Bacula 5.2.6 on Ubuntu 10.0.4.4 x64
Hi,
I am trying to troubleshoot why I suddenly have block size problems.
The tape drive passes HP diagnostics. It is supposed to support up to a
16M block size. I have never used or configured stinit.def. Tape drive
block size set to variable (0).
I have used 256K blocks up until now. Block sizes up to 214016 work,
but above 215040 and above fail.
configure looks like this:
basedir="/opt/bacula-5.2.6"
workindir="$basedir/working"
make distclean
CFLAGS="-g -O2 -Wall" \
./configure \
--sbindir=$basedir/bin \
--sysconfdir=$basedir/etc \
--mandir=$basedir/bin \
--with-pid-dir=$workingdir \
--with-subsys-dir=$workingdir \
--with-working-dir=$workindir \
--with-scriptdir=$basedir/bin \
--enable-smartalloc \
--enable-batch-insert \
--enable-largefile \
--disable-ipv6 \
--with-openssl \
--with-mysql
exit 0
bacula-sd.conf looks like this:
Device {
Name = Tape
Drive Index = 0
Device Type = Tape
Archive Device = /dev/nst0
Automatic Mount = yes
Removable Media = yes
Random Access = no
Media Type = Tape
Autochanger = no
Auto Select = yes
Always Open = yes
Minimum Block Size = 214016
#Minimum Block Size = 215040
Maximum Block Size = 214016
#Maximum Block Size = 215040
Maximum File Size = 4G
Maximum Spool Size = 80G
Maximum Job Spool Size = 10G
Maximum Network Buffer Size = 65536
Spool Directory = /data/spool/
}
Below is partial working and failing btape output.
** Running btape with debug on the working 214016 block looks like this: **
root@bacula:/opt/bacula/bin# ./btape -d200 -c ../etc/bacula-sd.conf Tape
Tape block granularity is 1024 bytes.
btape: stored_conf.c:704-0 Inserting director res: bacula-mon
btape: stored_conf.c:704-0 Inserting device res: Tape
btape: butil.c:290 Using device: "Tape" for writing.
btape: dev.c:311-0 init_dev: tape=1 dev_name=/dev/nst0
btape: device.c:277-0 start open_output_device()
btape: device.c:296-0 Opening device.
btape: dev.c:376-0 open dev: type=2 dev_name="Tape" (/dev/nst0) vol=
mode=OPEN_READ_ONLY
btape: dev.c:1896-0 Enter mount
btape: dev.c:426-0 Open dev: device is tape
btape: dev.c:441-0 Try open "Tape" (/dev/nst0) mode=OPEN_READ_ONLY
btape: dev.c:463-0 Rewind after open
btape: dev.c:2392-0 In set_os_device_parameters
btape: dev.c:2414-0 MTSETDRVBUFFER
btape: dev.c:515-0 open dev: tape 3 opened
btape: dev.c:389-0 preserve=0x0 fd=3
btape: device.c:302-0 open dev "Tape" (/dev/nst0) OK
btape: btape.c:471-0 Opening device
btape: dev.c:366-0 Close fd for mode change.
btape: dev.c:376-0 open dev: type=2 dev_name="Tape" (/dev/nst0) vol=
mode=OPEN_READ_WRITE
btape: dev.c:1896-0 Enter mount
btape: dev.c:426-0 Open dev: device is tape
btape: dev.c:441-0 Try open "Tape" (/dev/nst0) mode=OPEN_READ_WRITE
btape: dev.c:463-0 Rewind after open
btape: dev.c:2392-0 In set_os_device_parameters
btape: dev.c:2414-0 MTSETDRVBUFFER
btape: dev.c:515-0 open dev: tape 3 opened
btape: dev.c:389-0 preserve=0x0 fd=3
btape: btape.c:477 open device "Tape" (/dev/nst0): OK
btape: btape.c:311-0 Do tape commands
*test
=== Write, rewind, and re-read test ===
I'm going to write 10000 records and an EOF
then write 10000 records and an EOF, then rewind,
and re-read the data to verify that it is correct.
This is an *essential* feature ...
btape: btape.c:1155 Wrote 10000 blocks of 213916 bytes.
btape: dev.c:1524-0 === weof_dev="Tape" (/dev/nst0)
btape: btape.c:609 Wrote 1 EOF to "Tape" (/dev/nst0)
btape: btape.c:1171 Wrote 10000 blocks of 213916 bytes.
btape: dev.c:1524-0 === weof_dev="Tape" (/dev/nst0)
btape: btape.c:609 Wrote 1 EOF to "Tape" (/dev/nst0)
btape: btape.c:1213 Rewind OK.
10000 blocks re-read correctly.
Got EOF on tape.
10000 blocks re-read correctly.
=== Test Succeeded. End Write, rewind, and re-read test ===
** Running btape with debug on the failing 215040 block looks like this: **
root@bacula:/opt/bacula/bin# ./btape -d200 -c ../etc/bacula-sd.conf Tape
Tape block granularity is 1024 bytes.
btape: stored_conf.c:704-0 Inserting director res: bacula-mon
btape: stored_conf.c:704-0 Inserting device res: Tape
btape: butil.c:290 Using device: "Tape" for writing.
btape: dev.c:311-0 init_dev: tape=1 dev_name=/dev/nst0
btape: device.c:277-0 start open_output_device()
btape: device.c:296-0 Opening device.
btape: dev.c:376-0 open dev: type=2 dev_name="Tape" (/dev/nst0) vol=
mode=OPEN_READ_ONLY
btape: dev.c:1896-0 Enter mount
btape: dev.c:426-0 Open dev: device is tape
btape: dev.c:441-0 Try open "Tape" (/dev/nst0) mode=OPEN_READ_ONLY
btape: dev.c:463-0 Rewind after open
btape: dev.c:2392-0 In set_os_device_parameters
btape: dev.c:2414-0 MTSETDRVBUFFER
btape: dev.c:515-0 open dev: tape 3 opened
btape: dev.c:389-0 preserve=0x0 fd=3
btape: device.c:302-0 open dev "Tape" (/dev/nst0) OK
btape: btape.c:471-0 Opening device
btape: dev.c:366-0 Close fd for mode change.
btape: dev.c:376-0 open dev: type=2 dev_name="Tape" (/dev/nst0) vol=
mode=OPEN_READ_WRITE
btape: dev.c:1896-0 Enter mount
btape: dev.c:426-0 Open dev: device is tape
btape: dev.c:441-0 Try open "Tape" (/dev/nst0) mode=OPEN_READ_WRITE
btape: dev.c:463-0 Rewind after open
btape: dev.c:2392-0 In set_os_device_parameters
btape: dev.c:2414-0 MTSETDRVBUFFER
btape: dev.c:515-0 open dev: tape 3 opened
btape: dev.c:389-0 preserve=0x0 fd=3
btape: btape.c:477 open device "Tape" (/dev/nst0): OK
btape: btape.c:311-0 Do tape commands
*test
=== Write, rewind, and re-read test ===
I'm going to write 10000 records and an EOF
then write 10000 records and an EOF, then rewind,
and re-read the data to verify that it is correct.
This is an *essential* feature ...
btape: dev.c:1524-0 === weof_dev="Tape" (/dev/nst0)
btape: block.c:825-0 dir_update_volume_info max file size -- OK
btape: dev.c:1524-0 === weof_dev="Tape" (/dev/nst0)
btape: block.c:825-0 dir_update_volume_info max file size -- OK
btape: dev.c:1524-0 === weof_dev="Tape" (/dev/nst0)
btape: block.c:825-0 dir_update_volume_info max file size -- OK
btape: dev.c:1524-0 === weof_dev="Tape" (/dev/nst0)
(... repeated forever)
------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users
|
