Problem solved!
Steve Ellis pointed that there was a problem with my stinit command.
As you'll all notice below (my original message/post), the first output of
stinit is identifying /dev/nst0 and /dev/nst0l as mode 0 and 1, respectively.
But this output is not correct, it's a bug in version 0.9b (which is part of
Redhat and CentOS distributions, as well as others). Stinit shows that nst0 is
mode 0, but this device is actually using mode 1 (with compression enabled).
The device I was using to run backups (/dev/nst0l), was not really operating in
mode 1 (with compression), but in mode 2 (without compression), hence all the
problem.
Looks like the problem was filed as a bug by Debian users, but it clearly
affects other distributions. Here is the URL (thanks again to Steve):
http://www.mail-archive.com/debian-bugs-dist AT lists.debian DOT
org/msg379503.html
After removing mt-st version 0.9b, getting version 1.1 source code (there is no
rpm package of this version for Redhat/CentOS), compiling and installing it,
then stinit started to output the correct information:
[root@br01 ~]# stinit -v -v
Trying to open database '/etc/stinit.def'.
Open succeeded.
stinit, processing tape 0
Mode 1, name '/dev/nst0'
Mode 2, name '/dev/nst0l'
Mode 3, name '/dev/nst0m'
Mode 4, name '/dev/nst0a'
I know this is an operating system problem, but it would be helpful to have it
documented somewhere on bacula's website (maybe Wiki?) since it could be
affecting other RedHat/CentOS users.
AFAIK mt-st 1.1 will be part of Redhat version 6.
Cheers,
Rodrigo.
-----Original Message-----
From: Rodrigo Ferraz
Sent: Wednesday, September 01, 2010 9:23 AM
To: bacula-users AT lists.sourceforge DOT net
Cc: Suporte Concept Netservices
Subject: LTO-3 tape not compressing data/premature end of tape space
Hello
We've been facing this problem for a while and unfortunately we are still
unable to find an effective solution, so any help would be greatly appreciated.
The problem: bacula is only filling LTO-3 until exactly 394.38 GB and not going
any further a single byte. Looks like tape compression is not taking place.
Bconsole shows this error below and then keeps asking to mount another volume:
01-Sep 00:57 br01 JobId 546: End of medium on Volume "S002"
Bytes=423,463,799,808 Blocks=6,564,108 at 01-Sep-2010 00:57.
This is tapeinfo output:
[root@br01 ~]# tapeinfo -f /dev/sg1
Product Type: Tape Drive
Vendor ID: 'IBM '
Product ID: 'ULTRIUM-HH3 '
Revision: '93G7'
Attached Changer: No
SerialNumber: '1H10077966'
MinBlock:1
MaxBlock:16777215
SCSI ID: 6
SCSI LUN: 0
Ready: yes
BufferedMode: yes
Medium Type: 0x38
Density Code: 0x44
BlockSize: 0
DataCompEnabled: yes
DataCompCapable: yes
DataDeCompEnabled: yes
CompType: 0x1
DeCompType: 0x1
BOP: yes
Block Position: 0
[root@br01 ~]#
This is mt status:
[root@br01 ~]# mt -f /dev/nst0l status
SCSI 2 tape drive:
File number=0, block number=0, partition=0.
Tape block size 0 bytes. Density code 0x44 (no translation).
Soft error count since last status=0
General status bits on (41010000):
BOT ONLINE IM_REP_EN
[root@br01 ~]#
And this is bacula-sd configuration:
[root@br01 ~]# cat /etc/bacula/bacula-sd.conf Director {
Name = br01
Password = "abadede"
}
Storage {
Name = br01
SDPort = 9103
WorkingDirectory = "/var/lib/bacula"
Pid Directory = "/var/run"
Maximum Concurrent Jobs = 1
}
Device {
Name = LTO-3
Drive Index = 0
Autochanger = no
Archive Device = /dev/nst0l
Maximum Changer Wait = 3d
AutomaticMount = yes
AlwaysOpen = yes
Media Type = LTO-3
RemovableMedia = yes
Offline On Unmount = no
Spool Directory = /var/lib/bacula/spool
Maximum Spool Size = 35gb
RandomAccess = no
LabelMedia = yes
Hardware End of Medium = No
Fast Forward Space File = No
Two EOF = no
BSF at EOM = no
}
Messages {
Name = Standard
director = br01 = all
}
[root@br01 ~]#
As you'll notice we are using the mode1 device (nst0l), with compression
enabled so that we can go further than 400 GB. As per stinit:
[root@br01 ~]# stinit -v -v
Trying to open database '/etc/stinit.def'.
Open succeeded.
stinit, processing tape 0
Mode 0, name '/dev/nst0'
Mode 1, name '/dev/nst0l'
Mode 2, name '/dev/nst0m'
Mode 3, name '/dev/nst0a'
The manufacturer is 'IBM', product is 'ULTRIUM-HH3', and revision '93G7'.
Mode 1 definition: scsi2logical=1 can-bsr=1 auto-lock=0 two-fms=0
drive-buffering=1 buffer-writes read-ahead=1 async-writes=1 can-partitions=0
fast-mteom=1 timeout=800 long-timeout=14400 blocksize=0 density=0x44
compression=1
Mode 2 definition: scsi2logical=1 can-bsr=1 auto-lock=0 two-fms=0
drive-buffering=1 buffer-writes read-ahead=1 async-writes=1 can-partitions=0
fast-mteom=1 timeout=800 long-timeout=14400 blocksize=0 density=0x44
compression=0
Mode 3 definition: scsi2logical=1 can-bsr=1 auto-lock=0 two-fms=0
drive-buffering=1 buffer-writes read-ahead=1 async-writes=1 can-partitions=0
fast-mteom=1 timeout=800 long-timeout=14400 blocksize=0 density=0x42
compression=1
Mode 4 definition: scsi2logical=1 can-bsr=1 auto-lock=0 two-fms=0
drive-buffering=1 buffer-writes read-ahead=1 async-writes=1 can-partitions=0
fast-mteom=1 timeout=800 long-timeout=14400 blocksize=0 density=0x40
compression=1
stinit, processing tape 1
Initialized 1 tape device.
[root@br01 ~]#
Other information:
Bacula Version: 5.0.1 (24 February 2010)
Tape: Dell PowerVault LTO3-060 (IBM Ultrium HH3)
O.S.: Vmware ESXi 4 (fully patched) and tape configured in pass-through mode
Guest O.S.: CentOS 5.5
Btape tests work fine, without any error.
Thanks.
Rodrigo Ferraz
CONCEPT NETSERVICES
www.conceptnetservices.com
------------------------------------------------------------------------------
This SF.net Dev2Dev email is sponsored by:
Show off your parallel programming skills.
Enter the Intel(R) Threading Challenge 2010.
http://p.sf.net/sfu/intel-thread-sfd
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users
|