Good morning, we've got a debian sarge server running bacula-director 3.0.2.3
It's been running for about 4 years now but just recently it's started
developing a problem where it will stop for no apparent reason. There's no
error messages, nothing that I can find in the logs (although there's
Good morning, I'm running Bacula 5 on a Debian server and mostly, it's bloody
brilliant. However, there are four servers in two locations outside of our data
centre that I'm having massive problems backing up.
One location needs to backup 10 files with a total size of about 500mb, but
usually
On 07/06/11 10:08, dobbin wrote:
Hello there, we've got one particular server backing up to a slightly
older debian server running bacula 2.44.
There's way too much stuff on the server to backup but the only
important files are ones with *foo* in the filename
I want to backup every file in D
Hello there, we've got one particular server backing up to a slightly older
debian server running bacula 2.44.
There's way too much stuff on the server to backup but the only important files
are ones with *foo* in the filename
I want to backup every file in D:/media and all of it's
So bacula must be looking at this and deciding that it's already backed it up
because the modified attribute is older than the previous backup.
But it doesn't actually check or compare against what it's already backed up.
Is there any way around this?
Hi,
you could use accurate backups:
That looks like exactly the thing I'm looking for, I'm not having much luck
getting it working though. I'm testing on our bacula 5 server with a bacula
5.02 client.
Ignore that, I'm being thick. I should have uninstalled the old bacula client
before installing 5.02
I've had previous problems with bacula running backups and simply deciding not
to backup certain files.
http://www.backupcentral.com/phpBB2/two-way-mirrors-of-external-mailing-lists-3/bacula-25/bacula-has-only-backed-up-part-of-a-directory-110828/
I never fully got to the bottom ot that. Now
Right, I used a little tool to update the file modified attribute to the
current time and then ran the backup. Now it's backing up the massive file.
So bacula must be looking at this and deciding that it's already backed it up
because the modified attribute is older than the previous backup.
backup server is Debian lenny running bacula 5.02, client is debian sarge
running bacula 1.38.11.
The server is a web server and on the 22nd a copy was made of the main site,
if you look at the backup job for that day, you can see an extra 4000
files. We're running daily incrementals with a full
Hi there,
this is probably bad form as I'm essentially asking the same question twice but
I'm in a bit of a bind so here goes. Last week I posted this issue regarding
the fact that it looks like bacula just randomly decided not to backup a bunch
of files.
Hi, I considered that but volume retention is set to 65 days and the last full
backup was on the 1st
probably you have a problem with retention periods.
When you select 5 option on restore command you should able to see the entire
chain of backup (FDIII...)?
Maybe you full backup has already
I finally figured out bcscan and shunted the volumes onto another bacula
server, hoping it was just a database issue.
I rebuilt the catalog from what's in the volumes and the files are still not
there.
So it looks like bacula just decided not to backup these files. could it be
something to do
Hi there,
backup server is Debian lenny running bacula 5.02, client is debian sarge
running bacula 1.38.11.
The server is a web server and on the 22nd a copy was made of the main site, if
you look at the backup job for that day, you can see an extra 4000 files. We're
running daily incrementals
Hi there, I'm running bacula-dir in two locations, each backing up about 100
servers onto disk arrays, I'm not using tape.
One location had a bit of calamity a while back and I was forced to restart all
backups from scratch, that location is doing fine.
In the other location, backups repeatedly
Quote:
All the machine backups go into one pool. It is counterproductive to use
individual pools for machines as that means tapes aren't filled
regularly and you're having to either keep lots in the changer or have
to change tapes constantly.
yes but, as I said in my first post,
All the machine backups go into one pool. It is counterproductive to use
individual pools for machines as that means tapes aren't filled
regularly and you're having to either keep lots in the changer or have
to change tapes constantly.
yes but, as I said in my first post, I'm not using
There is no benefit on the creation of 100 pools for 100 different jobs
/ servers.
I don't understand what you mean, each server is backed up seperately, I have
100 servers, therefore I run 100 backup jobs and therefore have 100 pools
unless I'm misunderstanding the meaning of pool
when I say 100 backups, I mean 100 different servers with different pools.
Wait a minute this has to fail. I mean you said you do 100 backups in
12 hours but then your configuration has use the volume once (and hold
each volume for 65 days after using) so that at maximum 65 jobs can be
This may sound like a stupid question, but everything I've found relating to
this error referrs to problems with tape changers and our system is entirely
disk based. We've got two debian servers running bacula 2.44 and 3.02 (yes,
they're old, the 2.44 one is getting upgraded very soon)
19 matches
Mail list logo
