the
bacula FileChgr1-Dev1 and FileChgr1-Dev2 devices.
Your bacula virtual devices (FileChgr1-Dev1 and FileChgr1-Dev2) point to
actual files on the system. Mine in the default config (working, backs up
bacula database routinely) points to the folder that holds volume files. In
my configuration the volume filenames are never defined in the FileChgr*
resources.
Did you create these files yourself?
root@c1:~# ls -l /mnt/MylesDearDropBox/Backup/bacula/archive/
total 20971544
-rw-r--r-- 1 root root 21474860756 Dec 4 03:27 MylesMpwrware1
-rw-r--r-- 1 root root 0 Dec 4 03:04 MylesMpwrware1.l
-rw-r--r-- 1 root root 0 Dec 4 01:00 MylesMpwrware2
I would expect bacula to create volume files itself, and that it doesn't
need to be told what the volume names will be (except for "Vol-" defined in
configuration elsewhere, as you've noted).
in your bacula-sd.conf, for both devices (FileChgr1-Dev1 and
FileChgr1-Dev2) I recommend adjusting your ArchiveDevice= line to read the
same:
ArchiveDevice= /mnt/MylesDearDropBox/Backup/bacula/archive # no trailing
foreward slash! "/"
FileChgr1-Dev1 and FileChgr1-Dev2 should both point to the same folder!
There are two such devices defined in part so that one could read
information from a volume and another could write to another volume, such
as in a copy or migration job. There may be other reasons, but in any case
the FileChgr1-Dev* devices should certainly point to the same folder.
In your current configuration, I think the Vol-0001 and Vol-0002 are labels
written to the volumes themselves internally, for bacula's reference. If
these were tapes, bscan or some other bacula tool could read the tapes
(volumes) and find the labels. The label name internally to bacula doesn't
necessarily match the actual filename or label written on the tape.
However, it usually DOES match in a default configuration, like mine. I
think once you make this change to the ArchiveDevice paths and let bacula
create volumes, the bacula labels will match the filenames of the volumes.
In the case of Vol-0002 it might be that bacula intended to write to the
volume but couldn't create it because FileChgr1-Dev1 and FileChgr1-Dev2
both point to actual files and NOT to a folder. Not sure.
For reference, here's what I see on my bacula system. See how I have files
in /opt/bacula/archive/
[root@NSF-rocky]# find / -path /mnt -prune -o -type f -print | grep
"Vol-0"
/opt/bacula/archive/Vol-0001
/opt/bacula/archive/Vol-0040
[root@NSF-rocky EVO-Media]# ls -lah /opt/bacula/archive/
total 15G
drwxrwxr-x+ 3 bacula disk 4.0K Oct 20 12:25 .
drwxrwxr-x+ 11 root root 4.0K Sep 13 14:17 ..
-rw-r-----. 1 bacula disk 1.5G Sep 6 12:25 Vol-0001
-rw-r-----. 1 bacula disk 13G Dec 4 05:45 Vol-0040
in bconsole:
list volumes pool=File
Pool: File
+---------+------------+-----------+---------+----------------+----------+--------------+---------+------+-----------+-----------+---------+----------+---------------------+------------+
| mediaid | volumename | volstatus | enabled | volbytes | volfiles |
volretention | recycle | slot | inchanger | mediatype | voltype | volparts
| lastwritten | expiresin |
+---------+------------+-----------+---------+----------------+----------+--------------+---------+------+-----------+-----------+---------+----------+---------------------+------------+
| 1 | Vol-0001 | Read-Only | 1 | 1,248,082,548 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| 2023-06-27 23:10:02 | 17,729,032 |
| 23 | Vol-0022 | Purged | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 24 | Vol-0024 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 25 | Vol-0025 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 26 | Vol-0026 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 27 | Vol-0027 | Error | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 28 | Vol-0028 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 29 | Vol-0029 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 30 | Vol-0030 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 31 | Vol-0031 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 32 | Vol-0032 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 33 | Vol-0033 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 34 | Vol-0034 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 35 | Vol-0035 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 36 | Vol-0036 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 37 | Vol-0037 | Read-Only | 1 | 0 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| | 0 |
| 38 | Vol-0038 | Error | 1 | 15,246,254 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| 2023-09-13 14:10:21 | 24,435,851 |
| 39 | Vol-0039 | Error | 1 | 15,264,103 | 0 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| 2023-09-13 14:28:33 | 24,436,943 |
| 40 | Vol-0040 | Append | 1 | 13,471,721,811 | 3 |
31,536,000 | 1 | 0 | 0 | File1 | 1 | 0
| 2023-12-04 05:45:12 | 31,493,942 |
+---------+------------+-----------+---------+----------------+----------+--------------+---------+------+-----------+-----------+---------+----------+---------------------+------------+
(I have migrated from another server to this one, reinstalled bacula
improperly at first, and had filesystem permissions issues in a previous
instance of bacula, hence volumes 23-39, which were created automatically
as I struggled to correct the issue, ultimately reinstalling bacula,
correctly this time)
snippets from my bacula-sd.conf
Autochanger {
Name = "FileChgr1"
Device = "FileChgr1-Dev1"
Device = "FileChgr1-Dev2"
ChangerDevice = "/dev/null"
ChangerCommand = ""
}
Device {
Name = "FileChgr1-Dev1"
MediaType = "File1"
ArchiveDevice = "/opt/bacula/archive"
RemovableMedia = no
RandomAccess = yes
AutomaticMount = yes
LabelMedia = yes
AlwaysOpen = no
MaximumConcurrentJobs = 5
}
Device {
Name = "FileChgr1-Dev2"
MediaType = "File1"
ArchiveDevice = "/opt/bacula/archive"
RemovableMedia = no
RandomAccess = yes
AutomaticMount = yes
LabelMedia = yes
AlwaysOpen = no
MaximumConcurrentJobs = 5
}
Now here is my bacula-sd.conf referring to my Qualstar Q24 tape autochanger
and my LTO-8 drive (installed inside autochanger)
Autochanger {
Name = "Q24"
Device = "drive0"
ChangerDevice =
"/dev/tape/by-id/scsi-1BDT_FlexStor_II_00DE64107417_LL0-changer"
ChangerCommand = "/opt/bacula/scripts/mtx-changer %c %o %S %a %d"
}
Device {
Name = "drive0"
MediaType = "LTO-8"
ArchiveDevice = "/dev/tape/by-id/scsi-1097013295-nst"
RemovableMedia = yes
RandomAccess = no
AutomaticMount = yes
AlwaysOpen = no
Autochanger = yes
DriveIndex = 0
SpoolDirectory = /mnt/spool
MaximumFileSize = 75G
# Minimum Block Size = 1024
# Maximum Block Size = 1024
# Maximum Network Buffer Size = 65536
}
The biggest difference is that the LTO-8 drive "drive0" points to a system
device file, matching actual hardware. the FileChgr1-Dev1 and
FileChgr1-Dev2 devices point to the same folder(/opt/bacula/archive), which
as you can see holds two volume files Vol-0001 and Vol-0040. My point in
showing this to you is to illustrate how bacula "thinks" of disk volumes as
"tapes". Configuration is very similar.
Regards,
Robert Gerber
402-237-8692
r...@craeon.net
>
>>
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
