[Bacula-users] Trouble cancelling a job, 1.38.11

2006-09-23 Thread Ryan Novosielski
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I'm having trouble getting one of my jobs to cancel. Here is the
scenario... I'm running incremental backups on a number of machines
concurrently to the same tape. As it happens, I've added a new machine
recently and I have not yet started backing it up, so I disabled the
spec. I guess I must have restarted or something because it came back on
by itself -- I have no doubt that it did what it was supposed to in that
instance.

However, the result was that the backup ran on that host. There is no
prior incremental, so it upgraded to full. As it's inappropriate to run
a full on the incremental tape (no disk space), I want to cancel that
job and disable the spec again. However, the job will not cancel. This
is twice now that I've run into this. Here is the 'status storage':

Running Jobs:
 JobId Level   Name   Status
==
  1317 FullVistaBKsave.2006-09-22_23.00.05 is waiting for a mount
request
  1318 FullCatalog.2006-09-22_23.55.00 is waiting for higher
priority jobs to finish


Terminated Jobs:
 JobId  Level Files  Bytes Status   FinishedName

  1304  Incr  3,035237,327,564 OK   21-Sep-06 23:03 CFMX-dev
  1306  Incr  1,413  6,072,268 OK   21-Sep-06 23:04
BigBrother-Display
  1303  Incr  0  0 Cancel   22-Sep-06 01:30 NJMSCatalyst
  1310  Incr  8  7,001,600 OK   22-Sep-06 01:34 NJMSCatalyst
  1311  Full  1 57,944,677 OK   22-Sep-06 01:37 Catalog
  1316  Incr  9136,395 OK   22-Sep-06 23:00 SVNrepos
  1314  Incr  5 40,568,871 OK   22-Sep-06 23:00
sopris-WEBMAIL
  1312  Incr  7 17,351,680 OK   22-Sep-06 23:00 NJMSCatalyst
  1313  Incr 12 19,560,209 OK   22-Sep-06 23:01 CFMX-dev
  1315  Incr  1,358  5,823,160 OK   22-Sep-06 23:04
BigBrother-Display


*cancel 1317
Select Job:
 1: JobId=1317 Job=VistaBKsave.2006-09-22_23.00.05
 2: JobId=1318 Job=Catalog.2006-09-22_23.55.00
Choose Job to cancel (1-2): 1

...and it just hangs here. Anyone seen anything like this? Shall I try
to get this to happen again with the daemons running in debug mode? Any
other suggestions from the gallery? I can stop and start Bacula to get
this one stopped, but obviously not a solution. I'm going to let it run
overnight to see if it ever gets anyplace.

Thanks for your help!

- --
  _  _ _  _ ___  _  _  _
 |Y#| |  | |\/| |  \ |\ |  | |Ryan Novosielski - Systems Programmer III
 |$| |__| |  | |__/ | \| _| |[EMAIL PROTECTED] - 973/972.0922 (2-0922)
 \__/ Univ. of Med. and Dent.|IST/AST - NJMS Medical Science Bldg - C630
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFFM6bmb+gadEcsb4RAvX1AKDKY4TFv4KG+KFSDf1FIpgz0/vrXwCgwAXn
bN43kZxFlg3XHy/0H14tRjg=
=byti
-END PGP SIGNATURE-

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT  business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


Re: [Bacula-users] Manual ejection of tape confuses bacula

2006-09-23 Thread Kern Sibbald
On Friday 22 September 2006 23:53, Sarath Jayewardena wrote:
 
 On Fri, 22 Sep 2006, Kern Sibbald wrote:
 
  Unfortunately, you are out of luck.  You did the unmount/mount *after* you 
  ejected the tape, reloaded it in the drive, and started another job, which 
  started writing on the *beginning* of the tape rather than appending to 
it. 
  
  The simplest solution is to stop and restart the SD (probably not 
necessary), 
  delete the Volume from the catalog, relabel it, and re-run all the jobs 
that 
  were backed up to that Volume.
  
 Thank you for the straigt answer. So, in other words, all the data on the
 tape is useless, right? Assuming I had multiple backups over several
 days (or weeks), they are all gone. 

From a quick look at your output, this is my conclusion. If the data is 
important, you might want to verify this -- ask on the bacula-users list if 
you need help.

 How about if I had not started a new 
 backup after re-inserting the tape, yet?

All the data would have been OK.

 Will an unmount/mount before 
 running any jobs make bacula accept the tape? 

Yes, it *should* make Bacula re-examine/reposition the tape.  If you don't 
unmount/mount, Bacula assumes that the tape is positioned at the end of the 
data.

 Or at least, can I use any other utility to access the data?

If you didn't write too much data, all the old data, if any, *after* the new 
stuff *may* be recoverable. One user published how he did this using mt and 
dd.  It is not the easiest thing in the world to do.

 
 - Sarath
 
  On Friday 22 September 2006 21:00, Sarath Jayewardena wrote:
   To demonstrate the problem with manual ejection of tape, this is what
   I did:
   
 - start with a fresh database
 - start bconsole
 - add a volume to the Default pool
 - run two backup jobs
 - quit bconsole
 - * manually eject tape without unmounting
 - * re-insert tape
 - start bconsole
 - without unmounting/mounting try to run a job (fails)
 - do an unmount/mount
 - try to run a job (fails)
   
   The transcript of the session is attached to the end of this mail.
   Actually, I tried out a couple of variations. In some cases, bacula
   seemed to be able to run jobs after re-inserting the tape. It would
   even list them as completed jobs. But then, when I tried to restore,
   it had problems accessing the tape. These are the dangers I am
   afraid of. Are there any rescue methods in such cases? Again thank
   you very much for the help.
   
   - Sarath
   
   
   On Thu, 21 Sep 2006, Kern Sibbald wrote:
   
On Thursday 21 September 2006 19:08, Sarath Jayewardena wrote:
 Hi,
 
 I am experimenting to see whether bacula would be a good system for 
me 
  to
 backup our department's servers. I am running bacula director and 
  storage
 daemon on a Debian (sarge) machine which has a scsi DLT tape drive
 attached to it. The version of bacula is 1.36.2. It looks promising,
 except for one annoying, and in my view, serious problem:
 
 The tape drive allows me to eject the tape by pressing an 'unload'
 button. However, afterwards, when I re-insert the tape, bacula 
cannot 
  read
 it, so all the backup data is useless. (Of course, if I have 
'unmounted'
 the tape from within bconsole before  ejecting, bacula has no 
problem
 accessing the data afterwards.) The question is, is there any way to 
  make
 bacula accept the tape if the tape was accidientally, 'improperly'
 ejected?

As long as Bacula is not writing on the tape at the moment you ejected 
the 
tape, the data is fine and can be read by Bacula.  

However, when you eject a tape without unmounting the storage device 
  first, 
Bacula will not know or except that you have changed the tape until 
you 
unmount and re-mount the storage device.

 
 I have browsed the mailing list archive and seen the advice that I 
could
 use AlwaysOpen = no and OfflineOnUnmount = yes to guarantee that
 bacula unmounts the tape after each backup. But the advice comes 
with
 the warning that for each backup bacula will be rewinding and
 re-positioning the tape consuming a lot of time, so leaving the 
default
 behaviour is really preferred.
 
 If someone knows that this problem does not exist in newer versions 
of
 bacula, I could upgrade to the version (1.38.11) in the Debian 
testing
 distribution.
 
 I would greatly appreciate any help regarding this. Thanks.
 
 - Sarath
 
 
 
 
  -
 Take Surveys. Earn Cash. Influence the Future of IT
 Join SourceForge.net's Techsay panel and you'll get the chance to 
share 
  your
 opinions on IT  business topics through brief surveys -- and earn 
cash
 
  http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
 

Re: [Bacula-users] close Outlook on windows client

2006-09-23 Thread Jim Peters








I added the line Enable VSS = Yes
to the fileset and this is the result,



23-Sep 09:51 gig-fd:
Gig-NightlySave.2006-09-23_09.50.49 Warning: VSS was not initialized properly.
VSS support is disabled. ERR=The operation completed successfully.



23-Sep 09:51 gig-fd: Gig-NightlySave.2006-09-23_09.50.49
Error: Read error on file C:/Documents and Settings/jim/Local
Settings/Application Data/Microsoft/Outlook/Outlook.pst. ERR=The process cannot
access the file because another process has locked a portion of the file.



How do I properly initialize
VSS?



Thanks











From:
[EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Mantas Marciulaitis
Sent: Friday, September 22, 2006
3:07 AM
To:
bacula-users@lists.sourceforge.net
Subject: Re: [Bacula-users] close
Outlook on windows client





Are you using VSS? Volume Shadow Copy
service is actually designed to make backups of locked files like outlooks
.pst, so it should work with Enable VSS=Yes option in the FileSet config.











From:
[EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jim Peters
Sent: 22 September 2006 09:42
To: bacula-users@lists.sourceforge.net
Subject: [Bacula-users] close
Outlook on windows client





I searched through the archives and found
instructions on how use a run before job script to shut down
outlook so I can backup the pst file. I have a batch file in the bacula
directory on my windows client that I call from the job with this; 

Client RunBeforeJob=
c:/bacula/before_back.bat



This file has only one line
right now;

Cscript
c:\bacula\close_outlook.vbs



And here is the contents of
close_outlook.vbs;

Dim oOL

Set oOL =
CreateObject(Outlook.Application)

oOL.Quit



If I run the batch file from
the windows client immediately outlook exits and all is fine, but if I run the
job with bacula I see the command window open up on the windows client then a
windows error message appears saying this;

Cannot Start Microsoft Office
Outlook. A program Error has occurred please close outlook and windows then try
again.



If I shut down outlook at
this point I get a fatal error in the backup and it terminates the job. If I
just wait for about a min. the windows error message goes away and the backup
continues to work except that it cant back up the outlook pst file and this is
what I see in the message on the console; 



22-Sep 01:08 gig-fd: ClientRunBeforeJob: 

22-Sep 01:08 gig-fd: ClientRunBeforeJob:
C:\WINDOWS\system32Cscript c:\bacula\close_outlook.vbs 

22-Sep 01:08 gig-fd: ClientRunBeforeJob:
Microsoft (R) Windows Script Host Version 5.6

22-Sep 01:08 gig-fd: ClientRunBeforeJob:
Copyright (C) Microsoft Corporation 1996-2001. All rights reserved.

22-Sep 01:08 gig-fd: ClientRunBeforeJob: 

22-Sep 01:10 gig-fd: ClientRunBeforeJob:
c:\bacula\close_outlook.vbs(2, 1) Microsoft _vbscript_ runtime error: ActiveX
component can't create object: 'Outlook.Application'



Doesnt make sense to my pea brain why I can run this
file by clicking on it from windows and all is well yet when the Bacula-dir
executes it I get this problem. Its the same file executing either
way 

At any rate does anyone have a solution for this? I am
running office 2003, windows xp and winbacula-1.38.9 on the client and bacula
version 1.38.9 on the director



Thanks






-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT  business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


Re: [Bacula-users] Trouble cancelling a job, 1.38.11

2006-09-23 Thread Ryan Novosielski
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Found out the trouble. Lost communication with the -fd. As for the
reason why? I have no idea. Eventually it printed an error message that
it could not communicate with the file daemon and cancelled.

Ah well, I guess nevermind. :)

Ryan Novosielski wrote:
 I'm having trouble getting one of my jobs to cancel. Here is the
 scenario... I'm running incremental backups on a number of machines
 concurrently to the same tape. As it happens, I've added a new machine
 recently and I have not yet started backing it up, so I disabled the
 spec. I guess I must have restarted or something because it came back on
 by itself -- I have no doubt that it did what it was supposed to in that
 instance.
 
 However, the result was that the backup ran on that host. There is no
 prior incremental, so it upgraded to full. As it's inappropriate to run
 a full on the incremental tape (no disk space), I want to cancel that
 job and disable the spec again. However, the job will not cancel. This
 is twice now that I've run into this. Here is the 'status storage':
 
 Running Jobs:
  JobId Level   Name   Status
 ==
   1317 FullVistaBKsave.2006-09-22_23.00.05 is waiting for a mount
 request
   1318 FullCatalog.2006-09-22_23.55.00 is waiting for higher
 priority jobs to finish
 
 
 Terminated Jobs:
  JobId  Level Files  Bytes Status   FinishedName
 
   1304  Incr  3,035237,327,564 OK   21-Sep-06 23:03 CFMX-dev
   1306  Incr  1,413  6,072,268 OK   21-Sep-06 23:04
 BigBrother-Display
   1303  Incr  0  0 Cancel   22-Sep-06 01:30 NJMSCatalyst
   1310  Incr  8  7,001,600 OK   22-Sep-06 01:34 NJMSCatalyst
   1311  Full  1 57,944,677 OK   22-Sep-06 01:37 Catalog
   1316  Incr  9136,395 OK   22-Sep-06 23:00 SVNrepos
   1314  Incr  5 40,568,871 OK   22-Sep-06 23:00
 sopris-WEBMAIL
   1312  Incr  7 17,351,680 OK   22-Sep-06 23:00 NJMSCatalyst
   1313  Incr 12 19,560,209 OK   22-Sep-06 23:01 CFMX-dev
   1315  Incr  1,358  5,823,160 OK   22-Sep-06 23:04
 BigBrother-Display
 
 
 *cancel 1317
 Select Job:
  1: JobId=1317 Job=VistaBKsave.2006-09-22_23.00.05
  2: JobId=1318 Job=Catalog.2006-09-22_23.55.00
 Choose Job to cancel (1-2): 1
 
 ...and it just hangs here. Anyone seen anything like this? Shall I try
 to get this to happen again with the daemons running in debug mode? Any
 other suggestions from the gallery? I can stop and start Bacula to get
 this one stopped, but obviously not a solution. I'm going to let it run
 overnight to see if it ever gets anyplace.
 
 Thanks for your help!
 

- -
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT  business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

- --
  _  _ _  _ ___  _  _  _
 |Y#| |  | |\/| |  \ |\ |  | |Ryan Novosielski - Systems Programmer III
 |$| |__| |  | |__/ | \| _| |[EMAIL PROTECTED] - 973/972.0922 (2-0922)
 \__/ Univ. of Med. and Dent.|IST/AST - NJMS Medical Science Bldg - C630
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFFXILmb+gadEcsb4RAoqlAKC2/iavQHy2PXbD4ZKgJbXG4qxaDgCgnsEK
0u++YVtBJyZLBPH+vIgjomA=
=JejZ
-END PGP SIGNATURE-

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT  business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


Re: [Bacula-users] Purged volumes not getting used

2006-09-23 Thread Ryan Novosielski
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


Jo Rhett wrote:
 On Wed, Sep 20, 2006 at 10:31:33AM -0400, Ryan Novosielski wrote:
 I would like some input on the script from those that are having the
 problem. Most specifically, I'd like to know pool configs and retention
 periods. I think my setup is working OK, so I'd like to see what might not.
  
 It doesn't seem to matter.  Disk backup, volume use restricted by either
 size or time.  Set it up so that it should only need X volumes, but set
 maximum volumes to X*3 and watch it use all of them, ignoring the
 recycle-ability of the previous volumes.
 
 I'll send you a small config if you really need it, but the description
 above is enough.

Nope, that will do fine. I am just trying to make sure that no other
parameters that could affect this are set in the configs that are
showing problems. I suppose I'll find out when I start with a vanilla
config.

- --
  _  _ _  _ ___  _  _  _
 |Y#| |  | |\/| |  \ |\ |  | |Ryan Novosielski - Systems Programmer III
 |$| |__| |  | |__/ | \| _| |[EMAIL PROTECTED] - 973/972.0922 (2-0922)
 \__/ Univ. of Med. and Dent.|IST/AST - NJMS Medical Science Bldg - C630
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFFXIPmb+gadEcsb4RAqSHAKCmsodtypLNByj0lXf2CIB+0p/CzACgtNmJ
vcc6Qd0WhfXbOGVz1IGAi7A=
=rSt1
-END PGP SIGNATURE-

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT  business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


Re: [Bacula-users] How long should volume pruning take?

2006-09-23 Thread Kern Sibbald
On Thursday 21 September 2006 08:41, Uwe Schuerkamp wrote:
 On Wed, Sep 20, 2006 at 03:10:27PM +0200, Kern Sibbald wrote:
  
  10 GB is pretty big, but there are users with databases that large.  Take 
a 
  look at src/cats/make_mysql_tables.  There are a few tips in that file for 
  creating extra indexes if you have slow pruning.  Once you have the 
indexes 
  properly setup for your site requirements, pruning should run in a maximum 
of 
  a couple of minutes.
  
 
 Hello Ken et al., 
 
 I created another index on the File table as recommended in the
 documentation yesterday which took about one hour to create (the File
 table data file is around 6 GB), because I wasn't sure about MySQL's
 index display format, but it seems I created an identical index to one
 already there (the new one is called file_jfp_index, sorry for the
 crummy formatting): 
 
 mysql show index from File ; 
 
+---++--+--+-+---+-+--++--++-+
 | Table | Non_unique | Key_name | Seq_in_index | Column_name |
 Collation | Cardinality | Sub_part | Packed | Null | Index_type |
 Comment |
 
+---++--+--+-+---+-+--++--++-+
 | File  | 0  | PRIMARY  | 1| FileId  | A
   | 59778107| NULL ||  | BTREE  |
   |
 | File  | 1  | JobId| 1| JobId   | A
   | 318 | NULL ||  | BTREE  |
   |
 | File  | 1  | JobId_2  | 1| JobId   | A
   | 318 | NULL ||  | BTREE  |
   |
 | File  | 1  | JobId_2  | 2| PathId  | A
   | 4598315 | NULL ||  | BTREE  |
   |
 | File  | 1  | JobId_2  | 3| FilenameId  | A
   | 59778107| NULL ||  | BTREE  |
   |
 | File  | 1  | file_jfp_idx | 1| JobId   | A
   | 318 | NULL ||  | BTREE  |
   |
 | File  | 1  | file_jfp_idx | 2| FilenameId  | A
   | 29889053| NULL ||  | BTREE  |
   |
 | File  | 1  | file_jfp_idx | 3| PathId  | A
   | 59778107| NULL ||  | BTREE  |
   |
 
+---++--+--+-+---+-+--++--++-+
 8 rows in set (0.00 sec)
 
 
 So to me, it looks like the new index is identical in setup to the one
 that's been there all along called JobId_2. As a consequence, I don't
 expect any performance improvements when it comes to pruning the next
 time around. Is this correct? 
 
 As per Ken's recommendation, I checked the table creation script, and
 it seems not to mention any additional indices I should create that
 aren't already there: 
 
 
 #
 # Possibly add one or more of the following indexes
 #  to the above File table if your Verifies are
 #  too slow.
 #
 #  INDEX (PathId),
 #  INDEX (FilenameId),
 #  INDEX (FilenameId, PathId)
 #  INDEX (JobId),
 #
 
 I don't have a combined (for lack of a better expression) index on
 (FilenameId, PathId), but wouldn't that just be a subset of the
 (Job, Filename, Path) index? Also, the comment talks about slow
 verifies, but my problem is slow pruning ;-) 
 
 Should I drop the 2nd, identical index? Backup last night was nearly
 as fast as before (varying from 10MB/sec down to 3MB/sec depending on
 the client), so it doesn't seem to hurt in a major way.

I'm not convinced the indexes are the same. Some people say yes, some say no, 
and it seems to be also dependent on whether one is talking about MySQL as 
you are or PostgreSQL.  The best solution is to test for your particular 
case.

 
 Again, if I'm missing some crucial information, let me know and I'll
 be happy to provide what I can. Below are the my.cnf parameters
 relevant to the server: 
 
 #
 # * Fine Tuning
 #
 key_buffer  = 256M
 max_allowed_packet  = 16M
 thread_stack= 128K
 record_buffer   = 256M
 sort_buffer_size= 64M
 myisam_sort_buffer_size = 64M
 #
 # * Query Cache Configuration
 #
 query_cache_limit   = 6448576
 query_cache_size= 64777216
 query_cache_type= 1
 
 Mysqld uses around 500-700MB RSS memory of 1GB total (sorry, got that
 wrong in my last post, the server only has 1 GB RAM installed),
 depending on activity.
 
 All the best  thanks in advance for your comments, 
 
 Uwe 
 
 -- 
 Uwe Schuerkamp, NIONEX GmbH (http://www.nionex.com/)
 [EMAIL PROTECTED] Tel: +49 (0)5241 / 80 10 66 FAX: / 806 23 38
 Avenwedder Str. 55, D-33311 Guetersloh, Germany
 GnuPG KeyID: 5887047D, Fingerprint: 2E1320229A3F63 7F676FE9B1A836A461
 
 

[Bacula-users] Disk Backups - with offsite storage too

2006-09-23 Thread Angus Jordan
Hi there,I have a client with an aging DDS-3 tape drive that is currently way too small, and I'd like to eliminate the use of tapes altogether.The only problem is that this particular client requires offsite storage of the backup data for at least a week after the *full* backup is done.
Is anyone currently using this type of setup?Hope someone can help.Regards,Angus Jordan
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT  business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


[Bacula-users] Help with DVD storage

2006-09-23 Thread Jim Peters








I decided that the box I had bacula installed in was a little
week so I removed the 3 hard drives and stuck them in an old AMD Dual Athlon box
I use for testing and building linux stuff on but has been sitting Idle for a few
months. I love linux, try swapping windows hard drives to different hardware like
this and see the pretty blue screens come up! Anyway this new box also has a
DVD+RW drive so I thought I would experiment with using bacula to write to DVDs
, here is my configuration;



Bacula-sd.conf



#

# A DVD device

#

Device {

 Name =
DVD-RW

 Media Type = DVD

 Archive Device =
/dev/hdb

 LabelMedia =
yes;
# lets Bacula label unlabeled media

 Random Access = Yes;

 AutomaticMount =
no;
# when device opened, read it

 RemovableMedia = yes;

 AlwaysOpen = no;

 MaximumPartSize =
800M;

 Requires Mount = yes;

 Mount Point =
/mnt/dvd;

 Mount Command
= /bin/mount -t iso9660 -o ro %a %m;

 Unmount Command =
/bin/umount %m;

 Spool Directory =
/tmp/backup;

 Write Part Command =
/etc/bacula/dvd-handler %a write %e %v

 Free Space Command =
/etc/bacula/dvd-handler %a free

}

#



And my bacula-dir.conf



Job {

 Name =
BackupHome

 Type =
Backup

 Level =
Full

 Client =
Server-fd

 FileSet =
Home

 Storage =
dvd

 Pool =
Default

 Priority =
10

 Messages = Standard

 WriteBootstrap =
/var/bacula/Server-fd.bsr

 Write Part After Job =
yes

}



Storage {

 Name = dvd

 Address =
192.168.0.4

 SDPort = 9103

 Password =
password

 Device = DVD-RW

 Media Type = DVD

}



If you see anything that should be changed please let me know,
I really have no clue as to what Im doing here but am giving it a shot.

After setting up these files I stuck a brand new +RW disc in
the drive and ran the BackupHome job. It starts the job and never really gives
me any errors and doesnt end the job either all I see is this,



Job started. JobId=32

23-Sep 14:50 Server-dir: Start Backup JobId 32,
Job=BackupHome.2006-09-23_14.50.02

23-Sep 14:50 Server-sd: Please mount Volume
DefaultVolume-0002 on Storage Device DVD-RW (/dev/hdb)
for Job BackupHome.2006-09-23_14.50.02



Am I supposed to manually mount the DVD at this point? I
thought bacula wrote to the device in a raw format and didnt need to be
mounted? And if it does I would think the mount command =
derective would take care of this?



Here is what I see in my message log;



Sep 23 14:50:05 server
kernel: attempt to access beyond end of device

Sep 23 14:50:05 server
kernel: hdb: rw=0, want=68, limit=4

Sep 23 14:50:05 server
kernel: isofs_fill_super: bread failed, dev=hdb, iso_blknum=16, block=16

Sep 23 14:50:06 server
kernel: attempt to access beyond end of device

Sep 23 14:50:06 server
kernel: hdb: rw=0, want=68, limit=4

Sep 23 14:50:06 server
kernel: isofs_fill_super: bread failed, dev=hdb, iso_blknum=16, block=16

Sep 23 14:50:06 server
kernel: attempt to access beyond end of device

Sep 23 14:50:06 server
kernel: hdb: rw=0, want=68, limit=4

Sep 23 14:50:06 server
kernel: isofs_fill_super: bread failed, dev=hdb, iso_blknum=16, block=16

Sep 23 14:50:07 server
kernel: attempt to access beyond end of device

Sep 23 14:50:07 server
kernel: hdb: rw=0, want=68, limit=4

Sep 23 14:50:07 server
kernel: isofs_fill_super: bread failed, dev=hdb, iso_blknum=16, block=16



What have I done wrong or forgot to do at all? I am lost as
to where to look, any ideas appreciated. Oh and I am using version 1.38.9 of bacula,

Thanks










-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT  business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users