kely to fail if one or more jobs is running at
the time.
Has anyone else encountered this or similar behavior?
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ro
Error: Bacula backupsrv-dir 5.2.3
> (16Dec11):
>
> ...and wonder why does DIR get a "Network error" after 33 minutes.
33 minutes and 20 seconds, I'll bet. That's 2000 seconds.
So what has a 2000-second timeout?
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43
On 02/16/2012 05:01 AM, Thomas Mueller wrote:
> In theory encryption does just need the public-key. Encrypting needs the
> private key. but I don't know if it is possible to provide only the
> public key to bacula-sd.
I believe you mean that DEcrypting requires the private key
t; despite being able to immediately connect using those credentials
>
> mysqlshow -u bacula -h localhost --password="DbPass" bacula
> Database: bacula
Very curious. That certainly looks as though it SHOULD be working.
--
Phil Stracchino, CDK#2 DoD#
"DBPass"
> }
Notice anything...?
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, SQL wrangle
On 02/06/2012 05:45 PM, Phil Stracchino wrote:
> Stephen,
> Three suggestions here.
[...]
> Route 4:
...I'm sorry. We'll come in again.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinat
oute 3:
Look into an alternate DB backup solution like mydumper or Percona
XtraBackup.
Route 4:
Do you have the option of taking a snapshot of your MySQL datadir and
backing up the snapshot? This can be viable if you have a small DB and
fast copy-on-write snapshots. (It's the technique I
that was written for MySQL 5.0
but did not work completely normally with MySQL 5.1.
You might want to check before you downgrade.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Un
t and is now in extended support only (which is to say, only
critical bugs are being fixed). It will probably be end-of-lifed this year.
If you're trying to do it via mysqldump and it's not working, it most
likely means you're doing something wrong.
--
Phil Stracchino, CDK#2
#x27;m not sure I could in good conscience recommend any of
the maatkit tools for production work. Too many of them too often don't
work or produce useless results.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.
required to achieve the same
ends with mysqldump. It stores schema and data in separate files, and
supports both file and DB-connection compression.
In short, it's what mysqldump SHOULD have become by now, but hasn't.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71
ks handling these table sizes.
What my company does is run multiple Directors each with its own catalog
DB. The catalog DBs share the same DB server though (running PostgreSQL).
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.
y
frequently-repeated queries, update to MySQL 5.5, and seriously,
seriously consider converting to InnoDB. It is a MUCH higher
performance storage engine than MyISAM. Remember that MyISAM was
designed to yield *acceptable* performance in shared installations on
machines with less than 32MB
nity next arises. This
is particularly the case if jobs are running when the reload is executed.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker,
JOIN Pool AS d
> ON c.PoolId=d.PoolId AND d.Name='ARCH')
> AND a.EndTime>(now()-INTERVAL '10 DAY') ORDER BY a.StartTime ASC;
Drop the quotes around 10 DAY.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net
ck?
Are you getting useful information from it? If so, why unsubscribe?
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, SQL wrangler, Free Stater
n never upgraded and
never tuned. You don't want to know how many production websites I've
seen trying to run off MySQL instances with "small to medium server"
configuration files that were written back when a "large server" was one
that had more than 32MB of RAM.
--
all the volumes with the correct VolRetention?
Look for "Update pool from resource", then "Update volumes from pool".
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance M
the Linux forcedeth driver for the nVidia MCP77 GigE port on
my workstation, support jumbo frames.
Needless to say, I was annoyed.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Un
cula/ | head -4
total 3.1T
-rw-r- 1 root root 916G Dec 5 10:31 FULL-20111205-04:30
-rw-r- 1 root root 902G Nov 7 10:42 FULL-2007-04:30
-rw-r- 1 root root 901G Nov 5 05:47 FULL-2004-15:18
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerl
On 12/09/11 15:18, Tilman Schmidt wrote:
> Am 06.12.2011 10:07, schrieb Phil Stracchino:
>> On 12/06/11 03:10, Marcello Romani wrote:
>>> The closest thing to a custom "eject tape" builtin command in bconsole I
>>> could came up with is this:
> [..
estore the Dir.
Additionally, you may be in the position of "As long as we're rebuilding
the box anyway, let's make a few hardware changes, we'd get better
performance if we [fitb...]"
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerl
t;echo 'umount storage=Tape' | bconsole"
> RunAfterJob = "mt-st -f /dev/nst0 offl"
> }
>
> This works on my setup, where I have a single tape drive (LTO-1, for the
> record). No autochanger.
You know, personally I just set "OfflineOnUnmount = ye
it seems to me, is that this would break
compatibility with all older Bacula volumes and installations.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, S
writing to the 12TB ZFS disk array on babylon4-sd.
Can this be done? Or is spool enabling strictly job-by-job?
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker
up
> Recycle = yes
> AutoPrune = yes
> Volume Retention = 31 days
> Volume Use Duration = 72h
> Storage = NEO200S
> Cleaning Prefix = "CLN"
> }
It's working for me. However, keep in mind that the volume status may
not update until Bacula next examines th
rate. Sure, only one twentieth of
their indexes would fit in the cache ... but *those were the 1/20th
they were actively USING*.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Un
your job definitions. But we
can't tell you what without seeing your definitions.
If you would post the relevant sections of your configuration files, we
can tell you what you've done wrong.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net al
bscan - Bacula's 'Scan tape'
...
-t port
Specify database port (default: 0)
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix roni
he explanations in the howto guide. The bare database
> has been re-created and I can get to the "bconsole" but of course
> commands like "update" or "list" will come back empty.
Win,
What were you doing for MySQL catalog backup?
--
Phil Stra
$l = <$OUT>;
$l = <$OUT>;
print "Done.\nDeleting volume $file from disk ... " if
($opts{verbose});
if (-f $file)
{
$count++;
unlink ($file);
}
print "Done.\n" if ($opts{verbose});
}
On 10/30/11 12:59, John Drescher wrote:
> On Thu, Oct 27, 2011 at 4:07 PM, Phil Stracchino wrote:
>> I just compared what the script was doing to the existing schema and
>> made the necessary changes by hand. Wasn't a problem, and nothing
>> non-obvious showed up.
>
backup the
> resulting files.
It's certainly by far the simplest for small-to-moderate-sized DBs. The
main problem with it is that for sufficiently large databases, it fails
because mysqldump just physically takes too long to complete the dump.
--
Phil Stracchino, CDK#2
vement in available uptime is huge.
(To be fair, if using all-InnoDB, you can use mysqldump
--single-transaction to avoid locking any tables and still get a
consistent dump. However, that doesn't speed mysqldump up any...)
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71
n features to recover from
the error. MyISAM does not. MyISAM is, in fact, very vulnerable to
data corruption from any of a number of causes.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissa
s fast in Percona 5.1 as
InnoDB in MySQL 5.5. However, I rather believe they're somewhat
comparing apples to oranges.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin
On 10/24/11 11:05, Alan Brown wrote:
> Phil Stracchino wrote:
>
>> The NAS will be VERY much faster than the tape drive.
>
> Are you sure about that? My LTO5s make 160Mb/s+ and I've yet to see a
> non-SSD NAS which can match them.
Actually, someone else asked the
after it
completes? This will also free up your client faster.
Caveat: The NAS and the tape drive MUST be controlled by the same
storage daemon for this to work.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.
it innodb_buffer_pool_partitions.) It is one of the most
important configuration settings for obtaining optimum InnoDB
performance in MySQL 5.5.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
-the-box MySQL configs entirely. They are
worthless. They were written back when a "large" machine was one with
more than 32MB of RAM. If you want performance out of MySQL, learn to
tune and configure it properly yourself.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM
On 10/05/11 15:54, Ben Walton wrote:
> What mysql are you building against? The included one? More to the
> point, which compiler was it built with?
MySQL 5.5.8, Solaris 10 package from Oracle. Need to update that actually.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 4
bly be
a lot simpler if Solaris 10 came in all-32-bit and all-64-bit flavors,
instead of the hybrid mixture of 32-bit and 64-bit it is at present.
I'd hoped it would be easier using Sun Studio, but if anything it was
harder.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71
mod the MySQL 64-bit lib
directory 000. I'm reasonably certain this wasn't Bacula's fault.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, SQL wran
insertion by default.
>
> I'm still looking for mysql optimizations, or do you think it best to
> leave my.cnf at the default values (we don't do many restores) and use
> the maximum amount of RAM for the fs buffer cache as needed?
I'd look at performance with somethi
nt IDs.
Unfortunately, all the "enterprise" Linux distributions will probably
still be shipping MySQL 5.1 five years after MySQL 6.0 ships...
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
R
ot. Granted it would be nice if it overwrote
canceled jobs, but that's not always easy, because jobs may be
interleaved on the tape, so it doesn't try.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ord
t
once, then they will have to complete in roughly 12 minutes each if you
want your scheduled 02:00 job to be able to start on time.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man
at
you are describing. You should see an xfslogd process for each XFS
filesystem, and many of those other processes you have listed should
have one process per CPU core.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net
al system RAM should be allocated to MySQL, with 75% of
total system RAM in the InnoDB buffer pool.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Pe
.
>>
>
> Bacula does not support using more than 1 director. The developers
> will tell you this.
Which, I'm inferring, is why it took the OP a while to realize there was
a second Director running...
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
h time
Bacula automatically purges them and moves them into my scratch pool,
from where an admin job deletes the purged volumes once a week.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
each retention period.
> You can move empty buckets (volumes) between garages if you need more
> in one or the other. You may also set up a new garage when bigger
> buckets (larger tape capacity) become available and they require
> different handling.
I just want to say that this analog
ting ..."failed: database is full" error messages.
> Job completes, but fails at the end with this sequence of messages.
> Using sqlite.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p.
stem. There is no reasonable way to make it do
what you want to do. If you leave the machine around with an open
console and no password, you have no expectation whatsoever of security.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metr
e a 4 GHz processor.
mt can both get and set options including compression. It's all in the
man page.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix r
et up my message commands to use mutt as the
mailer. Example:
mailcommand = "/usr/bin/mutt -F /root/.muttrc-bacula -s \"Storage
alert\" %r"
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@
;s
requests address, as listed in the message headers:
List-Unsubscribe:
<https://lists.sourceforge.net/lists/listinfo/bacula-users>,
<mailto:bacula-users-requ...@lists.sourceforge.net?subject=unsubscribe>
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.3
fat enough pipe in between, you could have a redundant offsite
SAN and run copy jobs to it.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, SQL wrangler,
against previous jobs. No new incrementals
that are *not* started *after* the Full finishes will be based off the
new Full.
If this is what you want, then all you should need to do is turn off
'cancel duplicate jobs' and make sure that all applicable job
concurrency settings
On 07/06/11 10:41, Adrian Reyer wrote:
> On Wed, Jul 06, 2011 at 10:09:56AM -0400, Phil Stracchino wrote:
>> should I use for my tables?" is MyISAM.[1] At this point, wherever
>> possible, EVERYONE should be using InnoDB.
>
> I will, if the current backup ever finishes
ing your Fileset from a simple
script (which probably needn't be much more than a 'find / -iname \*foo\*').
Oh, and you should probably use the 'ignore fileset changes' option.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.n
is and enabling the
grant tables to also be stored in InnoDB is work in progress, and that
once this is accomplished, the entire MyISAM storage engine will
probably be deprecated.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net
rence.
Rule of thumb: Assume all out-of-the-box DB engine configurations, be
it MySQL, PostgreSQL or whatever, are worthless.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ron
of which is run
> in its own transaction.
> Someone else might be able to help more.
It always take longer - usually several times longer - to load any given
SQL symbolic dump back into any SQL DB than it took to dump it in the
first place. Just for starters, dumping the DB does not alter any
in
On 06/08/11 13:45, Jérôme Blion wrote:
> I had such issues with BAT... With Webacula, I am not able to reproduce
> this behaviour.
Very interesting.
> Perhaps a bad query which does not use an index.
Definitely worth study.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM
such as Bacula, you should still
expect to see a very significant performance improvement. As Jérôme
observed, if you're tied to 5.1, you should use the plugin InnoDB engine
rather than the built-in InnoDB engine if at all possible.
--
Phil Stracchino, CDK#2 DoD#29979
On 06/08/11 11:44, Jérôme Blion wrote:
> On Wed, 08 Jun 2011 10:50:58 -0400, Phil Stracchino
> wrote:
>> The very first thing I would do would be upgrade to MySQL 5.5.[current]
>> (5.5.13, right now) if you're not already using 5.5, making sure it's
>> properly
t;What
primary storage engine should I be using?" has any answer other than
"InnoDB". It's probably not too inaccurate to say that unless you
*NEED* either merge tables or full-text indices, you should be using InnoDB.
[1] Which is to say, not NDB Cluster
--
Phil Stracc
On 06/07/11 15:41, Rory Campbell-Lange wrote:
> On 07/06/11, Rory Campbell-Lange (r...@campbell-lange.net) wrote:
>> On 07/06/11, Phil Stracchino (ala...@metrocast.net) wrote:
>>> On 06/07/11 12:14, Rory Campbell-Lange wrote:
>>>> Does deleting the job remove the d
will need to zero some tapes.
Glad it came up, then. :)
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, SQL wrangler, Free Stater
It'
On 06/07/11 12:14, Rory Campbell-Lange wrote:
> On 07/06/11, Phil Stracchino (ala...@metrocast.net) wrote:
>> On 06/07/11 09:59, Rory Campbell-Lange wrote:
>>> I've written job 329 after job 315 on the same tape (CLW112L4). See below
>>>
>>> Two que
aximum Volume Jobs = 1" in the Pool.
You will have to update the Pool from the resource, then all the Volumes
from the Pool. Keep in mind, though, that this does mean you will only
ever be able to have one job running at once to that storage device.
--
Phil Stracchino, CDK#2 DoD#299792
use you
just do it in a single streaming read, but in practical terms you'll
never notice the difference.)
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix
aug | sep | oct | nov | dec | january |
february | ... | december
= 1 | 2 | ... 31
=
Likewise, the last day of the year is pretty much always Dec 31
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net
ven safer. There's always the chance
something mucked with a timestamp in an unexpected way.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, S
ing a
> diff to see what is different between the two locations?
There's another, simpler approach: Restore the entire damaged
filesystem, with REPLACE set to 'only if newer'.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@met
f a volume.
Now what you COULD do, which would achieve the same result, is to
copy/migrate only the good jobs on the volume to a new volume, then
purge the original volume.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net
you can just right-clock those volumes and select "List jobs on
volume". Then rerun those jobs as a differential (a Full is not
necessary). This will basically roll up all changes since the last
Full, thus resetting the "chain" of incrementals.
--
Phil Stracchin
ng the file and writing a null value into it : maybe
> something like echo "" > /Volumes/LabelOfTheVolume
> I want to keep the label in order to reuse the volume
In Bacula 5.0.3, you can use Action On Purge = truncate, iirc.
--
Phil Stracchino, CDK#2 DoD#2997
> 1 - Add D:\ into the FileSet
> 2 - Add C:\MyNewHardDrive into the FileSet
> 3 - Set option onefs=no in the filesystem
er ... the *fileset*, rather, not the filesystem
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocas
:\ into the FileSet
2 - Add C:\MyNewHardDrive into the FileSet
3 - Set option onefs=no in the filesystem
Pick one. Option 3 is not recommended, as it will cause all mountpoints
to be recursed into, which tends to be a bad thing on Windows since
Windows remounts bits of its own filesystem on mountpoin
then this should work on verify jobs too?
Of course, you need to make certain that the MySQL being stopped is not
the one running your catalog. ;)
Personally, I - both on my own systems, and on work systems - do not
back up MySQL data directories directly. I dump the DBs and back up the
dumps, o
, which
is backed up directly to LTO2 tape.) At this moment in time, the
smallest volume in spool/bacula (yes, it's a Solaris 10 box with a
12-disk ZFS array) is 5.8GB; the largest, 201GB. I have seen disk
volumes as large as 450GB in the past. I have not run into any problems
with this s
dup-enabled
> ZFS storage w/ Bacula.
I have not experimented with ZFS deduplication. ZFS deduplication is
implemented at block level, so the size of the file is unlikely to make
a great deal of difference.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caer
the Bacula DB, recreate it, restart
Bacula, bscan in the volume containing your last Catalog backup, restore
the Catalog.
However, if the device your Catalog is stored on is throwing block
errors, and you do this WITHOUT first replacing that device, you're
playing Russian roulette. Replace th
because version 12 of the DB is correct for Bacula 5.0.3.
If your Director wants DB version 10, it's a v2.something Director.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man,
>
> Thanks in advance.
>
> Rickifer Barros.
Open BAT. Click "Jobs Run" in the tree pane. Find the job you want.
Right-click it and select 'List JobMedia' from the context menu.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
a
e DVD-sized (or fractionally less than DVD-sized) disk
volumes, say 4.6GB for single-layer or 8.4GB for DL DVD blanks, then
write the volumes separately to DVD via an external process.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast
Director-per-client and
give each client their own storage area with a disk quota.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, SQL wrangler, Free Stater
le still exists on disk, and is still being appended
to, even though marked as deleted. Kill the process writing to it, so
that it gets closed, and it goes *poof*.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@
On 05/12/11 14:57, Phil Stracchino wrote:
> This is version 2.0 of my checkhost tool.
Except that somehow I accidentally attached the wrong version. Let's
try this again.
To make up for the error, here's a standalone lanwake tool using the
same DB backend as well.
--
Phil St
cript before
running it, and should set the broadcast address as well (default
broadcast address is set to 255.255.255.255).
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin,
no current
capability for direct communication between one storage daemon and another.)
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, SQL wrangler, Free S
On 05/07/11 10:17, Phil Stracchino wrote:
> On 05/07/11 07:22, Radosław Korzeniewski wrote:
>> Puting a whole instalation into /etc hierarchy is not a good idea. There
>> are /etc for config files, /bin and /sbin for binary app files and /var
>> for frequently changed and t
instead of
> /etc/bacula. It just my 0,02usd... :)
/opt/baclua/{etc,sbin,lib,} is actually the recommended official
Bacula installation layout.
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net ala...@metrocast.net p...@co.ordinate.org
Renaiss
On 05/06/11 03:59, J. Echter wrote:
> Hi,
>
> i have a fileset containing nfs mounted shares, but these seem to not get
> backed up.
Don't back up over NFS if you can avoid it. You'll get much better
performance backing up the NFS shares on the NFS server.
--
toward both of these is probably getting a good
understanding of how the base Windows 7 install is actually structured
and where all the junction points are. Call me cynical if you want, but
sometimes I have a hard time believing Microsoft does not intentionally
obfusca
ackup, then future incrementals and differentials should
run against that VirtualFull. The "incrementals forever" model is
pretty risky ... it can mean that a restore requires a HUGE number of
jobs restored, and can fail if any of them is missing.
--
Phil Stracchino, CDK#2 DoD#
he only
ways around this are either to educate your users, or to redesign your
backup strategy in a robust way (say, use rsync to copy the data to a
buffer area off the machine that you actually back up).
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.ne
isn't using compression so I really don't understand
> this.
Your *job* may not be set to use *software* compression, but is your
*tape library* automatically using *hardware* compression?
--
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
ala...@caerllewys.net
601 - 700 of 1450 matches
Mail list logo
