Re: [Bacula-users] Data loss with Hard Links

[Andrea Venturoli]

On 12/15/20 6:49 PM, Martin Simmons wrote:


What is the fileset definition?


Quite huge, both in terms of size (a full backup is around 400GB) and in 
terms of lines (more than 40 ZFS datasets).

I'm pasting here the relevant parts:

FileSet {
  Name="Xx"
  Include {
Options {
  signature = MD5
  Accurate=yes
}
...
File=/usr/jails/mail/.zfs/snapshot/bacula
File=/usr/jails/mail/var/spool/imap/.zfs/snapshot/bacula
...
  }
  Exclude {
...
  }
}





Perhaps you can track down a minimal set of jobs that trigger the problem by
using the "Select backup for a client before a specified time" option.  Start
with a time just after the end of the Full backup.  Does that restore without
errors?

> If so, then increase the time to include more and more
> Differential/Incremental backups until the problem occurs.

Restoring just the Full backup works without any error.
Same for just the Differential job.
As soon as the restore involves both, the errors are back.

I tried a different pair of Full/Differential and I'm seeing similar errors.



Any other hint from here?

This installation is several years old (although it has been upgraded 
along the way) and worked properly in the past.

Also, I tried the same on a different server and incurred in no such errors.
What to check?

 bye & Thanks
av.

P.S.
I'll run another Full this night, wait for a Differential or Incremental 
subsequent backup and try the restore again.



___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Deb packages for mariadb?

[Sven Hartge]

On 18.12.20 15:49, Daniel Rich wrote:

The bacula-director package that comes with Ubuntu 20.04, doesn’t 
include any database support as far as I can tell other than possibly 
sqlite. I couldn’t even get it to start with the database configs I 
currently have.


Yes, because you need the bacula-director-DBTYPE packages to select you 
SQL dialect.


But when installing the bacula-director package Postgres-Support is 
selected as default variant.


You need to manually install bacula-director-mysql if you want that variant.

Grüße,
Sven.



___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Deb packages for mariadb?

[Daniel Rich]

On Dec 17, 2020, 14:25 -0800, Sven Hartge , wrote:
> On 17.12.20 22:49, Daniel Rich wrote:
>
> > The ones that come with Ubuntu 20.04 are nearly a year old (9.4.2)
>
> That is not something we Debian Developers can directly influence. But
> you can take the Debian sources and rebuild them on a Ubuntu-System and
> it will work.

That is likely going to be what I end up doing. But again, they use different 
package names and layout from the bacula.org packages, so I have to convert my 
configs and scripts back to /etc/bacula instead of /opt/bacula as well as 
update my ansible cookbooks to use the Debian package names instead of the 
bacula.org package names.
> > and don’t include mysql support as far as I can tell. I don’t really
> > want to have to covert my configs back to the debian-supplied version
> > since I’ve been running the bacula.org version for over a years.
>
> Nobody uses MySQL anymore, every distribution has converted over to
> MariaDB by now. This will not go away.

Except the packages that are built at bacula.org that still have mysql as 
dependencies. And that’s in the most recent packages for 9.6.6. This is what 
started me down the path of looking for the source package in the first place 
so I can build my own package for 9.6.6.

The bacula-director package that comes with Ubuntu 20.04, doesn’t include any 
database support as far as I can tell other than possibly sqlite. I couldn’t 
even get it to start with the database configs I currently have. I have a vague 
memory that this is what had me convert over to the bacula.org packages from 
the OS provided ones a year or two ago.

> > Since I’m building a new server, I’m considering moving to 11.0, but
> > there aren’t packages for that yet, so I would still need the debian
> > package files to build my own.
>
> I would hold back on 11 for a bit. The database transition from 9 to 11
> is a bit rocky at the moment. So far, the official word on 11 is "it is
> a beta".
>

Hmm… I figured since it was actually announced those issues had been resolved...

Dan Rich 
http://www.employees.org/~drich/
"Step up to red alert!" "Are you sure, sir?
It means changing the bulb in the sign..."
- Red Dwarf (BBC)
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Data loss with Hard Links

[Eric Bollengier via Bacula-users]

Hello,

On 2020-12-18 15:16, Andrea Venturoli wrote:
> On 12/15/20 6:49 PM, Martin Simmons wrote:
> 
>> What is the fileset definition?
> 
> Quite huge, both in terms of size (a full backup is around 400GB) and in
> terms of lines (more than 40 ZFS datasets).
> I'm pasting here the relevant parts:
> 
> FileSet {
>   Name="Xx"
>   Include {
>     Options {
>   signature = MD5
>   Accurate=yes

Here you have a mistake (maybe not so big) in your fileset, the Accurate
directive you want to use is inside the Job definition. At the FileSet
Options level, this is used to control which parameter you want to use
to control if a file should be backed up or not. It is something like
"pins" by default ("yes" or "no" have no meaning). New 11.0 version
should control this field and report an error. You should probably also
have the directive like hardlinks=yes, else, they are not controlled.

Hope it helps!

Best Regards,
Eric


___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Deb packages for mariadb?

[Adolf Belka]

Hi Dan,

Sent from my Desktop Computer

On 18/12/2020 15:49, Daniel Rich wrote:

On Dec 17, 2020, 14:25 -0800, Sven Hartge , wrote:

On 17.12.20 22:49, Daniel Rich wrote:


The ones that come with Ubuntu 20.04 are nearly a year old (9.4.2)


That is not something we Debian Developers can directly influence. But

you can take the Debian sources and rebuild them on a Ubuntu-System and

it will work.


That is likely going to be what I end up doing. But again, they use different 
package names and layout from the bacula.org packages, so I have to convert my 
configs and scripts back to /etc/bacula instead of /opt/bacula as well as 
update my ansible cookbooks to use the Debian package names instead of the 
bacula.org package names.

and don’t include mysql support as far as I can tell. I don’t really

want to have to covert my configs back to the debian-supplied version

since I’ve been running the bacula.org version for over a years.


Nobody uses MySQL anymore, every distribution has converted over to

MariaDB by now. This will not go away.


Except the packages that are built at bacula.org that still have mysql as 
dependencies. And that’s in the most recent packages for 9.6.6. This is what 
started me down the path of looking for the source package in the first place 
so I can build my own package for 9.6.6.



Mariadb works a drop-in replacement for MySQL. So on Arch Linux I have Mariadb 
installed and then I build Bacula from source and specify with-mysql in the 
./configure command. For me everything just worked.  All the mysql names in bin 
are links to the equivalent in mariadb so bacula finds all the mysql files it 
needs. Hope this helps.




The bacula-director package that comes with Ubuntu 20.04, doesn’t include any 
database support as far as I can tell other than possibly sqlite. I couldn’t 
even get it to start with the database configs I currently have. I have a vague 
memory that this is what had me convert over to the bacula.org packages from 
the OS provided ones a year or two ago.

Since I’m building a new server, I’m considering moving to 11.0, but

there aren’t packages for that yet, so I would still need the debian

package files to build my own.


I would hold back on 11 for a bit. The database transition from 9 to 11

is a bit rocky at the moment. So far, the official word on 11 is "it is

a beta".



Hmm… I figured since it was actually announced those issues had been resolved...



It was released but as a beta. The announcement mentions "This is a major new 
release with many new features and a number of changes. Please take care to test this 
code carefully before putting it into production. Although the new features have been 
tested, they have not run in a production environment".




Dan Rich   
http://www.employees.org/~drich/ 
"Step up to red alert!" "Are you sure, sir?
It means changing the bulb in the sign..."
- Red Dwarf (BBC)



___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] Improving Bacula backup performance

[Philip Pemberton via Bacula-users]

Hi all,

I'm trying to improve the performance of my Bacula backups. I have a
configuration with two machines:

  - "A" - a small server running a few web services.
(Intel Celeron J1800 2.4GHz dual-core)

  - "B" - a 9TB NAS with a Quantum Superloader LTO-6 SAS tape robot
(Intel Q6600 3GHz quad-core)


My issues are twofold:

  - Backups of "B" are done by the local Bacula FD/SD/DIR and spooled
onto disk to reduce shoe-shining. The spool limit is 50GB, on a
solid-state disk.
  It takes about 6 minutes to fill the spool file, and between 5 and 7
to write it out to tape.
  This gives an effective data rate (quoted in the log) of about 50MB/s,
but the tape write rate (again, from the log) is closer 100-120MB/s.

  - Backups from A to B take a long time to spool to disk, but the tape
phase goes as fast as the local backup. Bacula reports about 7MB/sec. I
assume something is slowing down the network traffic.


I have a couple of questions --

  - Re. local "B" backups. Bacula seems to be writing to the spool file,
then dumping it to tape. There's no spooling happening when the tape is
being written to.
  Is there any way I can set Bacula up to do "A/B" or "ping-pong"
buffering, or something better than its current 50% duty cycle?
  Otherwise it seems my only


  - Re. slow transfers from "A" to "B". What can I do to speed up the
network transfers between the two?
  I find that SMB and NFS from my workstation to/from "A" or "B" is
quite fast, certainly far higher than the ~7MB/s I'm seeing (quoted in
the Bacula log). I'm not expecting to hit 100MB/s, but I was expecting
better than 7MB/s!


Both A and B are on the same gigabit network switch.
"A" (small server) has an Intel 82574L Ethernet controller.
"B" (NAS) has a Marvell 88E8056 Ethernet controller.


Thanks,
Phil
phil...@philpem.me.uk


___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users