Hi,
   I am trying to replicate bacula backup repository and MySQL database
on another system. Any idea what all files need to be replicated and
what is the procedure to update MySQL bacula database on the other node
in real-time.

Regards,
Arun.  

-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
[EMAIL PROTECTED]
Sent: Wednesday, July 11, 2007 12:32 AM
To: [email protected]
Subject: Bacula-users Digest, Vol 15, Issue 9

Send Bacula-users mailing list submissions to
        [email protected]

To subscribe or unsubscribe via the World Wide Web, visit
        https://lists.sourceforge.net/lists/listinfo/bacula-users
or, via email, send a message with subject or body 'help' to
        [EMAIL PROTECTED]

You can reach the person managing the list at
        [EMAIL PROTECTED]

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Bacula-users digest..."


Today's Topics:

   1. Re: New Exabyte Autochanger question (Maria McKinley)
   2. Re: New Exabyte Autochanger question (Christian Vogel)
   3. Pruning to the scratch pool. (Falk Sauer)
   4. Re: New Exabyte Autochanger question (tomasz dereszynski)
   5. Job Failed Fatal error: sql_create.c:93 Create DB Job record
      (Pieter (NL))
   6. Re: Job Failed Fatal error: sql_create.c:93 Create DB     Job
      record (Arno Lehmann)
   7. Re: btape and bsr sd.conf tweaks (Martin Simmons)
   8. Winows fd random shutdowns debunked (Mantas M.)
   9. Re: Winows fd random shutdowns debunked (Arno Lehmann)
  10. Re: Winows fd random shutdowns debunked (John Drescher)
  11. Re: Help with backup plan (Steen)
  12. Re: Winows fd random shutdowns debunked (Mantas M.)
  13. Re: Winows fd random shutdowns debunked (John Drescher)
  14. bacula lto-2 drive performance? (Bob Hetzel)
  15. Re: Job Failed Fatal error: sql_create.c:93 Create DB     Job
      record (Pieter (NL))
  16. Please mount volume...but volume is already mounted.
      (Brian Debelius)
  17. Re: Please mount volume...but volume is already   mounted. (mikee)
  18. Re: btape and bsr sd.conf tweaks (Ken Gunderson)
  19. confused by automatic purging (Craig White)
  20. Re: btape and bsr sd.conf tweaks (Martin Simmons)
  21. Re: New Exabyte Autochanger question (Richard White)
  22. Re: confused by automatic purging (John Drescher)
  23. Re: New Exabyte Autochanger question (Richard White)
  24. Re: Pruning to the scratch pool. (Andreas Helmcke)
  25. Unsure of btape output when testing autochanger (Shon Stephens)


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

Message: 1
Date: Mon, 09 Jul 2007 22:03:40 -0700
From: Maria McKinley <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] New Exabyte Autochanger question
To: bacula-users <[email protected]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii; format=flowed

John Drescher wrote:
> One other comment the manual states you should use nst0 instead of st0
see here:
> 
>
http://www.mail-archive.com/[email protected]/msg16139.
html
> 
> John
> 

I don't quite get this. When I tried to use nst0 instead of st0, it just

didn't work. And why is it better to use the "non-rewind" variant of the

device file name?

cheers,
maria



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

Message: 2
Date: Tue, 10 Jul 2007 09:11:20 +0200
From: Christian Vogel <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] New Exabyte Autochanger question
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=utf-8

Hi Richard, Hi Group,

On Mon, Jul 09, 2007 at 12:01:51PM -0700, Richard White wrote:
> > Are you 100% sure that  /dev/sg0 is your changer?
?
> You are right! /dev/sg0 yields some ATA nonsense, but
> /dev/sg1 tells me it is an Exabyte 1x10 1u. I will
> give it another try.

I recently had to struggle with a server with several
hot-pluggable SATA drives where depending on the number
of plugged drives the /dev/sg nodes seem to shuffle around.

Recent Linux distributions (namely the udev device management) seems to
have a nice solution to this problem (I am currently running Ubuntu
Feisty). They create symlinks named after the devices serial number,
so in Baculas sd.conf, I have:

        Changer Device = /dev/tape/by-id/scsi-200900de6055fa000
        Archive Device = /dev/tape/by-id/scsi-350060b000033c0d9-nst

unfortunately, the same udev rules miss the entry to adjust the
tape changer's permissions (/etc/udev/rules.d/40-permissions.rules):

        # SCSI devices
        SUBSYSTEMS=="scsi", GOTO="scsi_start"
        GOTO="scsi_end"
        LABEL="scsi_start"
        ATTRS{type}=="1",                       GROUP="tape"
----- ADD THE NEXT LINE: vvv
        ATTRS{type}=="8",                       GROUP="tape"
----- ADD THE PREVIOUS LINE: ^^^
        ATTRS{type}=="5",                       GROUP="cdrom"
        ATTRS{type}=="6",                       GROUP="scanner"
        ATTRS{type}=="3", ATTRS{vendor}=="HP",  GROUP="scanner"
        ATTRS{type}=="3", ATTRS{vendor}=="Epson", GROUP="scanner"
        LABEL="scsi_end"

Hope it's useful for someone of you.

Greetings from Germany,

        Chris

-- 
Saying that XP is the most stable MS OS is like saying that 
asparagus is the most articulate vegetable. -- Dave Barry



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

Message: 3
Date: Tue, 10 Jul 2007 09:12:22 +0200
From: Falk Sauer <[EMAIL PROTECTED]>
Subject: [Bacula-users] Pruning to the scratch pool.
To: bacula-users <[email protected]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain;  charset="us-ascii"

Hi,

Bacula 2-0-3 from rpm, os10.2 32bit x86, 
adic scalar24 with 2 SDLT320, 5,5TB NTEC IceBox over SCSI

I mean on any post in the last weeks i've read (only with one eye) over
a 
option thats allows a volume after pruning, automatically moving to the 
scratch pool.

That can be a very nice feature, in the moment i make this manually.

>From which version of bacula is this option usable and how is the
syntax?

And how is the pruning working in the case of reuse a volume while the
sd 
needs a pruned volume and have more outtimed volumes as he needs in the 
moment? Comes here the sd first and only momentarily unused volumes
going to 
the scratch pool or moves the pruning algo all pruned volumes to the
scratch 
pool and the sd reuse one of the volumes after them from scratch pool?

Why i ask? If the sd picks a self pruned volume, the volume believe in
the 
same pool and the use structure of the volume is every time  the same 
(monthly, weekly ...), if every time move the prune algo all of the
volumes 
to the scratch pool*, we become a better [stochastic/more evenly] use 
structure of the volumes when the volumes rotate between the pools.

*) i've noticed that's the scratch pool works like a fifo.

regards
   Falk




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

Message: 4
Date: Tue, 10 Jul 2007 09:21:52 +0100
From: tomasz dereszynski <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] New Exabyte Autochanger question
To: Richard White <[EMAIL PROTECTED]>,
        [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii

Richard White wrote:
>>>> On 7/9/2007 at 3:47 PM, in message
> <[EMAIL PROTECTED]>, "John
Drescher"
> <[EMAIL PROTECTED]> wrote:
>>> Please also note that, in the new and improved versions, the device
name
>>> is enclosed in double quotes, just like in the book and your
examples, but
>>> not, alas, like my originals.
>>>
>>> Now when I attempt the label barcode, I am presented with a choice
of
>>> devices, VXA and Auto. I choose 'Auto', but get exactly the same
result from
>>> Bacula. No matter what I do, it can't open the device "VXA3"
(/dev/nst0).
>>>
>> This might be a permissions issue. Depending on your install your
bacula-sd
>> may be running as user bacula. If so you need to make sure that user
bacula
>> has full access to the tape drive.
>>
>> John
> 
> Well, I just did the following:
> 
> # groupadd bacula
> # vi /etc/group
> 
> Make the last line look like this:
> 
> bacula:!:1000:root,bacula,richarw
> 
> # chown bacula:bacula /dev/nst0
> 
> At this point, presumably the user and group bacula own /dev/nst0.
> 
> # ls -l /dev/nst0
> # crw-rw----  1 bacula  bacula  9,  128 Jul  5 15:14  /dev/nst0
> 
> No change in result.
> 
> Most perplexing.

that was mention on group earlier
what is your system?
cos quite lot of systems is changing rights owners for devices after
restart

what means u should stick it in place like  /etc/devfs.conf on FreeBSD

i dont think changing owner to bacula i a best way

i will recommend to stick bacula in 'operator' group and then in
/etc/devfs.conf add privileges for group to device


-- 
bEsT rEgArDs            |       "Confidence is what you have before you
tomasz dereszynski      |       understand the problem." -- Woody Allen
TD840-RIPE              |



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

Message: 5
Date: Tue, 10 Jul 2007 01:56:27 -0700 (PDT)
From: "Pieter (NL)" <[EMAIL PROTECTED]>
Subject: [Bacula-users] Job Failed Fatal error: sql_create.c:93 Create
        DB      Job record
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii


Updated MySQL yesterday and all backups fail with the following
message(s):

Job failed.
10-Jul 10:05 backup-dir: sql_create.c:91 INSERT INTO Job
(Job,Name,Type,Level,JobStatus,SchedTime,JobTDate) VALUES
('halrouter.2007-07-10_10.05.22','halrouter','B','I','C','2007-07-10
10:05:20',1184054720)
10-Jul 10:05 backup-dir: halrouter.2007-07-10_10.05.22 Fatal error:
sql_create.c:93 Create DB Job record INSERT INTO Job
(Job,Name,Type,Level,JobStatus,SchedTime,JobTDate) VALUES
('halrouter.2007-07-10_10.05.22','halrouter','B','I','C','2007-07-10
10:05:20',1184054720) failed. ERR=

Any suggestions?

Pieter
-- 
View this message in context:
http://www.nabble.com/Job-Failed-Fatal-error%3A-sql_create.c%3A93-Create
-DB-Job-record-tf4054580.html#a11517075
Sent from the Bacula - Users mailing list archive at Nabble.com.




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

Message: 6
Date: Tue, 10 Jul 2007 11:59:42 +0200
From: Arno Lehmann <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] Job Failed Fatal error: sql_create.c:93
        Create DB       Job record
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi,

10.07.2007 10:56,, Pieter (NL) wrote::
> Updated MySQL yesterday and all backups fail with the following
message(s):
> 
> Job failed.
> 10-Jul 10:05 backup-dir: sql_create.c:91 INSERT INTO Job
> (Job,Name,Type,Level,JobStatus,SchedTime,JobTDate) VALUES
> ('halrouter.2007-07-10_10.05.22','halrouter','B','I','C','2007-07-10
> 10:05:20',1184054720)
> 10-Jul 10:05 backup-dir: halrouter.2007-07-10_10.05.22 Fatal error:
> sql_create.c:93 Create DB Job record INSERT INTO Job
> (Job,Name,Type,Level,JobStatus,SchedTime,JobTDate) VALUES
> ('halrouter.2007-07-10_10.05.22','halrouter','B','I','C','2007-07-10
> 10:05:20',1184054720) failed. ERR=

Really no error message printed?

> Any suggestions?

Yup... check the database. MySQL can fill its table space, the cure is 
simple.
It's also possible that the MySQL server disconnected after a long 
idle time, which can also easily be fixed.
Finally, you might have a more serious problem with the database, 
which you could analyze and hopefully fix using the MySQL tools.

Since you write you updated the database, but give no more details, I 
suggest looking through the MySQL related log files first.

Arno

> Pieter

-- 
Arno Lehmann
IT-Service Lehmann
www.its-lehmann.de



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

Message: 7
Date: Tue, 10 Jul 2007 11:12:35 +0100
From: Martin Simmons <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] btape and bsr sd.conf tweaks
To: [EMAIL PROTECTED]
Cc: [email protected]
Message-ID: <[EMAIL PROTECTED]>

>>>>> On Mon, 9 Jul 2007 20:27:23 -0600, Ken Gunderson said:
> 
> Greetings:
> 
> Since my baffled post I've been doing some testing w/various sd.conf
> tweaks.  btape results and fbsd ports message initially recommended:
> 
> Due to lack of some features in the FreeBSD tape driver
> implementation you MUST add some OS dependent options to
> the bacula-sd.conf file:
> 
>   Hardware End of Medium = no;
>   Backward Space Record  = no;
>   Backward Space File    = no;
> 
> With 2 filemarks at EOT (see man mt):
>   Fast Forward Space File = no;
>   BSF at EOM = yes;
>   TWO EOF    = yes;
> 
> 
> While btape completes w/above no problem, from man mt it looks like
FBSD
> supports bsr, however, and a few tests from command line seem to
> confirm so I trimmed sd.conf down thusly:
> 
>   Hardware End of Medium = no;
>   Fast Forward Space File = no;
>   BSF at EOM = yes;
>   TWO EOF = yes;
> 
> btape test completes fine and no longer recommend the bsr related
> tweaks.  Maybe they're not necessary and btape should be modified?
> Don't know about fill results yet - running now.  Hopefully I'll
> encounter the magic incantation to get my autochanger to actually
> change tapes...

In the current Bacula tape driver, Backward Space Record is only used to
verify that the last block has been written correctly when a tape gets
full,
it is not strictly necessary.  In addition, the setting of Backward
Space File
is ignored (bsf must work).

You might be able to use these settings on some FBSD tape drives:

  Fast Forward Space File = yes
  BSF at EOM = no
  TWO EOF = no

__Martin



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

Message: 8
Date: Tue, 10 Jul 2007 13:35:49 +0300
From: "Mantas M." <[EMAIL PROTECTED]>
Subject: [Bacula-users] Winows fd random shutdowns debunked
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hello,

I've just submited a bug report regarding this issue, but until it 
remains unsolved, i'll share my experience in case anyone else runs in 
to the same problem:

For a long period of time i had a strange issue with windows (xp pro, 
server 2003)  fds randomly shutting down. At last, today I figured out
why.

Issue occurs at these scenarios:

* User A logs in. Then user B connects to the machine using remote 
desktop. A dialog pops up asking if B wants to remotelly log off A. If 
accepted, A gets logged off and fd service shuts down.

* User A logs in, locks the computer (ctrl+alt+del > lock computer). 
Then  user B unlocks it with his account (B has to have administrative 
privileges for this to work). The same dialog pops up, if you choose to 
force log off of user A, the fd gets shut down.

I guess the problem is, that the file daemon service and the gui program

that resides in the systray are actually the same process. When the 
systray program gets closed down, the fd service goes down as well.



I hope this helps... You can check the status of the bug at 
http://bugs.bacula.org/view.php?id=893

Regards,
Mantas



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

Message: 9
Date: Tue, 10 Jul 2007 12:49:51 +0200
From: Arno Lehmann <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] Winows fd random shutdowns debunked
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hello,

10.07.2007 12:35,, Mantas M. wrote::
> Hello,
> 
> I've just submited a bug report regarding this issue, but until it 
> remains unsolved, i'll share my experience in case anyone else runs in

> to the same problem:
> 
> For a long period of time i had a strange issue with windows (xp pro, 
> server 2003)  fds randomly shutting down. At last, today I figured out
why.
> 
> Issue occurs at these scenarios:
> 
> * User A logs in. Then user B connects to the machine using remote 
> desktop. A dialog pops up asking if B wants to remotelly log off A. If

> accepted, A gets logged off and fd service shuts down.
> 
> * User A logs in, locks the computer (ctrl+alt+del > lock computer). 
> Then  user B unlocks it with his account (B has to have administrative

> privileges for this to work). The same dialog pops up, if you choose
to 
> force log off of user A, the fd gets shut down.
> 
> I guess the problem is, that the file daemon service and the gui
program 
> that resides in the systray are actually the same process. When the 
> systray program gets closed down, the fd service goes down as well.

That sounds reasonable.

Currently, development is under progress to separate the tray monitor 
from the FD itself (required for vista). That should fix this issue.

Arno

> 
> 
> I hope this helps... You can check the status of the bug at 
> http://bugs.bacula.org/view.php?id=893
> 
> Regards,
> Mantas
> 
>
------------------------------------------------------------------------
-
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Bacula-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/bacula-users

-- 
Arno Lehmann
IT-Service Lehmann
www.its-lehmann.de



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

Message: 10
Date: Tue, 10 Jul 2007 08:29:35 -0400
From: "John Drescher" <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] Winows fd random shutdowns debunked
To: [email protected]
Message-ID:
        <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

> Currently, development is under progress to separate the tray monitor
> from the FD itself (required for vista). That should fix this issue.
>
Sorry about the delay of that I had problems with my car that are
prevented me from working on this last week. At this point the tray
has been removed from the bacula-fd but the separate tray app is not
yet started. If anyone wants the executable I do have one I could
email that has the tray removed.

John



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

Message: 11
Date: Tue, 10 Jul 2007 14:17:57 +0200
From: Steen <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] Help with backup plan
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain;  charset="iso-8859-1"

On Friday 06 July 2007 06:23:22 Charles Sprickman wrote:
> Hi all,
>
> I'm very happy with Bacula so far.  I've got a bunch of clients setup
and
> have a basic config ready to go.  So far backing up to disk is working
> well and I need to figure out how to implement my real schedule and
tape
> usage now.  I'm coming from Amanda, so it's taking me a bit of time to
> digest the manual.
i Know nothing of Amanda - took me a bit of time too
>
> What I'd like to do is the following:
>
> -Full backup each sunday, keeping the tapes for a year.  After that I
> would likely manually relabel all the tapes but one.  These tapes
would
> all be in one pool.
>
> -Differential backup each following sunday, with 3 tapes that get
recycled
> each month.  These tapes are also in their own pool.
>
> -Incrementals mon-fri, with 5 tapes that get recycled each week.
Again,
> in their own pool.
>
> Does this seem like a sane plan? 
I use almost the same pattern - but only diskbased
> I'm on the fence about the differentials 
Don't really get that
> as we very rarely go back to backups for anything...  I'm also
wondering
> how these choices will impact the general ease of restoring something
if
> we need to.  Again, I'm coming from Amanda, and there's not much
thinking
> there.  When you restore, you just keep feeding it the tapes it wants.
Here it graps the disk volumes automatically - guess that would be
identical 
to your experience with Amanda with tapes
>
> Here's what I've tentatively got in my bacula-dir config.
>
> pools:
>
> Pool {
>    Name = Monthly
>    Pool Type = Backup
>    Recycle = no                        # Bacula can automatically
recycle
> Volumes AutoPrune = yes                     # Prune expired volumes
>    Volume Retention = 365 days         # one year
>    Volume Use Duration = 23h
>    Maximum Volumes = 12
>    LabelFormat = "Monthly"
> }
>
> Pool {
>    Name = Weekly
>    Pool Type = Backup
>    Recycle = yes                       # Bacula can automatically
recycle
> Volumes Recycle Oldest Volume = yes
>    AutoPrune = yes                     # Prune expired volumes
>    Volume Retention =  1 month         # one month
>    Volume Use Duration = 6 days
>    Maximum Volumes = 3
>    LabelFormat = "Weekly"
> }
>
> Pool {
>    Name = Daily
>    Pool Type = Backup
>    Recycle = yes                       # Bacula can automatically
recycle
> Volumes Recycle Oldest Volume = yes
>    AutoPrune = yes                     # Prune expired volumes
>    Volume Retention = 5 days           # 5 days
>    Volume Use Duration = 23h
>    Maximum Volumes = 5
>    LabelFormat = "Daily"
> }
>
My Pool defs look like these - except I have "Accept Any Volume = yes"

> I'm hoping what I've got there will force me to put a new tape in for
each
> run ("Volume Use Duration").  Am I on the right track with this or
not?
Yes
> Would it be simpler to let the incrementals all go to one tape and
reduce
> this to four tapes/month for those? 
Yes - no need to switch tapes
> I'm using 100GB LTO-2 tapes, so I 
> certainly have more than enough room, but I suppose I open myself up
to
> problems if a tape fails...
Yes - loose 5 days at most
> Does the above config look fairly sane?
>

To me - yes
> Thanks,
>
> Charles
>
>
------------------------------------------------------------------------
-
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Bacula-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/bacula-users



-- 
Regards

Steen



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

Message: 12
Date: Tue, 10 Jul 2007 16:23:42 +0300
From: "Mantas M." <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] Winows fd random shutdowns debunked
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi John,

Yes, it would be great if You could email the executable or upload it 
somewhere.

Thanks,

Mantas.



John Drescher wrote:
>> Currently, development is under progress to separate the tray monitor
>> from the FD itself (required for vista). That should fix this issue.
>>
> Sorry about the delay of that I had problems with my car that are
> prevented me from working on this last week. At this point the tray
> has been removed from the bacula-fd but the separate tray app is not
> yet started. If anyone wants the executable I do have one I could
> email that has the tray removed.
> 
> John
> 
>
------------------------------------------------------------------------
-
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Bacula-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/bacula-users
> 




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

Message: 13
Date: Tue, 10 Jul 2007 09:43:01 -0400
From: "John Drescher" <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] Winows fd random shutdowns debunked
To: "Mantas M." <[EMAIL PROTECTED]>
Cc: [email protected]
Message-ID:
        <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

On 7/10/07, Mantas M. <[EMAIL PROTECTED]> wrote:
> Hi John,
>
> Yes, it would be great if You could email the executable or upload it
> somewhere.
>

I was having a very difficult time trying to get gmail or comcast to
host the file so I tried a free binary host service.The link is below.
You go to this site and click on request download link on the upper
left. After that you click download file on the upper right.

http://mihd.net/mtqigz

John



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

Message: 14
Date: Tue, 10 Jul 2007 10:31:49 -0400
From: Bob Hetzel <[EMAIL PROTECTED]>
Subject: [Bacula-users] bacula lto-2 drive performance?
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi all,

I'm currently implementing a new bacula server running on a new dual 
core/dual cpu Dell PowerEdge 2900 server into a Dell Powervault 136T 
with a LTO-2 drive (IBM Ultrium-td2 model).  The tape drive and robot 
are on a dedicated Adaptec 29160 card.  The OS is OpenSuse 10.2

I think I should be able to get around 100 gigabytes/hour backup data 
rate with this hardware, but it looks like I'm getting about half that. 
  Of those of you running bacula on LTO-2 hardware, what kind of backup 
rates do you get?

Also, if you've got substantially better than 50 GB/hr on LTO-2 
hardware, what scsi card are you using?

Any other suggestions?

    Bob



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

Message: 15
Date: Tue, 10 Jul 2007 07:48:55 -0700 (PDT)
From: "Pieter (NL)" <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] Job Failed Fatal error: sql_create.c:93
        Create DB       Job record
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii


There was really no error message. Rebooted the server and recompiled
and
reinstalled mysql51-server and client and it's working again. Don't know
why
the problem disappeared.



Arno Lehmann wrote:
> 
> Hi,
> 
> 10.07.2007 10:56,, Pieter (NL) wrote::
>> Updated MySQL yesterday and all backups fail with the following
>> message(s):
>> 
>> Job failed.
>> 10-Jul 10:05 backup-dir: sql_create.c:91 INSERT INTO Job
>> (Job,Name,Type,Level,JobStatus,SchedTime,JobTDate) VALUES
>> ('halrouter.2007-07-10_10.05.22','halrouter','B','I','C','2007-07-10
>> 10:05:20',1184054720)
>> 10-Jul 10:05 backup-dir: halrouter.2007-07-10_10.05.22 Fatal error:
>> sql_create.c:93 Create DB Job record INSERT INTO Job
>> (Job,Name,Type,Level,JobStatus,SchedTime,JobTDate) VALUES
>> ('halrouter.2007-07-10_10.05.22','halrouter','B','I','C','2007-07-10
>> 10:05:20',1184054720) failed. ERR=
> 
> Really no error message printed?
> 
>> Any suggestions?
> 
> Yup... check the database. MySQL can fill its table space, the cure is

> simple.
> It's also possible that the MySQL server disconnected after a long 
> idle time, which can also easily be fixed.
> Finally, you might have a more serious problem with the database, 
> which you could analyze and hopefully fix using the MySQL tools.
> 
> Since you write you updated the database, but give no more details, I 
> suggest looking through the MySQL related log files first.
> 
> Arno
> 
>> Pieter
> 
> -- 
> Arno Lehmann
> IT-Service Lehmann
> www.its-lehmann.de
> 
>
------------------------------------------------------------------------
-
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Bacula-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/bacula-users
> 
> 

-- 
View this message in context:
http://www.nabble.com/Job-Failed-Fatal-error%3A-sql_create.c%3A93-Create
-DB-Job-record-tf4054580.html#a11522451
Sent from the Bacula - Users mailing list archive at Nabble.com.




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

Message: 16
Date: Tue, 10 Jul 2007 10:59:54 -0400
From: Brian Debelius <[EMAIL PROTECTED]>
Subject: [Bacula-users] Please mount volume...but volume is already
        mounted.
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

elkton-dir Version: 2.1.22 (26 June 2007) Linux Cross-compile Win32

Hey,
I am still getting this mounting problem.  Bacula loads the correct 
tape, but then says that I need to mount it.  After which I release, and

mount the same tape, and it then proceed to backup.

brian-


10-Jul 09:28 elkton-dir: Recycled volume "A0000009"
10-Jul 09:28 elkton-sd: 3307 Issuing autochanger "unload slot 3, drive 
0" command.
10-Jul 09:29 elkton-sd: 3304 Issuing autochanger "load slot 4, drive 0" 
command.
10-Jul 09:31 elkton-sd: 3305 Autochanger "load slot 4, drive 0", status 
is OK.
10-Jul 09:31 elkton-sd: 3301 Issuing autochanger "loaded? drive 0"
command.
10-Jul 09:31 elkton-sd: 3302 Autochanger "loaded? drive 0", result is 
Slot 4.
10-Jul 09:31 elkton-sd: Please mount Volume "A0000009" or label a new 
one for:
    Job:          Elkton_D-drive.2007-07-10_00.01.25
    Storage:      "VXA-172" (Tape0)
    Pool:         Tape-daily
    Media type:   X10
*update slots
Automatically selected Catalog: MyCatalog
Using Catalog "MyCatalog"
The defined Storage resources are:
     1: Disk0
     2: Disk1
     3: Tape
Select Storage resource (1-3): 3
Connecting to Storage daemon Tape at elkton.intelesys.local:9103 ...
3306 Issuing autochanger "slots" command.
Device "Packetloader0" has 10 slots.
Connecting to Storage daemon Tape at elkton.intelesys.local:9103 ...
3306 Issuing autochanger "list" command.
Catalog record for Volume "A0000004" updated to reference slot
1.*release
The defined Storage resources are:
     1: Disk0
     2: Disk1
     3: Tape
Select Storage resource (1-3): 3
3307 Issuing autochanger "unload slot 4, drive 0" command.
3922 Device "VXA-172" (Tape0) waiting for sysop.
*mount
The defined Storage resources are:
     1: Disk0
     2: Disk1
     3: Tape
Select Storage resource (1-3): 3
Enter autochanger slot: 4
3001 OK mount. Device="VXA-172" (Tape0)
Catalog record for Volume "A0000003" updated to reference slot 2.
Catalog record for Volume "A0000008" updated to reference slot 3.
Catalog record for Volume "B0000006" updated to reference slot 5.
Catalog record for Volume "A0000001" updated to reference slot 6.
Catalog record for Volume "CLNA0002" updated to reference slot 10.
Catalog record for Volume "A0000009" updated to reference slot 4.
10-Jul 10:50 elkton-sd: 3301 Issuing autochanger "loaded? drive 0"
command.
10-Jul 10:50 elkton-sd: 3302 Autochanger "loaded? drive 0", result: 
nothing loaded.
10-Jul 10:50 elkton-sd: 3304 Issuing autochanger "load slot 4, drive 0" 
command.
10-Jul 10:51 elkton-sd: 3305 Autochanger "load slot 4, drive 0", status 
is OK.
10-Jul 10:51 elkton-sd: 3301 Issuing autochanger "loaded? drive 0"
command.
10-Jul 10:51 elkton-sd: 3302 Autochanger "loaded? drive 0", result is 
Slot 4.
10-Jul 10:51 elkton-sd: Recycled volume "A0000009" on device "VXA-172" 
(Tape0), all previous data lost.
10-Jul 10:51 elkton-sd: New volume "A0000009" mounted on device 
"VXA-172" (Tape0) at 10-Jul-2007 10:51.




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

Message: 17
Date: Tue, 10 Jul 2007 10:02:02 -0500
From: mikee <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] Please mount volume...but volume is
        already mounted.
To: Brian Debelius <[EMAIL PROTECTED]>
Cc: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii

On Tue, 10 Jul 2007, Brian Debelius might have said:

> elkton-dir Version: 2.1.22 (26 June 2007) Linux Cross-compile Win32
> 
> Hey,
> I am still getting this mounting problem.  Bacula loads the correct 
> tape, but then says that I need to mount it.  After which I release,
and 
> mount the same tape, and it then proceed to backup.
> 
> brian-

How about umounting the mounting the volume?

Mike



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

Message: 18
Date: Tue, 10 Jul 2007 09:10:24 -0600
From: Ken Gunderson <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] btape and bsr sd.conf tweaks
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=US-ASCII

On Tue, 10 Jul 2007 11:12:35 +0100
Martin Simmons <[EMAIL PROTECTED]> wrote:

> >>>>> On Mon, 9 Jul 2007 20:27:23 -0600, Ken Gunderson said:
> > 
> > Greetings:
> > 
> > Since my baffled post I've been doing some testing w/various sd.conf
> > tweaks.  btape results and fbsd ports message initially recommended:
> > 
> > Due to lack of some features in the FreeBSD tape driver
> > implementation you MUST add some OS dependent options to
> > the bacula-sd.conf file:
> > 
> >   Hardware End of Medium = no;
> >   Backward Space Record  = no;
> >   Backward Space File    = no;
> > 
> > With 2 filemarks at EOT (see man mt):
> >   Fast Forward Space File = no;
> >   BSF at EOM = yes;
> >   TWO EOF    = yes;
> > 
> > 
> > While btape completes w/above no problem, from man mt it looks like
FBSD
> > supports bsr, however, and a few tests from command line seem to
> > confirm so I trimmed sd.conf down thusly:
> > 
> >   Hardware End of Medium = no;
> >   Fast Forward Space File = no;
> >   BSF at EOM = yes;
> >   TWO EOF = yes;
> > 
> > btape test completes fine and no longer recommend the bsr related
> > tweaks.  Maybe they're not necessary and btape should be modified?
> > Don't know about fill results yet - running now.  Hopefully I'll
> > encounter the magic incantation to get my autochanger to actually
> > change tapes...
> 
> In the current Bacula tape driver, Backward Space Record is only used
to
> verify that the last block has been written correctly when a tape gets
full,
> it is not strictly necessary.  In addition, the setting of Backward
Space File
> is ignored (bsf must work).
> 
> You might be able to use these settings on some FBSD tape drives:
> 
>   Fast Forward Space File = yes
>   BSF at EOM = no
>   TWO EOF = no
> 
> __Martin

Thanks. I know that it's optional and not needed.  But it seems btape's
suggestions regarding sd.conf tweaks may be a bit inconsistent depending
upon what flags are already set.

This is all a bit confusing on FBSD because per mtio man page various
calls referenced in sd.conf docs are supported, e.g. mtiocget, mtbsr,
mtfsf, etc.  Whether they're supported by SCSI device/driver remains
uncertain, as I don't know how to test.  In this case I'm using LSI
controller and mpt driver.  My assumption was that since btape said I
needed sd flags such as Backward Space Record, etc. that it did not.
But it looks like it does.  So I'm wondering whether the issue is with
btape erroneously reporting or the mpt driver.

To recap, btape and manual backups and restores work but Autochanger is
not changing tapes when it should so I suspected something was not
quite right in sd.conf.

To follow up, btape completed fine.  In this case drive is ARCVault12
w/LTO-2 Drive.  Drive in this one is an HP.  But since HP does no
longer make their own drives who knows...



-- 
Best regards,

Ken Gunderson
GPG Key -- 9F5179FD

"Never hold discussions with the monkey when the organ grinder is in
the room." - Sir Winston Churchill



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

Message: 19
Date: Tue, 10 Jul 2007 10:50:46 -0700
From: Craig White <[EMAIL PROTECTED]>
Subject: [Bacula-users] confused by automatic purging
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain

I have a 3 week rotation with a full backup each Friday and differential
backups every weekday.

Last night, upon execution, the 1_Monday_Week_1 AND 1_Monday_Week_2 were
both purged and it shouldn't have purged 1_Monday_Week_2

from my backup logs...

09-Jul 20:00 SRV1: Start Backup JobId 91,
Job=SRV1Backup.2007-07-09_20.00.00
09-Jul 20:00 SRV1: Pruned 3 Jobs on Volume "1_Monday_Week_1" from
catalog.
09-Jul 20:00 SRV1: ua_purge.c:611 All records pruned from Volume
"1_Monday_Week_1"; marking it "Purged"
09-Jul 20:00 SRV1: ua_purge.c:611 All records pruned from Volume
"1_Monday_Week_2"; marking it "Purged"
09-Jul 20:00 SRV1: Recycled volume "1_Monday_Week_1"
09-Jul 20:00 SRV1: Recycled volume "1_Monday_Week_1" on device
"DDS-4" (/dev/nst0), all previous data lost.

My MondayPool is defined as...
Pool {
  Name = MondayPool
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Volume Retention = 19d
  Volume Use Duration = 4d
  Maximum Volume Jobs = 5
}

and according to list media...
| MediaId | VolumeName      | VolStatus | Enabled | VolBytes       |
VolFiles | VolRetention | Recycle | Slot | InChanger | MediaType |
LastWritten         |
+---------+-----------------+-----------+---------+----------------+----
------+--------------+---------+------+-----------+-----------+---------
------------+
|       2 | 1_Monday_Week_2 | Purged    |       1 | 25,815,250,944 |
26 |    1,641,600 |       1 |    0 |         1 | DDS-4     | 2007-06-25
22:07:04 |

My 1_Monday_Week_2 was last written on 6/25/2007 which is clearly not 19
days of Volume Retention.

Is there something wrong with my understanding of this?

# rpm -qa|grep bacula
bacula-mysql-2.0.3-1
bacula-mtx-2.0.3-1
bacula-wxconsole-2.0.3-1
bacula-gconsole-2.0.3-1

-- 
Craig White <[EMAIL PROTECTED]>




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

Message: 20
Date: Tue, 10 Jul 2007 18:59:37 +0100
From: Martin Simmons <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] btape and bsr sd.conf tweaks
To: [EMAIL PROTECTED]
Cc: [email protected]
Message-ID: <[EMAIL PROTECTED]>

>>>>> On Tue, 10 Jul 2007 09:10:24 -0600, Ken Gunderson said:
> 
> On Tue, 10 Jul 2007 11:12:35 +0100
> Martin Simmons <[EMAIL PROTECTED]> wrote:
> 
> > >>>>> On Mon, 9 Jul 2007 20:27:23 -0600, Ken Gunderson said:
> > > 
> > > Greetings:
> > > 
> > > Since my baffled post I've been doing some testing w/various
sd.conf
> > > tweaks.  btape results and fbsd ports message initially
recommended:
> > > 
> > > Due to lack of some features in the FreeBSD tape driver
> > > implementation you MUST add some OS dependent options to
> > > the bacula-sd.conf file:
> > > 
> > >   Hardware End of Medium = no;
> > >   Backward Space Record  = no;
> > >   Backward Space File    = no;
> > > 
> > > With 2 filemarks at EOT (see man mt):
> > >   Fast Forward Space File = no;
> > >   BSF at EOM = yes;
> > >   TWO EOF    = yes;
> > > 
> > > 
> > > While btape completes w/above no problem, from man mt it looks
like FBSD
> > > supports bsr, however, and a few tests from command line seem to
> > > confirm so I trimmed sd.conf down thusly:
> > > 
> > >   Hardware End of Medium = no;
> > >   Fast Forward Space File = no;
> > >   BSF at EOM = yes;
> > >   TWO EOF = yes;
> > > 
> > > btape test completes fine and no longer recommend the bsr related
> > > tweaks.  Maybe they're not necessary and btape should be modified?
> > > Don't know about fill results yet - running now.  Hopefully I'll
> > > encounter the magic incantation to get my autochanger to actually
> > > change tapes...
> > 
> > In the current Bacula tape driver, Backward Space Record is only
used to
> > verify that the last block has been written correctly when a tape
gets full,
> > it is not strictly necessary.  In addition, the setting of Backward
Space File
> > is ignored (bsf must work).
> > 
> > You might be able to use these settings on some FBSD tape drives:
> > 
> >   Fast Forward Space File = yes
> >   BSF at EOM = no
> >   TWO EOF = no
> > 
> > __Martin
> 
> Thanks. I know that it's optional and not needed.  But it seems
btape's
> suggestions regarding sd.conf tweaks may be a bit inconsistent
depending
> upon what flags are already set.
> 
> This is all a bit confusing on FBSD because per mtio man page various
> calls referenced in sd.conf docs are supported, e.g. mtiocget, mtbsr,
> mtfsf, etc.  Whether they're supported by SCSI device/driver remains
> uncertain, as I don't know how to test.  In this case I'm using LSI
> controller and mpt driver.  My assumption was that since btape said I
> needed sd flags such as Backward Space Record, etc. that it did not.
> But it looks like it does.  So I'm wondering whether the issue is with
> btape erroneously reporting or the mpt driver.
> 
> To recap, btape and manual backups and restores work but Autochanger
is
> not changing tapes when it should so I suspected something was not
> quite right in sd.conf.

AFAIK, problems with the mechanics of tape changing in the autochanger
will
not be affected by any of the mtio options.  The changer is a separate
device
with its own protocol.

__Martin



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

Message: 21
Date: Tue, 10 Jul 2007 11:15:47 -0700
From: "Richard White" <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] New Exabyte Autochanger question
To: "tomasz dereszynski" <[EMAIL PROTECTED]>
Cc: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=US-ASCII

> what means u should stick it in place like  /etc/devfs.conf on FreeBSD

This is SUSE, so I have no /etc/devfs.conf.

Well, my problem is not with Bacula itself. I loaded a tape into the
drive and issued this at the command line:

# mt -f /dev/nst0 rewind
# mt: /dev/nst0: Input/output error

So the computer itself can't see the drive. However, commands issued to
the autchanger work just fine. I loaded a tape into the drive with this:

# ./mtx-changer /dev/sg1 load 1 /dev/nst0 0

Unloading works as well.

So I re-booted. After loading a tape, mt shows normal results. Still,
*label barcode has the same input/output errors.

After loading a tape, mt -f /dev/nst0 status shows this:
drive type = Generic SCSI-2 tape
drive status = -211392892
sense key error = 0
residue count = 0
file number = 0
block number = 0
Tape block size 1024 bytes. Density code 0x82 (unknown).
Soft error count since last status=01
General status bits on (41010000):
  BOT ONLINE IM_REP_EN

It does not make a CR-LF at the end of the last line, so the command
prompt appears right after '_EN'.

The tape drive /dev/nst0 passes the mt and tar test (tar cvf /dev/nst0 .
followed by a rewind and tar tvf /dev/nst0), so the OS seems to be happy
with the drive.

OK, Bacula will label the volumes one at a time. From bconsole I issue a
label command, select the autochanger device from the storage resource
list, give it the name of the tape in slot 1, and select the Daily pool.
Here's what bacula tells me:

Connecting to Storage daemon Auto at bacula-test:9103 ...
Sending label command for Volume "A0000001" Slot 0 ...
Invalid slot=0 defined, cannot autload Volume.
3301 Issuing autochanger "loaded? drive 0" command.
3302 Auchanger"loaded? drive 0", result is Slot 1
block.c:275 volume data error at 0:0! Wanded ID: "BB02", got "". Buffer
discarded.
3000 OK label. VolBytes=64512 DVD=0 Volume="A0000001" Device="VXA3"
(/dev/nst0)
Catalog record for Volume "A0000001", Slot 0 successfully creatged.
Requesting to mount VXA3 ...
3001 Device "VXA3" (/dev/nst0) is mounted with Volume "A0000001"
You have messages.
*m
10-Jul 08:55 bacula-dir:  Error: bnet.c:439 Write error sending 57 bytes
to client:127:0.0.2:36131:  ERR=Broken pipe

If, however, I issue this command:

*label storage=Auto volume=A0000002 slot=2

Bacula deals with it flawlessly. I do get the I/O error mentioned in the
manual, but otherwise it's OK. Much ado about nothing, I suppose, but
I'm under some pressure here.

Thanks to all.
-- 

Tks n rgds,
Richard White CNE6, Linux+
Network Engineer
Mason County, Washington
360-427-5501
[EMAIL PROTECTED]




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

Message: 22
Date: Tue, 10 Jul 2007 14:11:49 -0400
From: "John Drescher" <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] confused by automatic purging
To: "Craig White" <[EMAIL PROTECTED]>
Cc: [email protected]
Message-ID:
        <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="iso-8859-1"

On 7/10/07, Craig White <[EMAIL PROTECTED]> wrote:
>
> I have a 3 week rotation with a full backup each Friday and
differential
> backups every weekday.
>
> Last night, upon execution, the 1_Monday_Week_1 AND 1_Monday_Week_2
were
> both purged and it shouldn't have purged 1_Monday_Week_2
>
> from my backup logs...
>
> 09-Jul 20:00 SRV1: Start Backup JobId 91,
> Job=SRV1Backup.2007-07-09_20.00.00
> 09-Jul 20:00 SRV1: Pruned 3 Jobs on Volume "1_Monday_Week_1" from
> catalog.
> 09-Jul 20:00 SRV1: ua_purge.c:611 All records pruned from Volume
> "1_Monday_Week_1"; marking it "Purged"
> 09-Jul 20:00 SRV1: ua_purge.c:611 All records pruned from Volume
> "1_Monday_Week_2"; marking it "Purged"
> 09-Jul 20:00 SRV1: Recycled volume "1_Monday_Week_1"
> 09-Jul 20:00 SRV1: Recycled volume "1_Monday_Week_1" on device
> "DDS-4" (/dev/nst0), all previous data lost.
>
> My MondayPool is defined as...
> Pool {
>   Name = MondayPool
>   Pool Type = Backup
>   Recycle = yes
>   AutoPrune = yes
>   Volume Retention = 19d
>   Volume Use Duration = 4d
>   Maximum Volume Jobs = 5
> }
>
>
Quick question. Have you changed this pool definition after labeling
these
tapes? As each tape has an independent Volume Retention that is stored
in
the database when the tape is labeled.  This value  is  taken from the
loaded pool definition at that time.

John
-------------- next part --------------
An HTML attachment was scrubbed...

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

Message: 23
Date: Tue, 10 Jul 2007 11:28:13 -0700
From: "Richard White" <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] New Exabyte Autochanger question
To: <[email protected]>,       "tomasz dereszynski"
        <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=US-ASCII

I spoke too soon. If I issue this command:

*label storage=Auto volume=A0000002 slot=2

the command will fail. If, however, I load slot 2 into the drive (of
course, it will already be there if I issued the above command; if I
haven't, I can load it with './mtx-changer /dev/sg1 load 2 /dev/nst0 0'
at the shell prompt), then issue:

*label storage=VXA volume=A0000002

That will work.

Now that I have my tapes labeled, I need to figure out how to do a
backup with this library.
-- 

Tks n rgds,
Richard White CNE6, Linux+
Network Engineer
Mason County, Washington
360-427-5501
[EMAIL PROTECTED]




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

Message: 24
Date: Tue, 10 Jul 2007 20:59:31 +0200
From: Andreas Helmcke <[EMAIL PROTECTED]>
Subject: Re: [Bacula-users] Pruning to the scratch pool.
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1

Falk Sauer wrote:
> I mean on any post in the last weeks i've read (only with one eye)
over a 
> option thats allows a volume after pruning, automatically moving to
the 
> scratch pool.
> 
> That can be a very nice feature, in the moment i make this manually.
> 
> From which version of bacula is this option usable and how is the
syntax?
> 

>From the chapter: "The Pool Resource" from the actual bacula manual:

 The Pool Resource

 RecyclePool = <pool-resource-name>
      On versions 2.1.4 or greater, Bacula can recycle media in the pool
of
your choice. The most useful setup is to use Scratch.


Version 2.1.4 is not quite right ;-). This option is under rework by
Kern just
at the moment so it will hopefully be usable with next release (2.1.26
(beta
version) or 2.2.0 (release version))

> And how is the pruning working in the case of reuse a volume while the
sd 
> needs a pruned volume and have more outtimed volumes as he needs in
the 
> moment? Comes here the sd first and only momentarily unused volumes
going to 
> the scratch pool or moves the pruning algo all pruned volumes to the
scratch 
> pool and the sd reuse one of the volumes after them from scratch pool?

This is one of the points, which are changing at the moment ;-)

But in general bacula will try to prune volumes only when absolutely
necessary
and prefer the requested pool. So it will most likely use volumes from
the
"wrong" pool only when there are no usable volumes in neither the
"correct"
pool nor the scratch pool.

> Why i ask? If the sd picks a self pruned volume, the volume believe in
the 
> same pool and the use structure of the volume is every time  the same 
> (monthly, weekly ...), if every time move the prune algo all of the
volumes 
> to the scratch pool*, we become a better [stochastic/more evenly] use 
> structure of the volumes when the volumes rotate between the pools.
> 
> *) i've noticed that's the scratch pool works like a fifo.

Also it might be nice this is not a (main) feature of the pruning
algorithm.
But improving the tapemanagement is on the wishlist for bacula (see Item
25 on
the projekts-list: http://www.bacula.org/?page=projects )


Andreas




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

Message: 25
Date: Tue, 10 Jul 2007 15:01:50 -0400
From: "Shon Stephens" <[EMAIL PROTECTED]>
Subject: [Bacula-users] Unsure of btape output when testing
        autochanger
To: [email protected]
Message-ID:
        <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="iso-8859-1"

I am not sure of the expected behaviour of btape and am following the
User
Guide in configuring, testing my autochanger (Exabyte Magnum 224 LTO-3)
with
Bacula. I have already tested the tape device with mt and the changer
with
mtx. The mtx-changer script is working as described by the guide. I
start
with the tapes loaded in the magazine of the autochanger and no tape in
the
drive. If I run the first test

btape -c bacula-sd.conf /dev/rmt/0cbn
Tape block granularity is 1024 bytes.
btape: butil.c:286 Using device: "/dev/rmt/0cbn" for writing.
10-Jul 14:37 btape: 3301 Issuing autochanger "loaded? drive 0" command.
10-Jul 14:37 btape: 3302 Autochanger "loaded? drive 0", result: nothing
loaded.
10-Jul 14:37 btape: Fatal Error at device.c:280 because:
dev open failed: dev.c:425 Unable to open device "Ultrium-TD3"
(/dev/rmt/0cbn): ERR=I/O error
10-Jul 14:37 btape: btape Fatal error: butil.c:195 Cannot open
"Ultrium-TD3"
(/dev/rmt/0cbn)

It seems that btape is detecting that no tape is loaded, but doesn't try
to
load one. Is this its correct behavior?
I load the tape in Slot 1 into the drive directly with mtx and try
again.

btape -c bacula-sd.conf /dev/rmt/0cbn
Tape block granularity is 1024 bytes.
btape: butil.c:286 Using device: "/dev/rmt/0cbn" for writing.
10-Jul 14:51 btape: 3301 Issuing autochanger "loaded? drive 0" command.
10-Jul 14:51 btape: 3302 Autochanger "loaded? drive 0", result is Slot
1.
10-Jul 14:51 btape: 3301 Issuing autochanger "loaded? drive 0" command.
10-Jul 14:51 btape: 3302 Autochanger "loaded? drive 0", result is Slot
1.
btape: btape.c:368 open device "Ultrium-TD3" (/dev/rmt/0cbn): OK

This seems to work, so I run the "test" command in btape

btape: btape.c:1348 Now forward spacing 1 more file.
We should be in file 5. I am at file 5. This is correct!

=== End Forward space files test ===

Ah, I see you have an autochanger configured.
To test the autochanger you must have a blank tape
 that I can write on in Slot 1.

Do you wish to continue with the Autochanger test? (y/n):

=== Autochanger test ===

3301 Issuing autochanger "loaded" command.
Slot 1 loaded. I am going to unload it.
3302 Issuing autochanger "unload 1 0" command.
unload status=OK 0
3303 Issuing autochanger "load 1 0" command.
3303 Autochanger "load 1 0" status is OK.
10-Jul 14:55 btape: 3301 Issuing autochanger "loaded? drive 0" command.
10-Jul 14:55 btape: 3302 Autochanger "loaded? drive 0", result is Slot
1.
btape: btape.c:368 open device "Ultrium-TD3" (/dev/rmt/0cbn): OK
btape: btape.c:1235 Rewound "Ultrium-TD3" (/dev/rmt/0cbn)
btape: btape.c:1242 Wrote EOF to "Ultrium-TD3" (/dev/rmt/0cbn)

The test autochanger worked!!

So it seems that btape can issue a load command, and unload properly. So
whats the deal? Should btape has loaded a tape in the first place when
it
didn't detect one?

Thanks,
Shon
-------------- next part --------------
An HTML attachment was scrubbed...

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

------------------------------------------------------------------------
-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

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

_______________________________________________
Bacula-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-users


End of Bacula-users Digest, Vol 15, Issue 9
*******************************************

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Bacula-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-users

Reply via email to