Hello list,
I would like to tell you about a problem we're having after upgrading
bacula 2.2.6 to 5.0.3.
In our setup we have the main bacula server (director and storage) with
a LTO3 tape drive. All jobs are done on this server using his local
storage (storage1). The is also another storage daem
Michael FIG wrote:
> Hi,
>
> Cedric Tefft writes:
>
>
>> An autochanger config is probably overkill. You can implement autofs as
>> John already suggested, or you can use the "Mount Command", "Unmount
>> Command" (et. al.) directives
that prevents Bacula from calling the
unmount command for file resources. It's fixed in the git repository,
but I don't know if the fix made it into 3.0.3.
- Cedric
--
Come build with us! The BlackBerry(R) Devel
Steve Polyack wrote:
> I've recently enabled spooling on a storage director by turning it on in
> the Director and Storage sections of the dir's config. I've also
> enabled it in the schedules where it should be. However, for a
> particular schedule there is one job which I do not want to spoo
Gabriel - IP Guys wrote:
>
> I’ve just completed the installation and setup of a Bacula based
> backup system, and I'm happy with everything so far. I have two legacy
> machines, one running FC6, and one running FC3. I wish to build RPM’s
> for these systems, but is there anyway I can do this wi
#x27;t that question more appropriate
for a MSSQL forum than a Bacula forum?
- Cedric
P.S. If you get that far, the "pipe" symbol is "|", not ">".
--
Come build with us! The BlackBer
Matt Richards wrote:
> Hello,
>
> It is possible to get the SD to connect out to the FD for a single FD?
>
> I have a single FD outside of the network that bacula runs on and this
> FD is behind NAT for the rest of the network.
>
> I could start natting connections into the network that are from th
Ronan Eckelberry wrote:
> Hi guys,
>
> Quick scheduling question. I want to run Fulls 1st of every 3rd
> month and Incrementals on 2nd-31st. I think the incremental part is
> easy, but does anyone know of a good way to specify the Fulls?
>
> Looking for something to do the following:
>
DAve wrote:
> No change, same error.
>
> 30-Oct 07:09 director-dir: Allied-ex3.2009-10-30_01.00.02 Warning: Error
> updating job record. sql_update.c:194 Update problem: affected_rows=0
> 30-Oct 07:09 director-dir: Allied-ex3.2009-10-30_01.00.02 Warning: Error
> getting job record for stats: sql_
in practice most implementations will start
recycling volumes long before they hit 10,000.
- Cedric
--
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this
Fredrik Tegenfeldt wrote:
> Hello,
>
> is there any plans for using multithreading in the compression as in e.g
> pbzip2 ?
>
Your question is probably more appropriate for the developers mailing
list, bacula-d
Evan Fraser wrote:
> Attached are the excerpts of the log from last night, and the list media
> command I ran this morning whilst the job is hung.
>
> As you can see, I have setup a weekly pool rotation, but this problem was
> also occurring when I only had default and scratch pools.
>
> I forgot
Pascal Clermont wrote:
> I have added a new machine to be backed up with bacula and I cannot get a
> full backup, always have these weird errors.
> the current setup is all centos 5.3 86_64 and every other host besides this
> one have been working flawlessly for over 6 months.
>
> DIR Version: Ve
On Mon, Oct 26, 2009 at 1:36 PM, Cedric Tefft wrote:
>
>
> - Cedric
>
> These things usually turn out to be in some place you haven't
>
>
Doh! Sorry, that was supposed to be "... some place you haven't *looked*."
I hit send before I was finished edit
ight make sense to provide more information. I suggest you post:
1) Your entire director config
2) The part of your log file from the start of the job through the part
where it starts requesting tapes.
3) The output of 'list volumes', preferably taken while a job is on hold
requesting
y.
To me, this implies lzma compression is somewhere on the horizon, though
you'd probably have to check with Kern himself to know for sure.
- Cedric
--
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is
have five separate
catalogs, and this may be where you're getting into trouble. You need
to make sure your dump scripts correlate exactly with your clients -- in
other words make absolutely certain you aren't dumping the database to
the bac
glynd wrote:
> Hi Cedric,
>
> IS this what you are looking for?
>
> r...@mistral:~# ls -la /var/lib/bacula/bin/bacula.sql
> -rw-rw-r-- 1 bacula root 0 2009-10-22 08:20 /var/lib/bacula/bin/bacula.sql
> r...@mistral:~#
>
The size of the dump file is zero bytes. That&#
On Thu, Oct 22, 2009 at 12:23 AM, glynd wrote:
>
> When the backup runs I keep getting this error:
>
> 22-Oct 09:09 glyn-laptop-fd JobId 161: Could not stat
> "/var/lib/bacula/bin/bacula.sql": ERR=The system cannot find the path
> specified.
>
> The file exists and is owned by bacula.root and
I'm not certain, but here's my guess: In the pool definition for the
Pull_Inc pool, you have Label Format = "IncVol-". Although the manual does
not explicitly say so, I think this is equivalent to "IncVol-${NumVols+}"
which means the number Bacula automatically appends to the Label Format
string
2009/10/21 Serge Vostroknutov
>
> Hello
>
> *status dir
> bacula-dir Version: 2.0.3 (06 March 2007) x86_64-redhat-linux-gnu redhat
> Enterprise release
>
> I have a following warning messages in the bacula.log
> Those messages were appeared after some jobs and lolumes were manually
> purged from
2. Obviously, if that really is the problem it
wouldn't fix your restore issue right away -- you'd have to do a fresh
backup and try to restore from that.
- Cedric
--
Come build with us! The BlackBerry(R) Dev
the output of configure for
a reason? It may reveal the real reason make is failing.
Also, these may be of some use
http://www.pixelchaos.net/2009/07/14/bacula-3-0-1-for-mac-os-x/
http://www.pixelchaos.net/2008/11/25/bacula
Jesper Krogh wrote:
> Cedric wrote:
>
>> Do a Differential once every 90 days and Incrementals the rest of the time.
>> In theory, you only need to run one Full backup. Ever. In practice, you'll
>> probably want to create a new Full backup every year or so to clear
ly want to create a new Full backup every year or so to clear out the
obsolete data and make your differentials smaller (and faster).
Another way to do it that uses even less disk space would be to eliminate
the differentials completely and just create a new Full backup every 90
days.
- Cedric
-
gt;
> So according to this Volume strategy management, Bacula will NEVER
> create an eleventh Volume (Vol-0011) because would keep recycling
> Vol-0002, Vol-0003, ..., Vol-00010 (and then Vol-0001 again). Am I
> right?
>
Yeah, that looks about right.
- Cedric
reasons which otherwise make no sense. This seems to be exactly what
you're experiencing.
So short answer: ignore the error message. Your 2.4.4 daemon is not
fully compatible with 1.36.6 FD.
- Cedric
--
Come
anuals/en/install/install/Configuring_Director.html#SECTION006150000
- Cedric
--
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year.
om the FD?
>
>
My solution to essentially the same problem was to run two copies of the
FD: one with encryption and one without. You pull from the FD without
encryption when writing to disk, and the one with encryption when
ntial backup
> strategy?what cause it?
>
>
>
When you say the "time stamp is correct" you're probably only looking at
the modification time (mtime) of the file. By default, Bacula uses both
the mtime AND the ctime (Change Time) of a file to determ
-01 09:08:56 |
| Default | TAPE11 | 18494300160 | 2009-10-01 13:00:32 |
+-++-+-+
8 rows in set (0.00 sec)
With a little SQL knowledge you should be able to rework this into
whatever form you need.
- Cedric
Cedric Tefft wrote:
> Markus Falb wrote:
>> Eric Böse-Wolf wrote:
>>
>>
>>> Vladimir Doisan writes:
>>>
>>>
>>>> If you turn TLS and file encryption - the data will be double
>>>> encrypted
>>>>
ur backup
scheme you consider to be insecure. If you're worried about someone
getting hold of your backup media, you need "data encryption". If
you're worried about someone eavesdropping on communications between the
FD and SD, you need "TLS encryption". And obvious
r
that you can take advantage of. The only way to know for sure is to try it.
- Cedric
--
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumps
g this to no MIGHT solve
your problem. OTOH, it might actually make things worse. I suggest you
look up the Spool Attributes directive in the manual.
- Cedric
--
Come build with us! The BlackBerry(R) Developer Conferen
t; on one of those older files. Look at the ctime. Is it very
recent -- perhaps a day or less? By default, bacula uses both the mtime
and ctime of a file to determine if it has been changed. If EITHER one
of those is newer than the last backup, bacula will include the file in
an incrementa
don't know. If you are using either/both, you might try turning them
off just to see if it makes any difference.
- Cedric
--
Come build with us! The BlackBerry® Developer Conference in SF, CA
is the only develop
Proskurin Kirill wrote:
> Hello all.
>
> Im confused about recycle options in Bacula 2.4.4.
> I have such options on some pools:
>
> Recycle = yes
> AutoPrune = yes
> Volume Retention = 30 days
> Maximum Volumes = 3
> Maximum Volume Bytes = 10G
>
> As I understand - volume what older than 30 day wi
Reynier Pérez Mira wrote:
> Hi:
>
> Cedric Tefft wrote:
>> Reynier Pérez Mira wrote:
>> Probably /tmp. Run "show variables where Variable_name = 'tmpdir';"
>> in mysql to find out for sure.
>
> I use PostgreSQL instead of MySQL. What this qu
quot; in
mysql to find out for sure.
Be aware that if the problem really is a lack of temp space, you would
probably only see /tmp at or near capacity WHILE bacula is actively
trying to update the database (i.e. not before or after).
- Cedric
-
Federico Torres wrote:
> Good morning:
> I'm having a problem for quite some time with Bacula. I'm backing up about
> 30 servers (> 1Tb of info), and despite it all works, this problem is
> breaking my head a couple of times a month.
>
> I'm planning a scheme of jobs as follows: 1 Full on the fir
box. Sit back and let bacula do
its thing.
If you didn't already know that, you need to read the Concepts and
Overview section of the manual.
http://www.bacula.org/3.0.x-manuals/en/concepts/concepts/What_is_Bacula.html#SECTION0022
If you DID already know that, you need
Daniel Durgin wrote:
> I have searched google but haven't found a tape rotation calculator.
>
> Does anyone know of a way to use bconsole or the mysql database to
> calculate number of tapes needed. I was hoping for something dynamic
> so I could forecast different Tape Schedules.
>
No, there is
oing a SQL query on the database after a failed
job, something like "select * from Job where JobID =
;". If the error message is accurate, this query
should return no rows.
- Cedric
--
Come build with us!
ob (or JobDefs)
resource, then it will default to that level. If you DON'T specify the
level in Job or JobDefs, then it will use the implicit default which is
Incremental.
In the console, you have the option to override the level e
Joseph L. Casale wrote:
> When I run it the first time, it does a FULL, when I add data, and
> run it again, it does am Incremental?
>
> What am I missing? I was expecting a Differential?
>
>
When you ran the job the first and second time, do you mean you
initiated the job manually (i.e. with t
c problem as has been assumed. If I
were in your position, I would add a "me too" to the bug report,
including the relevant diagnostics and (obviously) pointing out that
unlike the reporter, your disk is not full. Perhaps your
cle
volumes and SD reads FD data as a command."
http://bugs.bacula.org/view.php?id=1356
- Cedric
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design,
f the manual that details the Job
resource, specifically the RunScript directive:
http://bacula.org/3.0.x-manuals/en/install/install/Configuring_Director.html#SECTION0063
- Cedric
--
Let Crystal Reports han
Joseph L. Casale wrote:
> The manual states
> variable expansion is for Volume Labels only
Really? Where does it say that, exactly? I looked at what I believe in
the relevant section of the manual, and I came away with a completely
different understanding.
-
hich can be completed within your 10-hour
window.
- Cedric
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
mp related: Stat shows
you the mtime, ctime, and atime (among other things):
> stat myfile
File: `myfile'
Size: 11603298Blocks: 22664 IO Block: 4096 regular file
Device: fe03h/65027dInode: 810095 Links: 1
Access: (0644/-rw-r--r--) Uid: ( 500/ cedric) Gid: ( 50
#x27;s. I would suggest verifying the
client and director's clocks more or less agree. This may be a
non-obvious issue if the client and director are in different time zones
-- or, more to the point, if one of them is in the wrong timezone.
- Cedric
---
pdated!
> need pass a username and password to the script somehow?
>
> Please help
>
>
Any arguments you pass to the update script are passed blindly on to
mysql, so if you use the -u and -p options to log in to mysq
edule {
Name = "MyCycle"
Run = Level=Full Pool=YearlyPool Jan 1st sun at 6:00
Run = Level=Full Pool=MonthlyPool feb-dec 1st sun at 6:00
Run = Level=Full Pool=WeeklyPool 2nd-5th sun at 6:00
Run = Level=Full Pool=DailyPool mon-sat at 6:00
}
- Cedric
Nicolas Jungers wrote:
> Le jeudi 03 septembre 2009 à 11:24 -0700, Cedric Tefft a écrit :
>
>> You might be able to accomplish this with a short script that identifies
>> the large files, run it before every backup, and use its output as an
>> include list for you
You might be able to accomplish this with a short script that identifies
the large files, run it before every backup, and use its output as an
include list for your large-files job and an exclude list for your
small-files job.
Here is an example for which I've assumed your users are backing up
Good idea.
Gavin McCullagh wrote:
> Hi,
>
> I'm not sure if this suggestion is well-known already, implemented already
> or maybe just a plain stupid idea that wouldn't work. So, I'm going to
> suggest it and don my flame-retardant suit. Feel free to flame or shoot
> it down.
>
> Suppose I do m
Graham Keeling wrote:
> Hello,
>
> On http://bugs.bacula.org/view.php?id=1356 Kern told me this...
>
> "...you are using an undocumented feature concerning the Storage definitions,
> and in any case, the way you are using it is definitely not supported and at
> some time will probably cause severe
terryc wrote:
> A job calls for an appendable tape
> A volume listing says the tape is appendable.
> The tape is mounted in the drive.
> It acknowledges that the tape is mounted and correctly reports the
> label, BUT director keeps squarking for hours for an appendable tape.
>
> Where this happens
tqz wrote:
> Hmm...I may be getting confused with another response I received from another
> forum...
>
> I have my keys stored off site in a secured location, as well as the bacula
> config files. So in the event of a disaster, if I reinstall bacula and my
> database management system, restore
ay be more helpful.
Also, just for grins, I would try temporarily disabling the Windows
box's other network card -- whichever one bacula is not connecting to.
This shouldn't make any difference, but if nothing changes it
d if I do something stupid like typing "rm -rf" in
the wrong directory. :-)
- Cedric
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and depl
so you need to make sure only trusted individuals have access to
both. For added security, I'd suggest you wrap your off-site encryption
keys in another layer of encryption. Using PGP (or GPG) to
conventionally encrypt the key files before writing them to the USB/
"purge"
command), and if THAT fails, use the "update" command to directly change
the tapes' status.
- Cedric
Edward Aronyk wrote:
> Hey All,
>
> I've been using Bacula to backup 7 or 8 servers for a while now and
> have been fairly pleased. I backup to a
OK, I posted new bug, # 1361 ( http://bugs.bacula.org/view.php?id=1361 )
I encourage anybody who is experiencing the same issue to have a look at
the bug report and add anything relevant.
Thx,
- Cedric
Thomas wrote:
> one year ago i ran into the same problem:
>
> http://www.adsm.
27;t even do it on an explicit
"umount" issued from the console.
Like you, I looked at bug 830 ( http://bugs.bacula.org/view.php?id=830 )
which was closed as "fixed" on 2007-07-16 with the comment:: "Fixed in
current SVN. You must have the appropriate Mount/Unmount commands
Hi,
Is there a way to initiate the backup from client side?
I have a special case where director is outside a lan... and it can't
connect to client directly... but client can connect to director...
Is it possible something like that? (without to add vpn solution).
--
Cedric G
Alan Brown wrote:
> On Thu, 20 Nov 2008, Cedric Devillers wrote:
>
>> too) you can use /dev/tape/by-id/scsi-XX-nst that should be fixed by
>> using specific id (instead of XX).
>
> Looking at that directory it's only been created for the first tape drive.
Alan Brown wrote:
> On Wed, 19 Nov 2008, Robert LeBlanc wrote:
>
>>> You can write a udev rule to lock the drives down. I wrote one awhile
>>> back to keep the changer device the same.
>>>
>>> dev6 ~ # cat /etc/udev/rules.d/55-bacula.rules
>>> SUBSYSTEM=="scsi",ATTRS{vendor}=="EXABYTE*",ATTRS{type
ll, even in trace file) may
be considered as a bug, but i've only seen this on a old 1.38 release.
Have a good day.
Cedric Devillers wrote:
> Cedric Devillers wrote:
>> Arno Lehmann wrote:
>>> Hi,
>>>
>>> 13.11.2007 12:54,, Cedric Devillers wrote
Cedric Devillers wrote:
> Arno Lehmann wrote:
>> Hi,
>>
>> 13.11.2007 12:54,, Cedric Devillers wrote::
>>> Hello,
>>>
>>> I have a little problem with one of our bacula installation.
>>>
>>> Let me explain the setup first.
>>
Arno Lehmann wrote:
> Hi,
>
> 13.11.2007 12:54,, Cedric Devillers wrote::
>> Hello,
>>
>> I have a little problem with one of our bacula installation.
>>
>> Let me explain the setup first.
>>
>> There is two server, the first has the data
Hello,
I have a little problem with one of our bacula installation.
Let me explain the setup first.
There is two server, the first has the data and the storage daemon
(meia). The second is the director/DB server (lucita). There is alsa two
client only servers (hr-accentv2 a windows client and da
Martin Simmons wrote:
>> On Wed, 19 Sep 2007 18:58:13 +0200, Marc Cousin said:
>> On Wednesday 19 September 2007 16:59:10 Martin Simmons wrote:
On Wed, 19 Sep 2007 11:54:37 +0200, Cousin Marc said:
I think the problem is linked to the fact dbcheck works more or less row
by ro
Martin Simmons wrote:
>> On Wed, 19 Sep 2007 11:54:37 +0200, Cousin Marc said:
>> I think the problem is linked to the fact dbcheck works more or less row by
>> row.
>>
>> If I understand correctly, the problem is that you have duplicates in the
>> path
>> table as the error comes from
>> S
Hello,
I must run dbcheck on my bacula database (postgresql backend) because of
"sql_create.c:767 More than one Path!" warning on every file backed up
(don't know where it comes from, actually).
The operation is extremely slow, it ran for the all week-end and only
deleted 2000 from the 51000 dup
Michel Meyers wrote:
> Aitor wrote:
>> Hi,
>> In Spanish: Español, Catalán, Inglés
>> In Catalan: Espanyol, Català, Anglés.
>
> In French: espagnol, catalan, anglais
Maybe you should add for French : espagnol (castillan), catalan, anglais ?
-
David Boyes wrote:
>> Of all the projects on the projects list, which 2 or 3 do you think
> are
>> most
>> important from an enterprise standpoint?
>
> That's a very open-ended question...8-) Careful what you wish for.
>
> IMHO, here's what my wish list would be:
>
> Copypools
> Extract capabi
Did you actually try mounting volume 84? If so, where's the
corresponding piece of the log?
- Cedric
---
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your j
Christoff Buch wrote:
Dear users,
as the subject says, bacula disables the python interpreter if I try
to import the time - module (which I need).
Has anyone any idea or experience with this?
Thanks a lot!
I can confirm the problem -- unfortunately no clue on the solution.
- Cedric
Mark Nienberg wrote:
Cedric Tefft wrote:
Mark Nienberg wrote:
I understand that I can't change a fileset definition without
triggering a full backup
Actually, you can. Just add the line:
Ignore FileSet Changes = yes
to your file set definition.
Does this mean that the fi
Mark Nienberg wrote:
I understand that I can't change a fileset definition without
triggering a full backup
Actually, you can. Just add the line:
Ignore FileSet Changes = yes
to your file set definition.
- Cedric
---
Using T
answer which I'm far too lazy to go into here, but Google
should be able to take you the rest of the way, and if you're a C
programmer, the bacula source code itself (in particular
src/lib/base64.c) should be helpful.
- Cedric
Robin Bowes wrote:
How can I guarantee that the verify job will run immediately after the
backup job?
Schedule both jobs to start at exactly the same time and set the
priority of the verify job to be one higher than the priority of the
backup job.
- Cedric
Chuck Bunn wrote:
Hi Cedric,
Thanks for responding. I looked at my logs and it looks like there are
a lot of issues. The README says nothing about LDAP or rights and
ownership on directories under bacula-web. Is there a document
somewhere that can guide me to do what is necessary to get
7;No'). Is all of this
normal or do I have a problem somewhere. I would think the page would
come up with at least the descriptions even though there is no data
yet to display? I have no jobs setup yet...
Thanks
Check your web server's error log. Mine's /var/lo
Juan Luis Frances wrote:
Hello Cedric,
If you use the last public version you musn't to have this problem:
http://prdownloads.sourceforge.net/bacula/bacula-gui-1.38.5.tar.gz?download
I have PHP 5.1.2 and bacula-web runs ok.
Best regards,
Juan Luis Francés
Much better.
T
sses.inc: $spr2 =
array_merge($spr2,array($row["Name"]=>$row["SUM(JobBytes)"]));
Pieter
Thanks, that did the trick, but yuck! Anybody got a reason why this
shouldn't be submitted as a bug?
Thx,
- Cedric
---
examples/test_setup.php
I'm not a PHP programmer and I don't play one on TV, but this looks like
a bug or a misleading error message. Short of debugging phplot, I'm not
quite sure where to go from here. Any suggestions?
Thx,
- Cedric
-
einitialize the bacula database:
http://www.bacula.org/rel-manual/Installing_Configurin_MySQL.html#SECTION000483000
- Cedric
---
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly
ou can end up with a
catalog that is unwieldy, if not outright unmanageable. Splitting the
data up into multiple catalogs is one way to address this problem.
- Cedric
---
Using Tomcat but need to do more? Need to support web services, security?
Ge
way" are you talking about the backup or
the daemon? In other words, when you try to start the service, can you
verify that it actually appears in the Services MMC as "started"?
- Cedric
---
Using Tomcat but need to do more? Nee
Cedric Tefft wrote:
Robert Nelson wrote:
You can also try "select Name from Filename where CAST(Name AS CHAR) =
'makedev';".
On my machine it was somewhat faster than CONVERT.
Out of curiosity, I plugged the CAST() function in to query.sql and
ran a real bacula
ual:
http://www.bacula.org/rel-manual/Getting_Started_with_Bacula.html#SECTION00092000
- Cedric
---
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to mak
#x27;. I did this three times and then did
the CONVERT() version three times.
Averages:
CAST: 62 seconds
CONVERT: 58 seconds
Obviously the CAST() version is preferable though, in practice, I don't
know if you'd be able to tell the dif
| Name|
+-+
| MAKEDEV |
+-+
1 row in set (1.85 sec)
I am using mysql 5.0.18, though, like you, I don't know why that should
make a difference.
- Cedric
---
This SF.Net email is sponsored by xPML, a groundbreakin
n't want to speak for the guy, and it
can't hurt to try, so by all means, give it a shot.
- Cedric
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile medi
27;re curious.
- Cedric
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new c
lPool, and the DifferentialPool
specifications permit overriding the Pool specification according to
what backup Job Level is in effect."
- Cedric
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that exten
1 - 100 of 111 matches
Mail list logo
