date:20150518

Re: [Bacula-users] Bacula Director CLUSTER

2015-05-18 Thread Ana Emília M . Arruda

Hello Carlo!

On Mon, May 18, 2015 at 3:09 PM, Carlo Filippetto <
carlo.filippe...@gmail.com> wrote:

>
> 2015-05-18 19:55 GMT+02:00 Ana Emília M. Arruda :
>
>> "The new copies can not be used for restoration unless you specifically
>> choose them by JobId."
>
>
> I read it, but propably I don't undertand it in the correct way... (I'm
> Italian)
>

Don't worry :). I'm neither an english native.

>
> I read that I can do a "restore copy" job that restore the original Job.
> You say me that I can do a normal "restore" job and there  select the ID
> of the copy instead of the original Job?
>

Yes you will can do the restore selecting the CopyJobId of the copy job
(this value is returned by a "list copies" from bconsole). Bacula will use
the CopyJobId and the volume used by this copy job to do the restore.

>
> If I'm in the situation to use the site2, I need to restore from the copy,
> because I will have only the secondary SD, not the primary one that is lost
> with all the Servers.
>

You will not have the volumes from site1 on site2, but you will have all
the catalog information, because they will be replicated. Supposing that
you will have an identical catalog and the copy jobs volumes on your site2,
you should propably have no problem for restores.

>
>
>
> Another question:
> May I use the Copy Job to create a "Virtual Full" backup?
>

I see your point here. In the case you need to start using Bacula on site2
because of a disaster on site1, you will need to start incremental and
differential jobs using the copy job. And then use the copy job plus the
diff and incr to have a new Virtual Full. I did some tests and it seems it
works fine. Also, since you will not have the volumes of your original
backup jobs on site2 (i'm supposing you will have on site2 just the volumes
of resulting copy jobs), it could be interesting to not replicate the
original backup jobs into your catalog on site2. This way bacula will
promote the copy jobs to "normal" backup jobs. And you will not have backup
jobs in your catalog for which you do not have the corresponding volumes.
Before starting the use of Bacula from site2 you could simply delete from
catalog in site2 all your jobs Type=F,D,I and all the copy jobs should be
promoted to normal backup jobs.

>
>
>
> Thank you ANA, I have to offer you an Italian's coffee
>

Thank you :)
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] confused about differentials

2015-05-18 Thread Kern Sibbald


  
  
On 18.05.2015 08:28, Luc Van der Veken
  wrote:


  
  
  
  
  
  
  
  
Hi,Kern,
 
Do you mean that exactly as you say it, “FD
that is more recent that the Dir and the SD […] is a big
problem”?
  


Yes, the above is correct.


  

I thought it was the other way around, that a
FD _older_ than dir/SD
would cause problems.
  


No, we attempt to maintain backward compatibility, otherwise, in an
installation of 1000 machines to backup, the user would have to
update *all* the File daemons when a new Dir/SD are employed.


  

 
The way you say it here, the FD Bacula Systems
provides for community users would force everyone who wants
to add Windows clients to an existing configuration, to
upgrade his
dir and SD to 7.0.5.
  


Yes, either you use an old windows binary produced by the community
for version 5.2.10 (or earlier) or if you are running the new 7.0.5
Bacula Director/SD, you can use the new Windows binaries that are at
this moment sold by Bacula Systems (but at a very reasonable price).

Regards,
Kern


  

 
 

  
From: Kern Sibbald
[mailto:k...@sibbald.com] 
Sent: 13 May 2015 18:13
To: Ian Young; Radosław Korzeniewski
Cc: bacula-users
Subject: Re: [Bacula-users] confused about
differentials
  

 

  
  We have never had any version of Bacula in which
  Differential backups were incomplete.  Running with a FD
  that is more recent that the Dir and the SD or a Dir and
  SD that are not identical is a big problem.  Until that is
  fixed it doesn't make much sense to speculate about any
  odd behavior.  Also, if you are doing something very
  unusual in your FileSet or using multiple FileSets in your
  backup, there could be a configuration problem or perhaps
  even an undiscovered bug, but the first thing to do is
  correct any possible version problems you have.
  
  Best regards,
  Kern
  
  On 13.05.2015 15:18, Ian Young wrote:


   
  

  
On 13 May 2015, at 07:20, Radosław
Korzeniewski 
wrote:
  
   
  

  

  

  

  I appear to be seeing
  the same problem with CentOS 6 /
  CentOS 6 combinations:


   

  

  
  
 
  
  
OS versions doesn't matter.
What is important: Bacula Dir/SD vs. File
Daemon versions. The supported configuration
require Dir/SD in the same version every
time and FD not newer.
  

  

  


   


  I should have been clearer. I meant that
  I was apparently seeing the same problem on at least
  one setup where the FD, DIR and SD versions were all
  the same.


   


  

  

  

  
I take your point,
though, that the Director/SD should not
be older than the clients, so I need to
fix that. Fortunately the (virtual)
machine running the Director and Storage
daemons is dedicated to that task, so it
should be relatively easy to build a new
CentOS 7 machine to get 5.2.13.

[Bacula-users] Copy Job && Create a Virtual Full

2015-05-18 Thread Carlo Filippetto

Hi all,
I want to use copy job to copy volumes to a secondary site used as disaster
recovery.

Can I use this copy job to restore a server in the second site?
Can I use Copy Job create Virtual Full Backup?

Thank you
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

2015-05-18 Thread Dimitri Maziuk

On 05/18/2015 12:27 PM, Florian Rist wrote:

> The whole thing about costs probably seams to be a bit ridiculous, but 
> unfortunately there is niter money nor understanding for the need here 
> right now. So I'm trying to figure out a cheap way to get to a working 
> system that can be handled somehow.

A reference point probably irrelevant since you're on the other side of
the pond where prices are different: I just finished setting up a

supermicro 36+2-bay chassis
(http://www.supermicro.com/products/system/4U/6048/SSG-6048R-E1CR36H.cfm_

-- for $7070 came with with 2x8 AMDs, 64GB RAM, 2x 1TB SSD, and 10x 4TB
Seagate NAS HDD.

That adds up to 22TB raidz (raid-5 equivalent) zfs storage that with our
backup sizes should last for 6-8 months. There's 26 empty drive bays for
when we need more space.

Backing up to a single filesystem is low-maintentance with bacula: you
get auto-labeling, auto-recycling, etc.

I have another similar system on my todo list, that one's a (slightly
cheaper) 24+2 with the requirement to archive the backups long-term. So
we're using Josh's vchanger to treat the 3.5 drives as removable
"magazines". This is a bit trickier because you have to plan for disk
failures and bacula has no built-in facility for writing two copies of
the same backup to 2 disks at once. Plus tape changers require labeling
tapes and re-inventorying slots manually, so more operator work.

That's kinda what the hardware costs are like. (c.f. EUR 5K+ tape
library with no tapes nor a computer to hook it up to.)

-- 
Dimitri Maziuk
Programmer/sysadmin
BioMagResBank, UW-Madison -- http://www.bmrb.wisc.edu

signature.asc
Description: OpenPGP digital signature
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula Director CLUSTER

2015-05-18 Thread Carlo Filippetto

2015-05-18 19:55 GMT+02:00 Ana Emília M. Arruda :

> "The new copies can not be used for restoration unless you specifically
> choose them by JobId."


I read it, but propably I don't undertand it in the correct way... (I'm
Italian)

I read that I can do a "restore copy" job that restore the original Job.
You say me that I can do a normal "restore" job and there  select the ID of
the copy instead of the original Job?

If I'm in the situation to use the site2, I need to restore from the copy,
because I will have only the secondary SD, not the primary one that is lost
with all the Servers.



Another question:
May I use the Copy Job to create a "Virtual Full" backup?



Thank you ANA, I have to offer you an Italian's coffee
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula Director CLUSTER

2015-05-18 Thread Ana Emília M . Arruda

On Mon, May 18, 2015 at 2:38 PM, Carlo Filippetto <
carlo.filippe...@gmail.com> wrote:

> Nice!
> but I read something that I don't like
> :(
>
> I read that the copy job is not like a normal Backup Job, it is used to
> 'restore' the original Job unless the original Job is purged (deleted), and
> in this case it is promote.
>

You can restore from copy jobs using "restore copies" from bconsole. This
will show you the jobids for the copy jobs. From Bacula´s manual:

"The new copies can not be used for restoration unless you specifically
choose them by JobId."


> I search a solution where the secondary site may restore the Virtual
> Machine starting from the copy!
>
> How can I do it?
> I have to make a second installation of the DIR and backup the machine
> twice?
>
> Thank you
>
>
>
>
>
>
>
> 2015-05-18 19:34 GMT+02:00 Ana Emília M. Arruda :
>
>> You can "copy" jobs based on various criterias. You can copy all jobs
>> that were backed up into one specific volume, all jobs from a client, etc.
>> There are a few options for that. Basically, you will create a copy job
>> that will copy your backups based on specifically conditions from one pool
>> (where you normally have your backups) to another pool (the next pool
>> defined in the original pool resource used for your original backups).
>>
>> An interesting directive for the migration/copy pool is:
>>
>> PoolUncopiedJobs This selection which copies all jobs from a pool to an
>> other pool which were not copied before is available only for copy Jobs.
>>
>> So, sure you can copy "a volume" (all the jobs that were copied in an
>> original volume) using copy jobs.
>>
>> You´re welcome!
>>
>> Best regards,
>> Ana
>>
>>
>> On Mon, May 18, 2015 at 2:07 PM, Carlo Filippetto <
>> carlo.filippe...@gmail.com> wrote:
>>
>>> Thank to you!
>>> I have another question: with the 'copy' job, SD to SD, I can copy a
>>> Volume or I have to copy every single job?
>>>
>>> Thank you
>>>
>>>
>>> 2015-05-18 19:04 GMT+02:00 Ana Emília M. Arruda 
>>> :
>>>
 Hello Carlo,

 Don´t worry!

 About to copying only the confs and db, I was supposing that on site2
 there would be an identical installation as it is on site1. Since
 clustering is not a builtin Bacula functionality (I´m affraid there is one
 that I´m not aware), I didn´t realize that it could be done between the two
 linux hosts. Without the bacula-dir clustering solution, it would be a
 manual process start bacula-dir on site2 when site1 becomes unavailable.

 Thank you!

 Best regards,
 Ana

 
 I'll put the two director in cluster, mysql in replication, and conf in
 sync.
 If you move only the conf and the db you can't do backup from site2,
 because the machine check the IP address, with cluster I want to sckip this
 problem.


 On Mon, May 18, 2015 at 12:13 PM, Carlo Filippetto <
 carlo.filippe...@gmail.com> wrote:

> Sorry for the delay!
>
>>
>>
>> On Fri, May 15, 2015 at 11:23 AM, Carlo Filippetto <
>> carlo.filippe...@gmail.com> wrote:
>>
>>> Thank you.
>>> I will use the sd2sd solution.
>>>
>>>
>>> 2015-05-15 15:54 GMT+02:00 Davide Giunchi 
>>> :
>>>
 problem: you need a lot of bandiwidth ( = € )! then you must keep
 always sync the servers, periodically check that everithing is working
 good, and then periodically to dr tests.
 benefit: fast dr

>>>
>>>
>>> I read about Virtual Full, I can use it to create a full on remote
>>> site instead off move the whole VM?
>>>
>>
>> Are you talking about Virtual Backup (Vbackup)? It is intended to
>> "consolidate" a full backup, the most recent differential and the 
>> following
>> incremental into one new full backup. This new full backup can be placed 
>> in
>> your site2.
>>
>
> Fantasctic, I read in the new features that I can consolidate not only
> the last, but the backup that I want.
>
>
>
>> I´m not sure if this is what you´re wondering. If you want to keep
>> site2 with the same volume data as site1, you will need daily copy jobs 
>> for
>> this or some other kind of data replication.
>>
>
> Yes, I want to have the same data on site1 and site2, but I can't
> migrate 500Gb of Backup every night
>
> So I will do a first full migration, and after I'll go on only with
> the daily and I will use the consololitate backup (Vbackup) to have a 
> full.
>

>
>
>>
>> The director master stay on site 1, the site 2 is offline since the
>>> firts goes down.
>>>
>>> The master on site 1 know the volumes on site 2 with SD to SD copy?
>>> (I think yes, it is an SD that he manage)
>>>
>>
>> Yes, if you use copy or migrate jobs. Are you also going to copy
>> your volumes data from site1 to site2?
>>
>>>

Re: [Bacula-users] Bacula Director CLUSTER

2015-05-18 Thread Carlo Filippetto

Nice!
but I read something that I don't like
:(

I read that the copy job is not like a normal Backup Job, it is used to
'restore' the original Job unless the original Job is purged (deleted), and
in this case it is promote.

I search a solution where the secondary site may restore the Virtual
Machine starting from the copy!

How can I do it?
I have to make a second installation of the DIR and backup the machine
twice?

Thank you







2015-05-18 19:34 GMT+02:00 Ana Emília M. Arruda :

> You can "copy" jobs based on various criterias. You can copy all jobs that
> were backed up into one specific volume, all jobs from a client, etc. There
> are a few options for that. Basically, you will create a copy job that will
> copy your backups based on specifically conditions from one pool (where you
> normally have your backups) to another pool (the next pool defined in the
> original pool resource used for your original backups).
>
> An interesting directive for the migration/copy pool is:
>
> PoolUncopiedJobs This selection which copies all jobs from a pool to an
> other pool which were not copied before is available only for copy Jobs.
>
> So, sure you can copy "a volume" (all the jobs that were copied in an
> original volume) using copy jobs.
>
> You´re welcome!
>
> Best regards,
> Ana
>
>
> On Mon, May 18, 2015 at 2:07 PM, Carlo Filippetto <
> carlo.filippe...@gmail.com> wrote:
>
>> Thank to you!
>> I have another question: with the 'copy' job, SD to SD, I can copy a
>> Volume or I have to copy every single job?
>>
>> Thank you
>>
>>
>> 2015-05-18 19:04 GMT+02:00 Ana Emília M. Arruda :
>>
>>> Hello Carlo,
>>>
>>> Don´t worry!
>>>
>>> About to copying only the confs and db, I was supposing that on site2
>>> there would be an identical installation as it is on site1. Since
>>> clustering is not a builtin Bacula functionality (I´m affraid there is one
>>> that I´m not aware), I didn´t realize that it could be done between the two
>>> linux hosts. Without the bacula-dir clustering solution, it would be a
>>> manual process start bacula-dir on site2 when site1 becomes unavailable.
>>>
>>> Thank you!
>>>
>>> Best regards,
>>> Ana
>>>
>>> 
>>> I'll put the two director in cluster, mysql in replication, and conf in
>>> sync.
>>> If you move only the conf and the db you can't do backup from site2,
>>> because the machine check the IP address, with cluster I want to sckip this
>>> problem.
>>>
>>>
>>> On Mon, May 18, 2015 at 12:13 PM, Carlo Filippetto <
>>> carlo.filippe...@gmail.com> wrote:
>>>
 Sorry for the delay!

>
>
> On Fri, May 15, 2015 at 11:23 AM, Carlo Filippetto <
> carlo.filippe...@gmail.com> wrote:
>
>> Thank you.
>> I will use the sd2sd solution.
>>
>>
>> 2015-05-15 15:54 GMT+02:00 Davide Giunchi :
>>
>>> problem: you need a lot of bandiwidth ( = € )! then you must keep
>>> always sync the servers, periodically check that everithing is working
>>> good, and then periodically to dr tests.
>>> benefit: fast dr
>>>
>>
>>
>> I read about Virtual Full, I can use it to create a full on remote
>> site instead off move the whole VM?
>>
>
> Are you talking about Virtual Backup (Vbackup)? It is intended to
> "consolidate" a full backup, the most recent differential and the 
> following
> incremental into one new full backup. This new full backup can be placed 
> in
> your site2.
>

 Fantasctic, I read in the new features that I can consolidate not only
 the last, but the backup that I want.



> I´m not sure if this is what you´re wondering. If you want to keep
> site2 with the same volume data as site1, you will need daily copy jobs 
> for
> this or some other kind of data replication.
>

 Yes, I want to have the same data on site1 and site2, but I can't
 migrate 500Gb of Backup every night

 So I will do a first full migration, and after I'll go on only with the
 daily and I will use the consololitate backup (Vbackup) to have a full.

>>>


>
> The director master stay on site 1, the site 2 is offline since the
>> firts goes down.
>>
>> The master on site 1 know the volumes on site 2 with SD to SD copy?
>> (I think yes, it is an SD that he manage)
>>
>
> Yes, if you use copy or migrate jobs. Are you also going to copy your
> volumes data from site1 to site2?
>
> I´m not sure about your needs, but if you´re going to replicate your
> volumes data from site1 to site2, I would prefer to replicate bacula´s
> configuration files and a catalog backup from site1 to site2. And when you
> need to use bacula on site2, you could just start it.
>


 
 I'll put the two director in cluster, mysql in replication, and conf in
 sync.
 If you move only the conf and the db you can't do backup from site2,
 because the machine check the

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

2015-05-18 Thread Florian Rist

Hi Rudolf

> If you want to avoid tape back hitching, you have to have atleast:
> HP LTO-5: 47 MB/sIBM LTO-5: 40 MB/s
> HP LTO-6: 54 MB/sIBM LTO-6: 40 MB/s
> 
> And I am afraid, that single 6 TB drive is not a good solution. It
> can be maybe sufficient if you first  spool the data, and then just
> unspool them to the tape, or using D2D2T, but if you have spooling
> and unspooling at the same time, I do not think that it will
> be considerate to your tape drive.

Thanks for the numbers and pointing me to the spool/unspool problem. 
This can be fixed by using two hard drives, right?

The whole thing about costs probably seams to be a bit ridiculous, but 
unfortunately there is niter money nor understanding for the need here 
right now. So I'm trying to figure out a cheap way to get to a working 
system that can be handled somehow. And I'm fully aware of the fact that 
the hardware investment costs will be a minor part in the actual long 
term overall costs.

See you
Flo


--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula Director CLUSTER

2015-05-18 Thread Ana Emília M . Arruda

You can "copy" jobs based on various criterias. You can copy all jobs that
were backed up into one specific volume, all jobs from a client, etc. There
are a few options for that. Basically, you will create a copy job that will
copy your backups based on specifically conditions from one pool (where you
normally have your backups) to another pool (the next pool defined in the
original pool resource used for your original backups).

An interesting directive for the migration/copy pool is:

PoolUncopiedJobs This selection which copies all jobs from a pool to an
other pool which were not copied before is available only for copy Jobs.

So, sure you can copy "a volume" (all the jobs that were copied in an
original volume) using copy jobs.

You´re welcome!

Best regards,
Ana


On Mon, May 18, 2015 at 2:07 PM, Carlo Filippetto <
carlo.filippe...@gmail.com> wrote:

> Thank to you!
> I have another question: with the 'copy' job, SD to SD, I can copy a
> Volume or I have to copy every single job?
>
> Thank you
>
>
> 2015-05-18 19:04 GMT+02:00 Ana Emília M. Arruda :
>
>> Hello Carlo,
>>
>> Don´t worry!
>>
>> About to copying only the confs and db, I was supposing that on site2
>> there would be an identical installation as it is on site1. Since
>> clustering is not a builtin Bacula functionality (I´m affraid there is one
>> that I´m not aware), I didn´t realize that it could be done between the two
>> linux hosts. Without the bacula-dir clustering solution, it would be a
>> manual process start bacula-dir on site2 when site1 becomes unavailable.
>>
>> Thank you!
>>
>> Best regards,
>> Ana
>>
>> 
>> I'll put the two director in cluster, mysql in replication, and conf in
>> sync.
>> If you move only the conf and the db you can't do backup from site2,
>> because the machine check the IP address, with cluster I want to sckip this
>> problem.
>>
>>
>> On Mon, May 18, 2015 at 12:13 PM, Carlo Filippetto <
>> carlo.filippe...@gmail.com> wrote:
>>
>>> Sorry for the delay!
>>>


 On Fri, May 15, 2015 at 11:23 AM, Carlo Filippetto <
 carlo.filippe...@gmail.com> wrote:

> Thank you.
> I will use the sd2sd solution.
>
>
> 2015-05-15 15:54 GMT+02:00 Davide Giunchi :
>
>> problem: you need a lot of bandiwidth ( = € )! then you must keep
>> always sync the servers, periodically check that everithing is working
>> good, and then periodically to dr tests.
>> benefit: fast dr
>>
>
>
> I read about Virtual Full, I can use it to create a full on remote
> site instead off move the whole VM?
>

 Are you talking about Virtual Backup (Vbackup)? It is intended to
 "consolidate" a full backup, the most recent differential and the following
 incremental into one new full backup. This new full backup can be placed in
 your site2.

>>>
>>> Fantasctic, I read in the new features that I can consolidate not only
>>> the last, but the backup that I want.
>>>
>>>
>>>
 I´m not sure if this is what you´re wondering. If you want to keep
 site2 with the same volume data as site1, you will need daily copy jobs for
 this or some other kind of data replication.

>>>
>>> Yes, I want to have the same data on site1 and site2, but I can't
>>> migrate 500Gb of Backup every night
>>>
>>> So I will do a first full migration, and after I'll go on only with the
>>> daily and I will use the consololitate backup (Vbackup) to have a full.
>>>
>>
>>>
>>>

 The director master stay on site 1, the site 2 is offline since the
> firts goes down.
>
> The master on site 1 know the volumes on site 2 with SD to SD copy? (I
> think yes, it is an SD that he manage)
>

 Yes, if you use copy or migrate jobs. Are you also going to copy your
 volumes data from site1 to site2?

 I´m not sure about your needs, but if you´re going to replicate your
 volumes data from site1 to site2, I would prefer to replicate bacula´s
 configuration files and a catalog backup from site1 to site2. And when you
 need to use bacula on site2, you could just start it.

>>>
>>>
>>> 
>>> I'll put the two director in cluster, mysql in replication, and conf in
>>> sync.
>>> If you move only the conf and the db you can't do backup from site2,
>>> because the machine check the IP address, with cluster I want to sckip this
>>> problem.
>>>
>>> Thank you
>>>
>>>
>>>
>>>
>>
>
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/

Re: [Bacula-users] Bacula Director CLUSTER

2015-05-18 Thread Carlo Filippetto

Thank to you!
I have another question: with the 'copy' job, SD to SD, I can copy a Volume
or I have to copy every single job?

Thank you


2015-05-18 19:04 GMT+02:00 Ana Emília M. Arruda :

> Hello Carlo,
>
> Don´t worry!
>
> About to copying only the confs and db, I was supposing that on site2
> there would be an identical installation as it is on site1. Since
> clustering is not a builtin Bacula functionality (I´m affraid there is one
> that I´m not aware), I didn´t realize that it could be done between the two
> linux hosts. Without the bacula-dir clustering solution, it would be a
> manual process start bacula-dir on site2 when site1 becomes unavailable.
>
> Thank you!
>
> Best regards,
> Ana
>
> 
> I'll put the two director in cluster, mysql in replication, and conf in
> sync.
> If you move only the conf and the db you can't do backup from site2,
> because the machine check the IP address, with cluster I want to sckip this
> problem.
>
>
> On Mon, May 18, 2015 at 12:13 PM, Carlo Filippetto <
> carlo.filippe...@gmail.com> wrote:
>
>> Sorry for the delay!
>>
>>>
>>>
>>> On Fri, May 15, 2015 at 11:23 AM, Carlo Filippetto <
>>> carlo.filippe...@gmail.com> wrote:
>>>
 Thank you.
 I will use the sd2sd solution.


 2015-05-15 15:54 GMT+02:00 Davide Giunchi :

> problem: you need a lot of bandiwidth ( = € )! then you must keep
> always sync the servers, periodically check that everithing is working
> good, and then periodically to dr tests.
> benefit: fast dr
>


 I read about Virtual Full, I can use it to create a full on remote site
 instead off move the whole VM?

>>>
>>> Are you talking about Virtual Backup (Vbackup)? It is intended to
>>> "consolidate" a full backup, the most recent differential and the following
>>> incremental into one new full backup. This new full backup can be placed in
>>> your site2.
>>>
>>
>> Fantasctic, I read in the new features that I can consolidate not only
>> the last, but the backup that I want.
>>
>>
>>
>>> I´m not sure if this is what you´re wondering. If you want to keep site2
>>> with the same volume data as site1, you will need daily copy jobs for this
>>> or some other kind of data replication.
>>>
>>
>> Yes, I want to have the same data on site1 and site2, but I can't migrate
>> 500Gb of Backup every night
>>
>> So I will do a first full migration, and after I'll go on only with the
>> daily and I will use the consololitate backup (Vbackup) to have a full.
>>
>
>>
>>
>>>
>>> The director master stay on site 1, the site 2 is offline since the
 firts goes down.

 The master on site 1 know the volumes on site 2 with SD to SD copy? (I
 think yes, it is an SD that he manage)

>>>
>>> Yes, if you use copy or migrate jobs. Are you also going to copy your
>>> volumes data from site1 to site2?
>>>
>>> I´m not sure about your needs, but if you´re going to replicate your
>>> volumes data from site1 to site2, I would prefer to replicate bacula´s
>>> configuration files and a catalog backup from site1 to site2. And when you
>>> need to use bacula on site2, you could just start it.
>>>
>>
>>
>> 
>> I'll put the two director in cluster, mysql in replication, and conf in
>> sync.
>> If you move only the conf and the db you can't do backup from site2,
>> because the machine check the IP address, with cluster I want to sckip this
>> problem.
>>
>> Thank you
>>
>>
>>
>>
>
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula Director CLUSTER

2015-05-18 Thread Ana Emília M . Arruda

Hello Carlo,

Don´t worry!

About to copying only the confs and db, I was supposing that on site2 there
would be an identical installation as it is on site1. Since clustering is
not a builtin Bacula functionality (I´m affraid there is one that I´m not
aware), I didn´t realize that it could be done between the two linux hosts.
Without the bacula-dir clustering solution, it would be a manual process
start bacula-dir on site2 when site1 becomes unavailable.

Thank you!

Best regards,
Ana


I'll put the two director in cluster, mysql in replication, and conf in
sync.
If you move only the conf and the db you can't do backup from site2,
because the machine check the IP address, with cluster I want to sckip this
problem.


On Mon, May 18, 2015 at 12:13 PM, Carlo Filippetto <
carlo.filippe...@gmail.com> wrote:

> Sorry for the delay!
>
>>
>>
>> On Fri, May 15, 2015 at 11:23 AM, Carlo Filippetto <
>> carlo.filippe...@gmail.com> wrote:
>>
>>> Thank you.
>>> I will use the sd2sd solution.
>>>
>>>
>>> 2015-05-15 15:54 GMT+02:00 Davide Giunchi :
>>>
 problem: you need a lot of bandiwidth ( = € )! then you must keep
 always sync the servers, periodically check that everithing is working
 good, and then periodically to dr tests.
 benefit: fast dr

>>>
>>>
>>> I read about Virtual Full, I can use it to create a full on remote site
>>> instead off move the whole VM?
>>>
>>
>> Are you talking about Virtual Backup (Vbackup)? It is intended to
>> "consolidate" a full backup, the most recent differential and the following
>> incremental into one new full backup. This new full backup can be placed in
>> your site2.
>>
>
> Fantasctic, I read in the new features that I can consolidate not only the
> last, but the backup that I want.
>
>
>
>> I´m not sure if this is what you´re wondering. If you want to keep site2
>> with the same volume data as site1, you will need daily copy jobs for this
>> or some other kind of data replication.
>>
>
> Yes, I want to have the same data on site1 and site2, but I can't migrate
> 500Gb of Backup every night
>
> So I will do a first full migration, and after I'll go on only with the
> daily and I will use the consololitate backup (Vbackup) to have a full.
>

>
>
>>
>> The director master stay on site 1, the site 2 is offline since the firts
>>> goes down.
>>>
>>> The master on site 1 know the volumes on site 2 with SD to SD copy? (I
>>> think yes, it is an SD that he manage)
>>>
>>
>> Yes, if you use copy or migrate jobs. Are you also going to copy your
>> volumes data from site1 to site2?
>>
>> I´m not sure about your needs, but if you´re going to replicate your
>> volumes data from site1 to site2, I would prefer to replicate bacula´s
>> configuration files and a catalog backup from site1 to site2. And when you
>> need to use bacula on site2, you could just start it.
>>
>
>
> 
> I'll put the two director in cluster, mysql in replication, and conf in
> sync.
> If you move only the conf and the db you can't do backup from site2,
> because the machine check the IP address, with cluster I want to sckip this
> problem.
>
> Thank you
>
>
>
>
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

2015-05-18 Thread Florian Rist

Hi Davide

> I will not advise any only-drive LTO solution: if you need to spend
> less money, you could take a look at the Tandbergdata's (opps:
> Overland) 1U storageloader with LTO4 or 5, it's very cheap.
> Any "no library" LTO solutions, will require a lot of operator work,
> it will make your solution quite unusable.

Well, I have to admit that I'm a bit concerned about that, too. I'm the 
one to handle the tapes and I know I will be board quite soon and forget 
to change tapes before I leave and so on.

Tandberg StorageLibrary T40 1x LTO-44400 EUR
Tandberg StorageLibrary T40+ 1x LTO-5   5200 EUR
Tandberg LTO-6  1700 EUR

I didn't think about the tape prices for now, but the library solution 
is at least 2800 EUR more expensive than a single LTO-6 drive.

The problem is, the whole backup thing is not considered to be imported 
here, yet it is, and the present situation is extremely risky and there 
is an obvious need to change things and it should be done before things 
go wrong.


The question is probably, what happens if a tape change is forgotten. 
Say I have scheduled backup job for the night and the tape is not change 
in the evening. Will this mess up everything or will the job 
automatically be run as soon as the needed tape becomes available?

I seams to be possible to get cheap old used libraries. Is it possible 
to change the drive?

I once did this a long time ago with a MOD library system. I was easy to 
do as the library was controlled via RS232 and did not at all 
communicate with the drive. Do modern LTO libraries work the same way?


> The official bacula online bacula documentation (
> http://www.bacula.org/7.0.x-manuals/en/main/index.html ) it's pretty
> good.

In deed it seams to be, I already had a look and it.

See you
Flo

--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] file set changes with accurate mode enabled

2015-05-18 Thread Radosław Korzeniewski

Hello,

2015-05-13 9:35 GMT+02:00 Pavel Bychikhin :

>  Hi Radoslaw,
> Thank you very much for your answer.
> I have one more in this regard. Can you please clarify it for me as well.
> One of my File definitions looks like:
> File = 
In this case if I add some new lines to that file, bacula will not force
> Full backup and will not backup that new files if accurate mode is disabled.
> I feel myself a little bit confused since in this situation bacula doesn't
> follow its logic of file set tracking. Looks like it just ignores such
> filesets while doing incremental backup and relies on a file set definition
> which was stored at the time of last Full backup. And Accurate mode
> magically solves this issue.
>

First of all the Bacula (to be more specific Director) is checking a
fileset changes by computing a md5 digest on the fileset Include/Exclude
definition and stores it in the catalog. If computed digest is the same as
a previous then Director decides that a fileset definition is not changed.
So incremental backup is not changing its level. In your example the
fileset is not changed but a File Daemon is responsible for listing files
to backup. If you make this king of configuration you should know what you
are doing.

best regards
-- 
Radosław Korzeniewski
rados...@korzeniewski.net
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

2015-05-18 Thread Dimitri Maziuk

On 05/18/2015 10:22 AM, Davide Giunchi wrote:

> Any "no library" LTO solutions, will require a lot of operator work,
it will make your solution quite unusable.

On the subject of amazon vtl: back when they announced it, they had a
rationale write-up detailing the real costs of tape backups. Including
things like tape storage costs/risks, real-life write speeds. read/write
contention, drive failures, and so on. Very useful read, unfortunately I
don't seem to be able to find it now. :(

-- 
Dimitri Maziuk
Programmer/sysadmin
BioMagResBank, UW-Madison -- http://www.bmrb.wisc.edu



signature.asc
Description: OpenPGP digital signature
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

2015-05-18 Thread Davide Giunchi

> I don't have the hardware yet. My plan is to get a small cheap server,
> some low cost CPU (some i3 Core), 6 TB HD (for spooling), a SAS
> controller and a LTO 5 or 6 drive (probably from Tandberg). I'll need to
> change tapes manually, any library solution seams to be to expensive.
> Changing tapes for the full backup will be bit annoying probably, and to
> do so I have to run it during the day, but this seams possible as there
> is not much going on on Fridays here.

I will not advise any only-drive LTO solution: if you need to spend less money, 
you could take a look at the Tandbergdata's (opps: Overland) 1U storageloader 
with LTO4 or 5, it's very cheap.
Any "no library" LTO solutions, will require a lot of operator work, it will 
make your solution quite unusable.

> Does any one know the book on Bacula from Philipp Storz? Is it useful to
> get started? I'm thinking about getting it.

The official bacula online bacula documentation ( 
http://www.bacula.org/7.0.x-manuals/en/main/index.html ) it's pretty good.

Regards

--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula Director CLUSTER

2015-05-18 Thread Carlo Filippetto

Sorry for the delay!

>
>
> On Fri, May 15, 2015 at 11:23 AM, Carlo Filippetto <
> carlo.filippe...@gmail.com> wrote:
>
>> Thank you.
>> I will use the sd2sd solution.
>>
>>
>> 2015-05-15 15:54 GMT+02:00 Davide Giunchi :
>>
>>> problem: you need a lot of bandiwidth ( = € )! then you must keep always
>>> sync the servers, periodically check that everithing is working good, and
>>> then periodically to dr tests.
>>> benefit: fast dr
>>>
>>
>>
>> I read about Virtual Full, I can use it to create a full on remote site
>> instead off move the whole VM?
>>
>
> Are you talking about Virtual Backup (Vbackup)? It is intended to
> "consolidate" a full backup, the most recent differential and the following
> incremental into one new full backup. This new full backup can be placed in
> your site2.
>

Fantasctic, I read in the new features that I can consolidate not only the
last, but the backup that I want.



> I´m not sure if this is what you´re wondering. If you want to keep site2
> with the same volume data as site1, you will need daily copy jobs for this
> or some other kind of data replication.
>

Yes, I want to have the same data on site1 and site2, but I can't migrate
500Gb of Backup every night

So I will do a first full migration, and after I'll go on only with the
daily and I will use the consololitate backup (Vbackup) to have a full.



>
> The director master stay on site 1, the site 2 is offline since the firts
>> goes down.
>>
>> The master on site 1 know the volumes on site 2 with SD to SD copy? (I
>> think yes, it is an SD that he manage)
>>
>
> Yes, if you use copy or migrate jobs. Are you also going to copy your
> volumes data from site1 to site2?
>
> I´m not sure about your needs, but if you´re going to replicate your
> volumes data from site1 to site2, I would prefer to replicate bacula´s
> configuration files and a catalog backup from site1 to site2. And when you
> need to use bacula on site2, you could just start it.
>


I'll put the two director in cluster, mysql in replication, and conf in
sync.
If you move only the conf and the db you can't do backup from site2,
because the machine check the IP address, with cluster I want to sckip this
problem.

Thank you
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

2015-05-18 Thread Cejka Rudolf

Florian Rist wrote (2015/05/17):
> I don't have the hardware yet. My plan is to get a small cheap server, 
> some low cost CPU (some i3 Core), 6 TB HD (for spooling), a SAS 
> controller and a LTO 5 or 6 drive (probably from Tandberg). I'll need to 
> change tapes manually, any library solution seams to be to expensive. 

Just try to ask several vendors and look at
http://www.oracle.com/us/products/servers-storage/storage/tape-storage/sl150-modular-tape-library/quantum-comp-i40-i500-wp-1704937.pdf
It is rather old (2012), but it shows that price differences may be
dramatical. For you, table 1 and line 1: Cheaply expandable 30-slot and one
LTO-5 drive SL150 cumulative cost 6000 USD against non-expandable 40-slot and
one LTO-5 drive i40 cumulative cost 13000 USD - but once again, it was 2012.

If you want to avoid tape back hitching, you have to have atleast:
HP LTO-5: 47 MB/sIBM LTO-5: 40 MB/s
HP LTO-6: 54 MB/sIBM LTO-6: 40 MB/s
(I'm sorry, I do not know anything about Tandberg LTOs.)
However, these speeds are just native speeds without compression,
so you have to be unpredictably faster in reality. Also, various
models can be somewhat different and there can be some prediction
logic, which can effectively change these requirements too. So the
best thing is to be as fast as possible :o) And I am afraid, that single
6 TB drive is not a good solution. It can be maybe sufficient if you first
spool the data, and then just unspool them to the tape, or using D2D2T,
but if you have spooling and unspooling at the same time, I do not think
that it will be considerate to your tape drive.

-- 
Rudolf Cejka  http://www.fit.vutbr.cz/~cejkar
Brno University of Technology, Faculty of Information Technology
Bozetechova 2, 612 66  Brno, Czech Republic

--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] confused about differentials

2015-05-18 Thread Luc Van der Veken

Hi Ana Emilia, thanks for clearing that up.

But then I think I may have a problem :(

I am using the Dir and FD from standard Ubuntu 12.04 repositories, 5.2.5.

I only have a few Windows clients. For those I got the Windows FD offered by 
Bacula Systems two years ago, which was at that time 6.0.6. That was the only 
one offered, as far as I remember. Today it would be 7.0.5.

Backups made with that combination run without errors or warnings.
I never had (or tried) to restore a full system, only some individual files, 
but that always worked too, so far.

I am also using community FD 5.2.10 on one machine (my own workstation), that 
looks OK too.

This is some logging output, from a job picked at random, from 6.0.6 FD with 
director/SD 5.2.5.

  Build OS:   x86_64-pc-linux-gnu ubuntu 12.04
  JobId:  29291
  Job:qvsrv.2015-05-15_20.05.01_50
  Backup Level:   Differential, since=2015-04-24 21:37:40
  Client: "qvsrv-fd" 6.0.6 (30Sep12) Microsoft Windows Server 
2008 R2 Standard Edition Service Pack 1 (build 7601), 64-bit,Cross-compile,Win64
  FileSet:"qvsrvSet" 2013-07-30 20:05:01
  Pool:   "File" (From Job DiffPool override)
  Catalog:"MyCatalog" (From Client resource)
  Storage:"File" (From Pool resource)
  Scheduled time: 15-May-2015 20:05:01
  Start time: 16-May-2015 00:53:13
  End time:   16-May-2015 01:37:31
  Elapsed time:   44 mins 18 secs
  Priority:   10
  FD Files Written:   4,044
  SD Files Written:   4,044
  FD Bytes Written:   8,663,741,214 (8.663 GB)
  SD Bytes Written:   8,664,580,063 (8.664 GB)
  Rate:   3259.5 KB/s
  Software Compression:   65.5 %
  VSS:yes
  Encryption: no
  Accurate:   no
  Volume name(s): FileStorage0133|FileStorage0134|FileStorage0135
  Volume Session Id:  284
  Volume Session Time:1430987205
  Last Volume Bytes:  4,089,208,805 (4.089 GB)
  Non-fatal FD errors:0
  SD Errors:  0
  FD termination status:  OK
  SD termination status:  OK
  Termination:Backup OK

I don’t see anything abnormal in it. The backup size may appear small even for 
a differential, but it’s an intranet reporting server with only little activity.


From: Ana Emília M. Arruda [mailto:emiliaarr...@gmail.com]
Sent: 18 May 2015 15:14
To: Luc Van der Veken
Cc: bacula-users
Subject: Re: [Bacula-users] confused about differentials

Hello Luc,

I think there is a misunderstooding here.

On Mon, May 18, 2015 at 3:28 AM, Luc Van der Veken 
mailto:luc...@wimionline.com>> wrote:
Hi,Kern,

Do you mean that exactly as you say it,

“FD that is more recent that the Dir and the SD […] is a big problem”?
I thought it was the other way around, that a FD _older_ than dir/SD would 
cause problems.



“FD that is more recent that the Dir and the SD […] is a big problem”
This is exactly what Kern said. Older versions of FD working with newer Dir and 
SD vesions are not a problem. Instead, "FD more recent than Dir and SD" are.


The way you say it here, the FD Bacula Systems provides for community users 
would force everyone who wants to add Windows clients to an existing 
configuration, to upgrade his dir and SD to 7.0.5.

FD versions older than Dir and SD are not a problem. I have a 5.2.10 Windows 
client working with a 7.0.5 Dir and SD. If you have 5.2.10 windows clients 
(the most recent community version), you should have Dir and SD running at 
least version 5.2.10.

Regards,
Ana



From: Kern Sibbald [mailto:k...@sibbald.com]
Sent: 13 May 2015 18:13
To: Ian Young; Radosław Korzeniewski
Cc: bacula-users
Subject: Re: [Bacula-users] confused about differentials


We have never had any version of Bacula in which Differential backups were 
incomplete.  Running with a FD that is more recent that the Dir and the SD or a 
Dir and SD that are not identical is a big problem.  Until that is fixed it 
doesn't make much sense to speculate about any odd behavior.  Also, if you are 
doing something very unusual in your FileSet or using multiple FileSets in your 
backup, there could be a configuration problem or perhaps even an undiscovered 
bug, but the first thing to do is correct any possible version problems you 
have.

Best regards,
Kern

On 13.05.2015 15:18, Ian Young wrote:

On 13 May 2015, at 07:20, Radosław Korzeniewski 
mailto:rados...@korzeniewski.net>> wrote:

I appear to be seeing the same problem with CentOS 6 / CentOS 6 combinations:


OS versions doesn't matter. What is important: Bacula Dir/SD vs. File Daemon 
versions. The supported configuration require Dir/SD in the same version every 
time and FD not newer.

I should have been clearer. I meant that I was apparently seeing the same 
problem on at least one setup where the FD, DIR and SD versions were all the 
same.

I take your

Re: [Bacula-users] FD external

2015-05-18 Thread Josh Fisher


On 5/14/2015 7:16 AM, kinomakino wrote:


Dear friends, first of all, thanks for everything.

Bacula infrastructure have working for my internal backups within the LAN.
Now I want to make a backup of an external VPS.

NAT have done in my LAN ports 9102 9101 9103.
He tells me not to find the DS.
in my bacula-sd.conf, the storage points to a local ip
in my bacula-dir.conf the storage points to a local ip

Then how  can i configure my bacula to perform local backups within 
the LAN, and add equipment that  are on the Internet.?




Bacula does have TLS encryption capabilities, so backup over a direct 
remote connection in a safe manner is possible. The DIR machine must be 
able to connect to the FD machine's TCP port 9102 and the FD must be 
able to connect to the SD machine's TCP port 9103. That said, in many 
cases the remote FD will not have a static IP address or will be behind 
a firewall. I have found that it is easier to connect remote FDs through 
an OpenVPN VPN that is configured to assign the FD an IP that is 
reachable from the DIR.



--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] confused about differentials

2015-05-18 Thread Ana Emília M . Arruda

Hello Luc,

I think there is a misunderstooding here.

On Mon, May 18, 2015 at 3:28 AM, Luc Van der Veken 
wrote:

>  Hi,Kern,
>
>
>
> Do you mean that exactly as you say it,
> 
> “FD that is more recent that the Dir and the SD […] is a big problem”?
>
> I thought it was the other way around, that a FD _*older*_ than dir/SD
> would cause problems.
>
>


“FD that is more recent that the Dir and the SD […] is a big problem”
This is exactly what Kern said. Older versions of FD working with newer Dir
and SD vesions are not a problem. Instead, "FD more recent than Dir and SD"
are.


>
>
> The way you say it here, the FD Bacula Systems provides for community
> users would force everyone who wants to add Windows clients to an existing
> configuration, to upgrade his dir and SD to 7.0.5.
>

FD versions older than Dir and SD are not a problem. I have a 5.2.10
Windows client working with a 7.0.5 Dir and SD. If you have 5.2.10 windows
clients (the most recent community version), you should have Dir and SD
running at least version 5.2.10.

Regards,
Ana


>
>
>
>
> *From:* Kern Sibbald [mailto:k...@sibbald.com]
> *Sent:* 13 May 2015 18:13
> *To:* Ian Young; Radosław Korzeniewski
> *Cc:* bacula-users
> *Subject:* Re: [Bacula-users] confused about differentials
>
>
>
>
> We have never had any version of Bacula in which Differential backups were
> incomplete.  Running with a FD that is more recent that the Dir and the SD
> or a Dir and SD that are not identical is a big problem.  Until that is
> fixed it doesn't make much sense to speculate about any odd behavior.
> Also, if you are doing something very unusual in your FileSet or using
> multiple FileSets in your backup, there could be a configuration problem or
> perhaps even an undiscovered bug, but the first thing to do is correct any
> possible version problems you have.
>
> Best regards,
> Kern
>
> On 13.05.2015 15:18, Ian Young wrote:
>
>
>
>  On 13 May 2015, at 07:20, Radosław Korzeniewski <
> rados...@korzeniewski.net> wrote:
>
>
>
>   I appear to be seeing the same problem with CentOS 6 / CentOS 6
> combinations:
>
>
>
>
>
> OS versions doesn't matter. What is important: Bacula Dir/SD vs. File
> Daemon versions. The supported configuration require Dir/SD in the same
> version every time and FD not newer.
>
>
>
> I should have been clearer. I meant that I was apparently seeing the same
> problem on at least one setup where the FD, DIR and SD versions were all
> the same.
>
>
>
> I take your point, though, that the Director/SD should not be older
> than the clients, so I need to fix that. Fortunately the (virtual) machine
> running the Director and Storage daemons is dedicated to that task, so it
> should be relatively easy to build a new CentOS 7 machine to get 5.2.13.
>
>
>
> Recommended version in May 2015 is Bacula 7.0.5, not 5.2.13.
>
>
>
> I understand that, but deploying 7.0.5 would be significantly harder in my
> environment than moving to 5.2.13 so it would be something of a last resort.
>
>
>
> I don't think you're saying that I need to move to the latest version to
> get reliable backups, are you?
>
>
>
> If anyone knew of a bug in 5.2.x (or for that matter in 5.0.x) that caused
> differentials to be incomplete, I'd obviously feel differently (but then, I
> imagine Red Hat would too, as 5.2.13 is what they are shipping in their
> most recent release).
>
>
>
> I don't think I actually have a version mismatch problem (as I'm
> seeing the same issue with matched versions), but there are all sorts of
> reasons this might make my problem go away: there may be a bug in the
> version of 5.0 shipped with RHEL/CentOS, or I may have a configuration
> problem. Either way, starting from scratch and transitioning clients over
> may help.
>
>
>
> First of all. Did you ever test that it is not working?
>
>
>
> As stated, I have seen significant data loss when attempting to restore a
> production system. This is not theoretical, although perhaps the subject
> line led you astray.
>
>
>
>Differential backup does not backup a files which were deleted in the
> mean time. So in real system it is very unlikely (it must meet a specific
> conditions) you get the same number of files backed up in Incremental and
> Differential levels.
>
>
>
> I don't believe this is the problem I'm seeing. For example, from my
> original mail:
>
>
>
> | 9,929 | srv-c701-backup | 2015-05-01 23:05:04 | B| I |   91,030
> | 3,450,906,888 | T
> | 9,938 | srv-c701-backup | 2015-05-02 23:05:03 | B| D |  112
> | 2,184,302 | T
>
>
>
> The 91,000 files in job 9929 were NOT all deleted before job 9938 was run,
> but do not appear in that job.
>
>
>
> -- Ian
>
>
>
>
>
>
>
>
>
>  
> --
>
> One dashboard for servers and applications across Physical-Virtual-Cloud
>
> Widest out-of-the-box monitoring support with 50+ applications
>
> Performance metrics, stats and reports that give you Acti

Re: [Bacula-users] Unable to send mail with mutt as mailcommand

2015-05-18 Thread Heitor Faria

Hey Kelvin/Alex, 

You could also just set postfix to send ssl/tls authenticated emails: 
http://www.bacula.com.br/?p=1160 

Regards, 
=== 
Heitor Medrado de Faria - LPIC-III | ITIL-F | Bacula Systems Certified 
Administrator II 
15 a 26 de junho: Treinamento Telepresencial Bacula: 
http://www.bacula.com.br/?p=2174 
61 8268-4220 
Site: www.bacula.com.br | Facebook: heitor.faria | Gtalk: heitorfa...@gmail.com 
 

> De: "Alex Domoradov" 
> Para: "Kelvin Minter" 
> Cc: "bacula-users" 
> Enviadas: Segunda-feira, 18 de maio de 2015 4:54:05
> Assunto: Re: [Bacula-users] Unable to send mail with mutt as mailcommand

> You can use swaks to send emails with ssl/tls. It's a very useful command line
> utility which I have used with nagios

> /usr/bin/printf "$MESSAGE$" | /usr/bin/swaks -4 --server 
> smtp.googlemail.com:587
> --from no-re...@example.net --to nagios-ad...@example.net -tls --auth PLAIN
> --auth-user no-re...@example.net --auth-password 7654321 --h-Subject
> "$SUBJECT$" --body - --hide-all

> The other way is to setup stunnel, if your command line utility doesn't 
> support
> ssl/tls, but email server require it.

> On Thu, May 14, 2015 at 8:40 PM, Kelvin Minter < kb.min...@gmail.com > wrote:

>> I am now seeing logs within the maillog after following Bill's advice.

>> Thank you for your replys and happy coding!

>> On Thu, May 14, 2015 at 9:04 AM, Richard < 
>> lists-bac...@listmail.innovate.net >
>> wrote:

>>>  Original Message 
>>> > Date: Thursday, May 14, 2015 09:11:10 AM -0400
>>> > From: Bill Arlofski < waa-bac...@revpol.com >

>>> > On 05/14/2015 01:33 AM, Kelvin Minter wrote:
>>> >> Hello I am running bacula 5.2.13 on CentOS 6.6.

>>> >> I have the following configuration for my messages in
>>> >> /etc/bacula/bacula-dir.conf

>>> >> Messages {
>>> >> Name = Standard
>>> >> mailcommand = "/usr/bin/mutt -F /root/.muttrc-bacula -s
>>> >> \"Bacula: %t %e of %c %l\" %r"
>>> >> operatorcommand = "/usr/bin/mutt -F /root/.muttrc-bacula -s
>>> >> \"Bacula: Intervention needed for %j\" %r"

>>> >> mail = myem...@domain.com  = all,
>>> >> !skipped operator = myem...@domain.com
>>> >>  = mount console = all, !skipped,
>>> >> !saved

>>> >> append = "/var/log/bacula/bacula.log" = all, !skipped
>>> >> catalog = all
>>> >> }

>>> >> When I run mutt on the command line it proceeds and I am able to
>>> >> receive the mail at my destination address.
>>> >> However, when running a job I receive no email upon completion of
>>> >> the job and no log is created within bacula or maillog.

>>> >> I am using tls for postfix but this should be no issue as I can
>>> >> run the above mailcommand from the command line and receive an
>>> >> email.

>>> >> Using bsmtp (Unable to send because I need to run through tls)
>>> >> will produce an erroneous entry in the maillog file.

>>> >> I am stumped on how to debug further to determine where the mail
>>> >> is being caught up at. It seems to have to be something on
>>> >> bacula's end.

>>> >> Any help would be appreciated.

>>> > Hi Kelvin,

>>> > I would first check to see what user that the bacula-dir daemon is
>>> > running as:

>>> ># ps axuf | grep bacula-dir

>>> > On some distributions, the director runs as user "bacula", on
>>> > others (like Gentoo) it runs as user "root". The director does not
>>> > need to run as root, but some distributions package Bacula so that
>>> > it does.

>>> > You are pointing mutt to a config file in the root user's home
>>> > directory, perhaps mutt, running as "bacula" when called by the
>>> > director simply can not read the config file and exits?


>>> > Bill

>>> >From your slightly obscured configuration, I'm assuming that you're
>>> trying to deliver the mail directly from bacula to your remote mail
>>> server, not localhost, which is why you have the bsmtp/TLS issue. I
>>> believe that if you use the default Bacula bsmtp setup to deliver to
>>> localhost and then alias the recipient off (with an entry in
>>> /etc/aliases) you should be ok. That approach will invoke the
>>> (postfix) MTA -- which will handle TLS just fine. [and avoids the
>>> issue of trying to configure a different mailer.] E.g.,

>>> mailcommand = "/usr/sbin/bsmtp -h localhost -f ...

>>> mail = @localhost = all, !skipped
>>> operator = @localhost = mount

>>> which is the default Bacula configuration and what I use on my
>>> centos box.

>>> - Richard

>>> --
>>> One dashboard for servers and applications across Physical-Virtual-Cloud
>>> Widest out-of-the-box monitoring support with 50+ applications
>>> Performance metrics, stats and reports that give you Actionable Insights
>>> Deep dive visibility with transaction tracing using APM Insight.
>>> http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
>>> __

Re: [Bacula-users] Bacula Amazon S3 off-site Backups

2015-05-18 Thread Ana Emília M . Arruda

Yes Alan. I had suggested it and this certainly solve this issue. But we´re
trying to adjust the timeouts.

On Mon, May 18, 2015 at 8:05 AM, Alan Brown  wrote:

> On 17/05/15 02:22, Ana Emília M. Arruda wrote:
>
>  I'm aware of a case of Bacula working with an Amazon VTL service. We
>> are having occasional drive timeout issues. IMHO this is causing some
>> issues with tapes being unloaded to different slots than that they were
>> originally.
>>
>
> If this is the case, adding an "update slots" to the mount/unmount scripts
> would solve it, at cost of adding a few seconds to the process.
>
>
>
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

2015-05-18 Thread Ana Emília M . Arruda

Hello Florian,

OK. So you will not need to much retention periods. With two fulls +
differentials you have enough.

On Sun, May 17, 2015 at 12:22 PM, Florian Rist  wrote:

>
> > Your full backups could be divided into smaller ones to fill in one
> > tape.
>
> Is this something that I would have to take care of? I guess Bracula can
> automatically span backups over multiple tapes.
>

Yes. Bacula can automatically span backups over multiple tapes. In the
case of using a standalone tape drive, you will need to change tapes
manually. So, the idea of spliting your backups into smaller ones would
allow you to have nightly full backups and not daily ones. And you would
just need to change the tapes every day for the next nightly full backup
job.

>
> > And you could run daily differential and full backups from
> > different data. This way you could use a low cost storage array and
> > your tape drive (which is your technology?).
>
> I don't have the hardware yet. My plan is to get a small cheap server,
> some low cost CPU (some i3 Core), 6 TB HD (for spooling), a SAS
> controller and a LTO 5 or 6 drive (probably from Tandberg). I'll need to
> change tapes manually, any library solution seams to be to expensive.
> Changing tapes for the full backup will be bit annoying probably, and to
> do so I have to run it during the day, but this seams possible as there
> is not much going on on Fridays here.
>

OK. In this case, I would do a not very usual configuration, but I´m quite
sure that would work for you:

1) Have full backups running into tapes. One job per night using a
"splited" fileset that fits in one tape.
2) Daily differential backup jobs for what is not being backup up in the
full backup job going into the disk.

Supposing you are going to use LTO-6 tapes and the native capacity for
this technology, in 8 days you will have a full set of your full backup
jobs. And the other days you will have differential ones in disk.

>
> Does any one know the book on Bacula from Philipp Storz? Is it useful to
> get started? I'm thinking about getting it.
>

I would recommend you to look for tutorials on Internet. There are lots of
them.

>
> See you
> Flo
>
> Regards,
Ana

>
>
>
>
>
> --
> One dashboard for servers and applications across Physical-Virtual-Cloud
> Widest out-of-the-box monitoring support with 50+ applications
> Performance metrics, stats and reports that give you Actionable Insights
> Deep dive visibility with transaction tracing using APM Insight.
> http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
> ___
> Bacula-users mailing list
> Bacula-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>
--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula Amazon S3 off-site Backups

2015-05-18 Thread Alan Brown

On 17/05/15 02:22, Ana Emília M. Arruda wrote:

> I'm aware of a case of Bacula working with an Amazon VTL service. We
> are having occasional drive timeout issues. IMHO this is causing some
> issues with tapes being unloaded to different slots than that they were
> originally.

If this is the case, adding an "update slots" to the mount/unmount 
scripts would solve it, at cost of adding a few seconds to the process.





--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Unable to send mail with mutt as mailcommand

2015-05-18 Thread Alex Domoradov

You can use swaks to send emails with ssl/tls. It's a very useful command
line utility which I have used with nagios

/usr/bin/printf "$MESSAGE$" | /usr/bin/swaks -4 --server
smtp.googlemail.com:587 --from no-re...@example.net --to
nagios-ad...@example.net -tls --auth PLAIN --auth-user no-re...@example.net
--auth-password 7654321 --h-Subject "$SUBJECT$" --body - --hide-all

The other way is to setup stunnel, if your command line utility doesn't
support ssl/tls, but email server require it.

On Thu, May 14, 2015 at 8:40 PM, Kelvin Minter  wrote:

> I am now seeing logs within the maillog after following Bill's advice.
>
> Thank you for your replys and happy coding!
>
> On Thu, May 14, 2015 at 9:04 AM, Richard <
> lists-bac...@listmail.innovate.net> wrote:
>
>>
>>
>>  Original Message 
>> > Date: Thursday, May 14, 2015 09:11:10 AM -0400
>> > From: Bill Arlofski 
>> >
>> > On 05/14/2015 01:33 AM, Kelvin Minter wrote:
>> >> Hello I am running bacula 5.2.13 on CentOS 6.6.
>> >>
>> >> I have the following configuration for my messages in
>> >> /etc/bacula/bacula-dir.conf
>> >>
>> >> Messages {
>> >>   Name = Standard
>> >>   mailcommand = "/usr/bin/mutt -F /root/.muttrc-bacula -s
>> >>   \"Bacula: %t %e of %c %l\" %r"
>> >>   operatorcommand = "/usr/bin/mutt -F /root/.muttrc-bacula -s
>> >>   \"Bacula: Intervention needed for %j\" %r"
>> >>
>> >>   mail = myem...@domain.com  = all,
>> >>   !skipped operator = myem...@domain.com
>> >>    = mount console = all, !skipped,
>> >>   !saved
>> >>
>> >>   append = "/var/log/bacula/bacula.log" = all, !skipped
>> >>   catalog = all
>> >> }
>> >>
>> >> When I run mutt on the command line it proceeds and I am able to
>> >> receive the mail at my destination address.
>> >> However, when running a job I receive no email upon completion of
>> >> the job and no log is created within bacula or maillog.
>> >>
>> >> I am using tls for postfix but this should be no issue as I can
>> >> run the above mailcommand from the command line and receive an
>> >> email.
>> >>
>> >> Using bsmtp (Unable to send because I need to run through tls)
>> >> will produce an erroneous entry in the maillog file.
>> >>
>> >> I am stumped on how to debug further to determine where the mail
>> >> is being caught up at. It seems to have to be something on
>> >> bacula's end.
>> >>
>> >> Any help would be appreciated.
>> >
>> > Hi Kelvin,
>> >
>> > I would first check to see what user that the bacula-dir daemon is
>> > running as:
>> >
>> ># ps axuf | grep bacula-dir
>> >
>> > On some distributions, the director runs as user "bacula", on
>> > others (like Gentoo) it runs as user "root". The director does not
>> > need to run as root, but some distributions package Bacula so that
>> > it does.
>> >
>> > You are pointing mutt to a config file in the root user's home
>> > directory, perhaps mutt, running as "bacula" when called by the
>> > director simply can not read the config file and exits?
>> >
>> >
>> > Bill
>>
>> >From your slightly obscured configuration, I'm assuming that you're
>> trying to deliver the mail directly from bacula to your remote mail
>> server, not localhost, which is why you have the bsmtp/TLS issue. I
>> believe that if you use the default Bacula bsmtp setup to deliver to
>> localhost and then alias the recipient off (with an entry in
>> /etc/aliases) you should be ok. That approach will invoke the
>> (postfix) MTA -- which will handle TLS just fine. [and avoids the
>> issue of trying to configure a different mailer.] E.g.,
>>
>>   mailcommand = "/usr/sbin/bsmtp -h localhost -f ...
>>
>>   mail = @localhost = all, !skipped
>>   operator = @localhost = mount
>>
>> which is the default Bacula configuration and what I use on my
>> centos box.
>>
>>
>>- Richard
>>
>>
>>
>>
>> --
>> One dashboard for servers and applications across Physical-Virtual-Cloud
>> Widest out-of-the-box monitoring support with 50+ applications
>> Performance metrics, stats and reports that give you Actionable Insights
>> Deep dive visibility with transaction tracing using APM Insight.
>> http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
>> ___
>> Bacula-users mailing list
>> Bacula-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/bacula-users
>>
>
>
>
> --
> One dashboard for servers and applications across Physical-Virtual-Cloud
> Widest out-of-the-box monitoring support with 50+ applications
> Performance metrics, stats and reports that give you Actionable Insights
> Deep dive visibility with transaction tracing using APM Insight.
> http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
> ___
> Bacula-users mailing list
> Bacula-users@lists.sourceforge.net
> http

[Bacula-users] How to compile bacula with LZO

2015-05-18 Thread ladolf

Thank you both for the answer and sorry for late response. I successfuly 
compiled bacula with LZO support.

The conclusion is: 
To compile bacula with LZO support install LZO development libraries 
(liblzo2-dev on Ubuntu) and compile bacula as usual, which means wihout 
--with-lzo (this parameter should probably be omitted when running "./configure 
--help").

+--
|This was sent by lado.lan...@gmail.com via Backup Central.
|Forward SPAM to ab...@backupcentral.com.
+--



--
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula Director CLUSTER

Re: [Bacula-users] confused about differentials

[Bacula-users] Copy Job && Create a Virtual Full

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

Re: [Bacula-users] Bacula Director CLUSTER

Re: [Bacula-users] Bacula Director CLUSTER

Re: [Bacula-users] Bacula Director CLUSTER

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

Re: [Bacula-users] Bacula Director CLUSTER

Re: [Bacula-users] Bacula Director CLUSTER

Re: [Bacula-users] Bacula Director CLUSTER

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

Re: [Bacula-users] file set changes with accurate mode enabled

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

Re: [Bacula-users] Bacula Director CLUSTER

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

Re: [Bacula-users] confused about differentials

Re: [Bacula-users] FD external

Re: [Bacula-users] confused about differentials

Re: [Bacula-users] Unable to send mail with mutt as mailcommand

Re: [Bacula-users] Bacula Amazon S3 off-site Backups

Re: [Bacula-users] Is Bacula and a LTO drive right for me?

Re: [Bacula-users] Bacula Amazon S3 off-site Backups

Re: [Bacula-users] Unable to send mail with mutt as mailcommand

[Bacula-users] How to compile bacula with LZO

26 matches

Site Navigation

Mail list logo

Footer information