Backup Central

Hi,

I wonder why I see situations that a client is waiting for another job
to complete, that is only despooling, i. e. that does not block any
resources on the client:

53100 Increme <job 1> SD despooling Data
53101 Increme <job 2> is waiting on max Client jobs

This is with bacula 5.0.3 as distributed with Fedora 15.

Settings:

- in FD and SD: Maximum Concurrent Jobs = 20
- in Job: Maximum Concurrent Jobs = 6
- in Client: Maximum Concurrent Jobs = 1

I am aware that I'm limiting the number of jobs on the client to 1,
and this is intentional. But the "SD despooling Data" is something
that involves the DIR and the SD only, so it should not block the
client from starting the next backup job.

Seems I'm missing something here. Any ideas are highly welcome.

Thanks in advance.


Best regards,

Wolfgang Denk

--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd < at > denx.de
Severe culture shock results when experts from another protocol suite
[...] try to read OSI documents. The term "osified" is used to refer
to such documents. [...] Any relationship to the word "ossified" is
purely intentional. - Marshall T. Rose

------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev
_______________________________________________
Bacula-users mailing list
Bacula-users < at > lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

On Thu, Apr 5, 2012 at 8:18 AM, Wolfgang Denk <wd < at > denx.de> wrote:

Hi,

I wonder why I see situations that a client is waiting for another job
to complete, that is only despooling, i. e. that does not block any
resources on the client:

 53100 Increme  <job 1> SD despooling Data
 53101 Increme  <job 2> is waiting on max Client jobs

This is with bacula 5.0.3 as distributed with Fedora 15.

Settings:

- in FD and SD: Maximum Concurrent Jobs = 20
- in Job:       Maximum Concurrent Jobs = 6
- in Client:    Maximum Concurrent Jobs = 1

I am aware that I'm limiting the number of jobs on the client to 1,
and this is intentional.  But the "SD despooling Data" is something
that involves the DIR and the SD only, so it should not block the
client from starting the next backup job.

Seems I'm missing something here.  Any ideas are highly welcome.

Thanks in advance.

Do you have any restrictions on how many jobs go per volume?

John

------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev
_______________________________________________
Bacula-users mailing list
Bacula-users < at > lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Dear John Drescher,

In message <CAEhu1-7L0Q=jNQqBf_d_5VPaRHxsNKN-Szh7Zh_h=Nk8Op-vRw < at > mail.gmail.com> you wrote:

Do you have any restrictions on how many jobs go per volume?

No.

Best regards,

Wolfgang Denk

--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd < at > denx.de
What is research but a blind date with knowledge? -- Will Harvey

------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev
_______________________________________________
Bacula-users mailing list
Bacula-users < at > lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

On Thu, Apr 5, 2012 at 2:19 PM, Wolfgang Denk <wd < at > denx.de> wrote:
Dear John Drescher,

In message <CAEhu1-7L0Q=jNQqBf_d_5VPaRHxsNKN-Szh7Zh_h=Nk8Op-vRw < at > mail.gmail.com> you wrote:

Do you have any restrictions on how many jobs go per volume?

No.

Is the same volume used by both clients? I mean you are not using a
different pool per client or something like that?

John

------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev
_______________________________________________
Bacula-users mailing list
Bacula-users < at > lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

In the message dated: Thu, 05 Apr 2012 14:18:22 +0200,
The pithy ruminations from Wolfgang Denk on
<[Bacula-users] parallelizing jobs> were:
=>
=> Hi,
=>
=> I wonder why I see situations that a client is waiting for another job
=> to complete, that is only despooling, i. e. that does not block any
=> resources on the client:

This has been discussed several times. Check the list archives for "concurrent
spooling":

https://www.google.com/search?q=bacula+mailing+list+concurrent+spooling

Unfortunately, there does not seem to be a solution.

Mark

=>
=> 53100 Increme <job 1> SD despooling Data
=> 53101 Increme <job 2> is waiting on max Client jobs
=>
=> This is with bacula 5.0.3 as distributed with Fedora 15.
=>
=> Settings:
=>
=> - in FD and SD: Maximum Concurrent Jobs = 20
=> - in Job: Maximum Concurrent Jobs = 6
=> - in Client: Maximum Concurrent Jobs = 1
=>
=> I am aware that I'm limiting the number of jobs on the client to 1,
=> and this is intentional. But the "SD despooling Data" is something
=> that involves the DIR and the SD only, so it should not block the
=> client from starting the next backup job.
=>
=> Seems I'm missing something here. Any ideas are highly welcome.
=>
=> Thanks in advance.
=>
=>
=> Best regards,
=>
=> Wolfgang Denk
=>
=> --
=> DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
=> HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
=> Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd < at > denx.de
=> Severe culture shock results when experts from another protocol suite
=> [...] try to read OSI documents. The term "osified" is used to refer
=> to such documents. [...] Any relationship to the word "ossified" is
=> purely intentional. - Marshall T. Rose
=>
=> ------------------------------------------------------------------------------
=> Better than sec? Nothing is better than sec when it comes to
=> monitoring Big Data applications. Try Boundary one-second
=> resolution app monitoring today. Free.
=> http://p.sf.net/sfu/Boundary-dev2dev
=> _______________________________________________
=> Bacula-users mailing list
=> Bacula-users < at > lists.sourceforge.net
=> https://lists.sourceforge.net/lists/listinfo/bacula-users
=>



------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev
_______________________________________________
Bacula-users mailing list
Bacula-users < at > lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Dear John Drescher,

In message <CAEhu1-6y0CYS2sOWuUYg2DRizRy3sTt=PsewHwFCD+AAgJMekQ < at > mail.gmail.com> you wrote:

Do you have any restrictions on how many jobs go per volume?

No.

Is the same volume used by both clients? I mean you are not using a
different pool per client or something like that?

All these jobs use the same pool, so both jobs and thus both clients
will use the same volume. There are other jobs running in parallel,
too (but not as many as to run into the max job limit).

Best regards,

Wolfgang Denk

--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd < at > denx.de
The only perfect science is hind-sight.

------------------------------------------------------------------------------
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2
_______________________________________________
Bacula-users mailing list
Bacula-users < at > lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Dear Mark,

In message <3158.1333653400 < at > localhost> you wrote:

=> I wonder why I see situations that a client is waiting for another job
=> to complete, that is only despooling, i. e. that does not block any
=> resources on the client:

This has been discussed several times. Check the list archives for "concurrent
spooling":

https://www.google.com/search?q=bacula+mailing+list+concurrent+spooling

I'm not sure if this is the same problem. In my case, there is
actualy no concurrency. The job on the client has terminated, the
client says it is running no jobs at this time. Only SD and DIR
are still processing this job.

Best regards,

Wolfgang Denk

--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd < at > denx.de
"An open mind has but one disadvantage: it collects dirt."
- a saying at RPI

------------------------------------------------------------------------------
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2
_______________________________________________
Bacula-users mailing list
Bacula-users < at > lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

On Thu, Apr 5, 2012 at 6:24 PM, Wolfgang Denk <wd < at > denx.de> wrote:
Dear John Drescher,

In message <CAEhu1-6y0CYS2sOWuUYg2DRizRy3sTt=PsewHwFCD+AAgJMekQ < at > mail.gmail.com> you wrote:

Do you have any restrictions on how many jobs go per volume?

No.

Is the same volume used by both clients? I mean you are not using a
different pool per client or something like that?

All these jobs use the same pool, so both jobs and thus both clients
will use the same volume.  There are other jobs running in parallel,
too (but not as many as to run into the max job limit).


I have not seen this however I do not look that closely. I know my
jobs do spool concurrently at least while the jobs are already
started. I use a 5GB spool.

John

------------------------------------------------------------------------------
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2
_______________________________________________
Bacula-users mailing list
Bacula-users < at > lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users