Hi!
when using a new pool (file storage, is that the same?) for the first
time i get this error:
Fatal error: catreq.c:487 Attribute create error. Pool record not
found in Catalog.
do i have to initiate the pool? how do i do that?
/andreas
John Drescher escribió:
How long was each tape used?
Forget that I see what you are talking about. I think this warrants a
bug report.
John
Thanks for the reply. After updating the tapes, this weekend didn't
happen, possibly it's going to be hard to reproduce...
Cheers,
--
José E.
Looks similar to the issue I reported last April:
Date: Fri, 11 Apr 2008 11:08:06 +0200
Subject: tape use period counted from when use should have started instead of
when it did
Had your job been running for more than 12 hours at the time this
happened? Then this behaviour is apparently
Hi list,
I've got a quite simple question regarding some output
modification... since I'm managing my backup clients with FQDN (for
many different reasons), the output of status director is more or
less unreadable because the names are longer than the default column
with, for example:
Scheduled
Hi,
I have two newbie questions..
1. While all my verify jobs work manualy, none seem to work when scheduled...
Each day I get Verify Fatal Error on my 4 jobs.
Start Verify JobId=719 Level=Incremental Job=filer_Verify.2009-03-14_05.00.57
Fatal error: verify.c:277 Unimplemented Verify level
Hi folks,
thanks for your suggestions, I tried your tar suggestion and indeed it
turns out that transfer rates drop to the dozens of kb/sec in one
special directory stored on the ocfs2 filesystem.
I'm now in contact with the ocfs2 devs on the users list to see if
they have any suggestions.
All
Hi folks,
I'd like to add a 2nd tape library to one of our bacula servers which
is running out of space. Is it possible to define jobs in such a way
that bacula itself will take care of which storage the job actually
ends up in / on?
My idea was to label the tapes in the new library so that
On Mon, Mar 16, 2009 at 9:16 AM, Uwe Schuerkamp hoo...@nionex.net wrote:
Hi folks,
I'd like to add a 2nd tape library to one of our bacula servers which
is running out of space. Is it possible to define jobs in such a way
that bacula itself will take care of which storage the job actually
On Mon, Mar 16, 2009 at 4:36 AM, Andreas Schuldei
schuldei+bacula-us...@spotify.com wrote:
Hi!
when using a new pool (file storage, is that the same?) for the first
time i get this error:
Fatal error: catreq.c:487 Attribute create error. Pool record not
found in Catalog.
do i have to
Hello,
how can I recover my files and database?
After fixing some mysql-issues I can't recover my old files. So I tried
the tool bextract with no luck :-(
I did a myisamchk and mysql found some errors and fixed it.
So it begun:
After my vacation the backup had a 'no space left on device'. The
I'm wondering if there is a way to resume (rather than restart from the
beginning) a backup job that has aborted with an error?
My situation is that I am backing up one client over an unreliable
connection (an SSH tunnel that goes through a cable modem). A full
backup of this client would take
Is there a way in bconsole to run a job from a schedule?
Of course I could use the run command, and then modify all the
parameters. But what I'd like to be able to do is instead to pick a
schedule entry and have it run with all the values configured the way
the schedule would configure them.
Hi to everybody !
Kevin, you're saying that by reducing the retention times, old backups are
being deleted. But, according to the documentation, *Job Retention*
affects only records in the catalog and not data in your archive backup.
See
On Mon, Mar 16, 2009 at 3:04 PM, John Drescher dresche...@gmail.com wrote:
On Mon, Mar 16, 2009 at 4:36 AM, Andreas Schuldei
schuldei+bacula-us...@spotify.com wrote:
Hi!
when using a new pool (file storage, is that the same?) for the first
time i get this error:
Fatal error: catreq.c:487
Yes, you are right. I was sloppy in how I phrased it. First of all, we
are talking about volume retention here. Job retention does affect
volume retention, but only indirectly.
Second, you are right, technically speaking, the old backups aren't
deleted. Instead, the same volume is reused. When
On Fri, 13 Mar 2009 23:15:39 -0600, Devin Reade g...@gno.org said:
DR Am I correct that the only way to add 'speed=1' to the growisofs
DR command is to modify the dvd-handler script?
Yep. The script has lots of hard coded values. I also recommend using
-dvd-compat as long as you're modifying
Hi All,
Has anyone tried GPG encrypted backup on tapes?
If yes, can you please point me to some documentation.
GPG encryption works fine on my server, just want to embed it in bacula,
and encrypt the files for offisite storage,
Kind regards,
Prashant
First, a word of caution: I did this when I still used tar for backups.
Eventually I had to revert to unencrypted backups because the backup was
horrendously slow. Eliminating the GPG encryption sped up a full backup
of my server from 3 days to an hour or two. BTW, part of the slowdown is
due
Andreas Schuldei wrote:
On Mon, Mar 16, 2009 at 3:04 PM, John Drescher dresche...@gmail.com wrote:
On Mon, Mar 16, 2009 at 4:36 AM, Andreas Schuldei
schuldei+bacula-us...@spotify.com wrote:
Hi!
when using a new pool (file storage, is that the same?) for the first
time i get this
I'm doing backups on hard drive so bacula write everything in one big file
and I want to externalize my backups for security.
How can I do it without sending each time the growing file to the other
datacenter ?
I also want to keep a copy in each datacenter.
I was thinking of a diff between the
On Fri, Mar 13, 2009 at 5:13 PM, John Drescher dresche...@gmail.com wrote:
On Fri, Mar 13, 2009 at 4:46 PM, Victor Hugo dos Santos
[...]
Did you install the same bacula version?
mmm.. I have clients with distinct versions:
1.38.11 = 10% (debian 4.0)
2.2.8 = 20% (RHEL 4 and windows)
2.4.2 =
On Mon, Mar 16, 2009 at 11:59 AM, kevin_am ke...@alphamosa.fr wrote:
I'm doing backups on hard drive so bacula write everything in one big file
and I want to externalize my backups for security.
How can I do it without sending each time the growing file to the other
datacenter ?
First step:
On Mon, Mar 16, 2009 at 4:47 PM, Kevin Keane subscript...@kkeane.com wrote:
Andreas Schuldei wrote:
On Mon, Mar 16, 2009 at 3:04 PM, John Drescher dresche...@gmail.com wrote:
On Mon, Mar 16, 2009 at 4:36 AM, Andreas Schuldei
schuldei+bacula-us...@spotify.com wrote:
when using a new pool
On Mon, 16 Mar 2009 07:47:18 -0700, Kevin Keane wrote:
I'm wondering if there is a way to resume (rather than restart from the
beginning) a backup job that has aborted with an error?
My situation is that I am backing up one client over an unreliable
connection (an SSH tunnel that goes
On Mon, 16 Mar 2009 08:59:54 -0700, kevin_am wrote:
I'm doing backups on hard drive so bacula write everything in one big
file and I want to externalize my backups for security.
How can I do it without sending each time the growing file to the other
datacenter ?
I also want to keep a copy
Andreas Schuldei wrote:
On Mon, Mar 16, 2009 at 4:47 PM, Kevin Keane subscript...@kkeane.com wrote:
Andreas Schuldei wrote:
On Mon, Mar 16, 2009 at 3:04 PM, John Drescher dresche...@gmail.com wrote:
On Mon, Mar 16, 2009 at 4:36 AM, Andreas Schuldei
On Mon, Mar 16, 2009 at 12:35 PM, Andreas Schuldei
schuldei+bacula-us...@spotify.com wrote:
On Mon, Mar 16, 2009 at 4:47 PM, Kevin Keane subscript...@kkeane.com wrote:
Andreas Schuldei wrote:
On Mon, Mar 16, 2009 at 3:04 PM, John Drescher dresche...@gmail.com wrote:
On Mon, Mar 16, 2009 at
On Mon, Mar 16, 2009 at 12:33 PM, Victor Hugo dos Santos
listas@gmail.com wrote:
On Fri, Mar 13, 2009 at 5:13 PM, John Drescher dresche...@gmail.com wrote:
On Fri, Mar 13, 2009 at 4:46 PM, Victor Hugo dos Santos
[...]
Did you install the same bacula version?
mmm.. I have clients with
Thomas Mueller wrote:
On Mon, 16 Mar 2009 07:47:18 -0700, Kevin Keane wrote:
I'm wondering if there is a way to resume (rather than restart from the
beginning) a backup job that has aborted with an error?
My situation is that I am backing up one client over an unreliable
connection (an
On Mon, Mar 16, 2009 at 1:02 PM, Kevin PREVOST ke...@alphamosa.fr wrote:
I'm ok to make others volumes but a backup for 1 machin is approximatively
50Gb and I have 30 machins to backup.
So you are trying to backup locally then move the volume file(s) to
some other remote storage instead of just
I'm ok to make others volumes but a backup for 1 machin is
approximatively 50Gb and I have 30 machins to backup.
Le 16 mars 09 à 17:32, John Drescher a écrit :
On Mon, Mar 16, 2009 at 11:59 AM, kevin_am ke...@alphamosa.fr wrote:
I'm doing backups on hard drive so bacula write everything in
Thanks for the reply John,
Here's my hosts file:
r...@daffy:~# cat /etc/hosts
127.0.0.1 localhost
127.0.1.1 daffy.gateway.2wire.net daffy
192.168.1.73mickey.gateway.2wire.net mickey
I'm sure the file daemon is started. Here's what I get with a status fd
from console
*status fd
2009/3/16 Chris Cameron chrisacame...@gmail.com:
Thanks for the reply John,
Here's my hosts file:
r...@daffy:~# cat /etc/hosts
127.0.0.1 localhost
127.0.1.1 daffy.gateway.2wire.net daffy
This is the problem. Fix this with the external ip address instead of
127.0.0.1. The reason
In that case, use Maximum Volume Jobs = 1, and also limit the size of a
volume.
This way, you will end up with several individual files for each of your
backups (it's a lot easier to manage 10 5GB files than 1 50GB file), and
you can also be sure that you will never have two jobs in the same
Hello all.
I have Bacula 2.4.2 running on openSUSE 11.1, and I seem to have developed a
tape issue I can't seem to resolve.
I needed to remove some full tapes from the library and replace them with empty
ones, and then take those new tapes and do a rewind/weof/label on them. It
seems Bacula
A router has nothing to do with the loopback address. When you talk
about a router supporting loopback, it refers to a very different
feature that has nothing to do with the problem here. When you use an IP
address that starts with 127, traffic will never get to the router in
the first place;
2009/3/16 Chris Cameron chrisacame...@gmail.com:
Does the router not supporting loop-back make a difference?
---
No. If bacula is only listening and accepting connections on an
ipaddress that external machines can not connect to then there is
On Mon, Mar 16, 2009 at 3:07 PM, Christopher Dick jcdi...@yahoo.com wrote:
Hello all.
I have Bacula 2.4.2 running on openSUSE 11.1, and I seem to have developed a
tape issue I can't seem to resolve.
I needed to remove some full tapes from the library and replace them with
empty ones, and
I have one client that I am backing up through an SSH tunnel. This turns
out to be quite unreliable; the connection errors out after a while.
Sometimes it lasts a whole hour, but in my most recent attempt, the
connection collapsed after only 7 minutes. I have never been able to
complete a full
Does the router not supporting loop-back make a difference?
---
Chris Cameron
On Mon, Mar 16, 2009 at 1:47 PM, John Drescher dresche...@gmail.com wrote:
2009/3/16 Chris Cameron chrisacame...@gmail.com:
Thanks for the reply John,
Here's
GPG directly to Bacula tapes? - Can't do.
Default encryption module that is already included with Bacula using
OpenSSL with certificates works just fine.
If using GPG is a requirement:
1. Backup to a file based volume
2. Encrypt backup volume with GPG
3. Write backup.gpg to the tape (tar or
On Mon, Mar 16, 2009 at 10:41 AM, Kevin Keane subscript...@kkeane.com wrote:
First, a word of caution: I did this when I still used tar for backups.
Eventually I had to revert to unencrypted backups because the backup was
horrendously slow. Eliminating the GPG encryption sped up a full backup
Kevin Keane wrote:
Is there a way in bconsole to run a job from a schedule?
enter run and chose fom the list of jobs defined in bacula-dir.conf
Of course I could use the run command, and then modify all the
parameters. But what I'd like to be able to do is instead to pick a
schedule entry
And it is not entered in the honstname-fd.conf file.
I am having a brain fart atm and can not remember where I entered/how I
told the windows client FD what password it uses to talk to the
bacula-director.
In all my working windows client hostname-fd.conf files the password
line is;
Password
44 matches
Mail list logo
