Hi.
Among my backup clients there's a Win2k8R2 virtual machine with MSSQL on it,
that executes a dump script before making a backup.
The script runs for ~30 min and from time to time (20-30% chance) the
backup job fails with:
11-May 04:43 backup-dir JobId 10787: Fatal error: Socket error on
Ru
On 10/05/12 19:53, John Drescher wrote:
> On Thu, May 10, 2012 at 3:50 AM, Alex Lucas
> wrote:
>> Dears,
>>
>> I was trying to get one volume per job with this pool configuration:
>>
>> Pool {
>> Name = File
>> Pool Type = Backup
>> Recycle = no # Bacula can automatically recy
On 05/10/2012 05:16 PM, Troy Kocher wrote:
> Oh wise ones..
>
> Demand on my bacula installation is growing and I need to increase the
> amount of simultaneous inbound data. One option that occurred to me was
> is it possible to launch multiple daemons on different ports of the same
> physical
Oh wise ones..
Demand on my bacula installation is growing and I need to increase the
amount of simultaneous inbound data. One option that occurred to me was
is it possible to launch multiple daemons on different ports of the same
physical server?
All the data is written to disk only, no tape.
I think you have two options...
Probably the simplest and most straight forward would be to create a copy job
for each pool, then a simple SQL select query for that pool to grab the latest
full backup. This seems to be consistent with the rest of your configuration
and probably the recommended
Il giorno gio, 10/05/2012 alle 16.37 +0200, Radosław Korzeniewski ha
scritto:
> It depends how your disk polls are arranged (sorry I don't have your
> very first email). If you have only one pool, then it could be
> achieved by only one job with a proper selection criteria and pattern
> (especiall
Hello,
2012/5/10 Gandalf Corvotempesta
>
> Now the hard part:
>
> how can I copy the latest complete backup from every client to tape?
>
> Should I create a copy job for each client or can I create a single job?
>
> Any help?
>
>
It depends how your disk polls are arranged (sorry I don't have yo
Dear all,
for the installation of Bacula 5.2.5 on our QNAP TS-859-RP+ we followed this
tutorial
http://www.vizoland.com/default.aspx
where we got this compilation error:
==>Entering directory /opt/bacula-5.2.5/src/filed
make[1]: Entering directory
`/share/MD0_DATA/.qpkg/Optware/bacula-5.2.5/src/f
>> http://www.bacula.org/en/dev-manual/main/main/Autochanger_Resource.html
>
> I've read that many times with no succes, but now i've made some little
> progress.
>
We can help if you post the terminal output of your testing with mtx.
John
-
Il giorno gio, 10/05/2012 alle 09.36 -0400, John Drescher ha scritto:
> I would start with the testing the autochanger part of the manual:
>
> http://www.bacula.org/en/dev-manual/main/main/Autochanger_Resource.html
I've read that many times with no succes, but now i've made some little
progress.
On Thu, May 10, 2012 at 9:21 AM, Gandalf Corvotempesta
wrote:
> Il giorno gio, 10/05/2012 alle 07.20 -0400, John Drescher ha scritto:
>> Yes. Your tape functionality will be greatly increased if you have a
>> barcode reader on your archive and use barcode labels.
>
> Ok, i've added a sample barcod
Il giorno gio, 10/05/2012 alle 07.20 -0400, John Drescher ha scritto:
> Yes. Your tape functionality will be greatly increased if you have a
> barcode reader on your archive and use barcode labels.
Ok, i've added a sample barcode to the tape.
Now library is able to detect the tape, but Bacula does
Hi all,
I am a bit confused about this option:
TLS Require = yes
On the client side it is set to "yes", on the server side "no". That
way everything works. But is the communication still encrypted then? If
I set it to "yes" on the server as well then I got an error:
Fatal error: Authorization pr
Hello,
Few days before I had a bad behaviour from bacula. It pruned lots of
backups from my archives ONLY for one client in one pool before a
specified date!.
I couldn't reread tapes because there was a lot of tapes.
I couldn't restore the full catalog dump because lots of jobs had run
since this
> Lately I have large backups of 10 - 15 Tb that we run weekly because it
> takes days to complete on an LT04.
> Some times the backup stops for some reason near completion.
>
> Is there a way to resume from where it stop?
No. I suggest splitting the job into smaller Jobs if possible.
> I mean I
On Thu, May 10, 2012 at 3:50 AM, Alex Lucas
wrote:
> Dears,
>
> I was trying to get one volume per job with this pool configuration:
>
> Pool {
> Name = File
> Pool Type = Backup
> Recycle = no # Bacula can automatically recycle Volumes
> AutoPrune = yes # Prun
> Very stupid question, is the first time that I use
> a tape library: should I put a physical label with
> a barcode on the tape?
>
Yes. Your tape functionality will be greatly increased if you have a
barcode reader on your archive and use barcode labels.
After you do that. use the bacula update
Dears,
This started happening recently:
I run a job and it finishes. Can see it in list jobs and related volume
in list volumes.
I then run another job in 2 minutes -- the volume from the first job
gets auto-pruned!
All my retentions are set to 90 days (file, job and volume) and I have
not change
Hi all,
Lately I have large backups of 10 - 15 Tb that we run weekly because it takes
days to complete on an LT04.
Some times the backup stops for some reason near completion.
Is there a way to resume from where it stop?
I mean I can rewrite one tape (the one that got error) if it is that,
Il giorno mer, 09/05/2012 alle 14.20 -0700, Tim Krieger ha scritto:
> I'm not certain if the TL2000 supports barcodes, or if you are using them, if
> so, you should be able to use the following :
>
> Label barcodes storage=TL2000-LTO-1
>
> To view the tapes in the library :
>
> status slots sto
Il giorno mer, 09/05/2012 alle 19.29 +0100, Martin Simmons ha scritto:
> That explains why the job can't find any appendable volumes in the Tape pool.
>
> Should the job be using the Tape pool or something else?
>
> Or maybe you specified a different pool when you labeled the new volume?
Could y
On 05/10/2012 09:46 AM, Radosław Korzeniewski wrote:
> Media Type attribute indicates Bacula that it can use any "Media Type"
> volume on any "Media Type" storage device.
> If both physical tape libraries has the same "Media Type" set to LTO5
> then both libraries has to be able to mount _any_ LTO5
Dears,
I was trying to get one volume per job with this pool configuration:
Pool {
Name = File
Pool Type = Backup
Recycle = no # Bacula can automatically recycle Volumes
AutoPrune = yes# Prune expired volumes
Volume Retention = 90 days # how long to
Helo,
2012/5/9 Jan Behrend
>
> (...)
> requests manual mounting of the volume
> with the differential from the primary library on a drive in the
> secondary which of course is not possible. Both libraries have the same
> media type "LTO5" configured, whcih is actually true. I hope this is not
> t
24 matches
Mail list logo
