[Bacula-users] Volume issues with 11.0.6 running in docker
I recently migrated my bacula director and storage daemon to version 11 and am running them in docker. Ever since, I have started getting occasional volume errors that can only be resolved by restarting the bacula-sd container or in some cases deleting and relabeling a volume. Thinks will run fine for a week or two (or more) and then start throwing errors. Has anyone else seen anything like this? These volumes are file volumes that live on my Synology and are mounted over NFS. Here’s a example of what I’m seeing: 07-Aug 06:10 morpheus-dir JobId 3671: shell command: run BeforeJob "/opt/bacula/etc/scripts/make_catalog_backup.pl MyCatalog" 07-Aug 06:11 morpheus-dir JobId 3671: Start Backup JobId 3671, Job=BackupCatalog.2023-08-07_06.10.00_18 07-Aug 06:11 morpheus-dir JobId 3671: Using Device "vDrive-03" to write. 07-Aug 06:11 morpheus-sd JobId 3671: Fatal error: block_util.c:478 Volume data error at 0:0! Wanted ID: "BB02", got "". Buffer discarded. 07-Aug 06:11 morpheus-fd JobId 3671: Fatal error: job.c:3012 Bad response from SD to Append Data command. Wanted 3000 OK data , got len=257 msg="3903 Error append data: Read label block failed: requested Volume "morpheus-0076" on File device "vDrive-03" (/srv/backup) is not a Bacula labeled Volume, because: ERR=block_util.c:478 Volume data error at 0:0! Wanted ID: "BB02", got "". Buffer discarded." 07-Aug 06:11 morpheus-dir JobId 3671: Error: Bacula morpheus-dir 11.0.6 (10Mar22): Build OS: x86_64-redhat-linux-gnu-bacula redhat (Core) JobId: 3671 Job: BackupCatalog.2023-08-07_06.10.00_18 Backup Level: Full Client: "morpheus-fd" 9.6.7 (10Dec20) x86_64-pc-linux-gnu,ubuntu,20.04 FileSet: "Catalog" 2022-08-07 06:10:00 Pool: "morpheus" (From Job resource) Catalog: "MyCatalog" (From Client resource) Storage: "File" (From Job resource) Scheduled time: 07-Aug-2023 06:10:00 Start time: 07-Aug-2023 06:11:39 End time: 07-Aug-2023 06:11:40 Elapsed time: 1 sec Priority: 10 FD Files Written: 0 SD Files Written: 0 FD Bytes Written: 0 (0 B) SD Bytes Written: 0 (0 B) Rate: 0.0 KB/s Software Compression: None Comm Line Compression: None Snapshot/VSS: no Encryption: no Accurate: no Volume name(s): Volume Session Id: 72 Volume Session Time: 1690744006 Last Volume Bytes: 1 (1 B) Non-fatal FD errors: 1 SD Errors: 1 FD termination status: Error SD termination status: Error Termination: *** Backup Error *** Dan Rich | http://www.employees.org/~drich/ | "Step up to red alert!" "Are you sure, sir? | It means changing the bulb in the sign..." | - Red Dwarf (BBC) signature.asc Description: Message signed with OpenPGP using AMPGpg ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Frequently Changing Pools For Media
Hi Bill,
Sorry for the delay in responding, thank you for your response. I was
out of work for a couple of days and wasn't in a position to respond.
On Mon, Jul 17, 2023 at 01:50:45PM +, Bill Arlofski via Bacula-users wrote:
> On 7/15/23 06:13, Rich Fox wrote:
> > Hi All,
> >
> > I have a couple of different tape changers that are used to run jobs in
> > parallel and all share the same media. I've discovered by accident that
> > if I put all of the media into a single pool, the media selection
> > algorithm won't take into account which changer a tape is loaded in and
> > will try to mount tapes from a different changer which causes problems.
> > A simple solution is to have one pool for each changer and update the
> > pool for a given media depending on which changer it gets loaded into.
> > This appears to work in practice but I was wondering if anybody knows
> > of any unintended issues that might arise from doing this on a routine
> > basis.
> >
> > Thanks for any advice,
> > Rich.
>
>
> Hello Rich,
>
> Without seeing any configuration, my initial guess is that your different
> tape changers (well, the Director Storage resources
> and their associated changers' device resources) have the same media types.
> This would (
> might) explain what you are describing.
Yes, that's right. When I first started using the single pool technique,
I had multiple tape libraries running simultaneously
all loaded with LTO-5 media from a single pool.
At this time, the configuration is slightly different I have two LTO-5
and two LTO-9 libraries. Under the single-pool convention, there are
still cross library load conflicts among the libraries that share the
same media type, it just doesn't happen as often. Nevertheless, I've had
to abandon it and return to a four pool configuration.
I was confident that I had tracked down the code that identifies the
next available loaded media and I remember it didn't take into account
which changer the media was loaded in. I was thinking of modifying the
code to add that to the selection criteria but I remember that not being
as straightforward as I would like so I put it off. Now I can't seem to
find my notes on the topic.
> Can you show us some configurations for your Director Storage resources and
> their related Autochanger and Device resources in
> the SD(s)?
Yes, they follow. I use includes where possible so I've
manually merged these files and filtered site-centric comments,
hopefully I didn't make any mistakes but if something doesn't make
sense, it might be a result of the merging, or my ignorance.
Storage {
Name = sl01
Address = fsarchive.bpcservers.private
SDPort = 9103
Device = sl01 # must be same as Device in Storage daemon
Media Type = LTO-9 # must be same as MediaType in Storage daemon
Autochanger = yes # enable for autochanger device
}
Autochanger {
Name = sl01
Device = sl01-d1
Changer Command = "/usr/local/bacula/scripts/mtx-changer %c %o %S %a %d"
Changer Device = /dev/tape/by-id/scsi-1BDT_FlexStor_II_00DE64004865_LL0
}
Device {
Name = sl01-d1
Archive Device = /dev/tape/by-id/scsi-35000e111cb6bb001
Maximum block size = 524288
Maximum File Size = 50g
Maximum Network Buffer Size = 65536
Maximum Changer Wait = 3600
Maximum Rewind Wait = 600
Maximum Open Wait = 600
Maximum Open Wait = 1800
Media Type = LTO-9
AutomaticMount = yes # when device opened, read it
AlwaysOpen = yes
RemovableMedia = yes
LabelMedia = yes # lets Bacula label unlabeled media
RandomAccess = no
Maximum File Size = 20GB
Changer Command = "/usr/local/bacula/scripts/mtx-changer %c %o %S %a %d"
AutoChanger = yes
Alert Command = "sh -c 'smartctl -H -l error -d scsi %c'"
}
Storage {
Name = sl02
Address = fsarchive.bpcservers.private # N.B. Use a fully qualified name
here
SDPort = 9103
Device = sl02 # must be same as Device in Storage daemon
Media Type = LTO-9 # must be same as MediaType in Storage daemon
Autochanger = yes # enable for autochanger device
}
Autochanger {
Name = sl02
Device = sl02-d1
Changer Command = "/usr/local/bacula/scripts/mtx-changer %c %o %S %a %d"
Changer Device = /dev/tape/by-id/scsi-1BDT_FlexStor_II_00DE64005002_LL0
}
Device {
Name = sl02-d1
Archive Device = /dev/tape/by-id/scsi-35000e111c9469001
Volume Poll Interval = 3600
Maximum block size = 524288
Maximum File Size = 50g
Maximum Network Buffer Size = 65536
Maximum Changer Wait = 3600
Maximum Rewind Wait = 600
Maximum Open Wait = 600
Maximum Open Wait = 1800
Media Type = LTO-9
AutomaticMount = yes # when device opened, read it
AlwaysOpen = yes
[Bacula-users] Frequently Changing Pools For Media
Hi All, I have a couple of different tape changers that are used to run jobs in parallel and all share the same media. I've discovered by accident that if I put all of the media into a single pool, the media selection algorithm won't take into account which changer a tape is loaded in and will try to mount tapes from a different changer which causes problems. A simple solution is to have one pool for each changer and update the pool for a given media depending on which changer it gets loaded into. This appears to work in practice but I was wondering if anybody knows of any unintended issues that might arise from doing this on a routine basis. Thanks for any advice, Rich. ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] New bacula install
Did you drop the database and recreate it after you tried to create the tables with the root user? That sounds like there were already tables there when you ran that command, and it couldn’t recreate them because of the constraints on the existing tables. If you don’t have any data it in, I would drop and recrecreate the database; then run that script again. Dan Rich | http://www.employees.org/~drich/ | "Step up to red alert!" "Are you sure, sir? | It means changing the bulb in the sign..." | - Red Dwarf (BBC) On April 5, 2023 at 08:14:06, Andrew Meyer via Bacula-users (bacula-users@lists.sourceforge.net) wrote: I'm trying to set up bacula on AlmaLinux 8.7 w/ MariaDB 10.11 and getting the following error: $ sudo /usr/libexec/bacula/make_mysql_tables -u bacula -p Enter password: ERROR 1005 (HY000) at line 112: Can't create table `bacula`.`Job` (errno: 150 "Foreign key constraint is incorrectly formed") ERROR 1005 (HY000) at line 286: Can't create table `bacula`.`Pool` (errno: 150 "Foreign key constraint is incorrectly formed") Creation of Bacula MySQL tables succeeded. The error was much worse when I tried to run this as mysql root user. Can someone point me in the right direction to resolve this? ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users signature.asc Description: Message signed with OpenPGP using AMPGpg ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Cloud storage and bacula 11
Has anyone been able to use cloud storage reliably with Bacula 11? I am trying
to convert from using volumes on local disk to using minio as a cloud service
without much luck.
The odd thing is, it will work for a time. But if bacula-sd or bacula-dir
restarts, I end up in a state where the device status shows things like:
> Device Cloud (S3 Plugin): "cloudDrive-01" (/tmp/cloudDrive-01) is not open.
> Device is BLOCKED waiting for mount of volume "rpi-0002",
> Pool: rpi
> Media type: CloudType
> Drive 0 is not loaded.
and I don't see any obvious way to force it to mount the cloud storage into
that device node -- short of running s3fs to mount the volume as if I was doing
a restore. I would think that would not be needed since this mount request is
from a backup.
Restores are another problem, as I have yet to get one to work reliably with
cloud media. Maybe I just need to tune my settings a bit more, but doing a
mount of a volume with lots of small files via. s3fs is painful at best, and
doesn't work at all in the worst case. Having to mount that under a specifc
directory when I'm running bacula-sd in a docker container is even more
painful, since it requires reconfiguing and restarting bacula-sd any time I
want to restore something.
Here's a sample of my config (with passwords/hostnames obsfucated of course):
> Device {
> Name = cloudDrive-01
> Device Type = Cloud
> Media Type = CloudType
> Archive Device = /tmp/cloudDrive-01
> AutomaticMount = yes
> AlwaysOpen = no
> RemovableMedia = yes
> Label Media = yes
> Random Access = yes
> Autochanger = yes
> Maximum Part Size = 50MB
> Maximum File Size = 50MB
> Cloud = minio-nas
> }
> Autochanger {
> Name = CloudAutoChangerS3
> Changer Device = /dev/null
> Changer Command = ""
> Device = cloudDrive-01, cloudDrive-02, cloudDrive-03, cloudDrive-04
> }
> Cloud {
> Name = "minio-nas"
> Driver = S3
> HostName = "nas.xxx.com:9000"
> BucketName = "bacula"
> AccessKey = "bacula"
> SecretKey = "xx"
> Protocol = HTTP
> UriStyle = "Path"
> Truncate Cache = "AfterUpload"
> Upload = "EachPart"
> Region = us-xxx-1
> MaximumUploadBandwidth = 100MB/s
> }
>
Dan Rich
http://www.employees.org/~drich/
"Step up to red alert!" "Are you sure, sir?
It means changing the bulb in the sign..."
- Red Dwarf (BBC)
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Ubuntu 20.04, Bacula, and MariaDB
I asked a very similar question nearly 2 years ago (https://sourceforge.net/p/bacula/mailman/bacula-users/thread/4ef25e47-7fc9-49c9-9d8b-c8b73414c9a3@Spark/) — is there any hope of getting Debian packages for Bacula built with MariaDB in addition to MySQL? Virtually everything else I use is built with the MariaDB dependencies these days, and I can’t upgrade to Bacula 9.6 w/o building my own packages at this point. I was hoping to migrate from local disk to s3 storage, but that can’t happen w/o a newer version of bacula that what comes with Ubuntu. And even the debian.org packages are still mysql-only. Dan Rich http://www.employees.org/~drich/ "Step up to red alert!" "Are you sure, sir? It means changing the bulb in the sign..." - Red Dwarf (BBC) ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Deb packages for mariadb?
On Dec 17, 2020, 14:25 -0800, Sven Hartge , wrote: > On 17.12.20 22:49, Daniel Rich wrote: > > > The ones that come with Ubuntu 20.04 are nearly a year old (9.4.2) > > That is not something we Debian Developers can directly influence. But > you can take the Debian sources and rebuild them on a Ubuntu-System and > it will work. That is likely going to be what I end up doing. But again, they use different package names and layout from the bacula.org packages, so I have to convert my configs and scripts back to /etc/bacula instead of /opt/bacula as well as update my ansible cookbooks to use the Debian package names instead of the bacula.org package names. > > and don’t include mysql support as far as I can tell. I don’t really > > want to have to covert my configs back to the debian-supplied version > > since I’ve been running the bacula.org version for over a years. > > Nobody uses MySQL anymore, every distribution has converted over to > MariaDB by now. This will not go away. Except the packages that are built at bacula.org that still have mysql as dependencies. And that’s in the most recent packages for 9.6.6. This is what started me down the path of looking for the source package in the first place so I can build my own package for 9.6.6. The bacula-director package that comes with Ubuntu 20.04, doesn’t include any database support as far as I can tell other than possibly sqlite. I couldn’t even get it to start with the database configs I currently have. I have a vague memory that this is what had me convert over to the bacula.org packages from the OS provided ones a year or two ago. > > Since I’m building a new server, I’m considering moving to 11.0, but > > there aren’t packages for that yet, so I would still need the debian > > package files to build my own. > > I would hold back on 11 for a bit. The database transition from 9 to 11 > is a bit rocky at the moment. So far, the official word on 11 is "it is > a beta". > Hmm… I figured since it was actually announced those issues had been resolved... Dan Rich http://www.employees.org/~drich/ "Step up to red alert!" "Are you sure, sir? It means changing the bulb in the sign..." - Red Dwarf (BBC) ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Deb packages for mariadb?
The ones that come with Ubuntu 20.04 are nearly a year old (9.4.2) and don’t include mysql support as far as I can tell. I don’t really want to have to covert my configs back to the debian-supplied version since I’ve been running the bacula.org version for over a years. Since I’m building a new server, I’m considering moving to 11.0, but there aren’t packages for that yet, so I would still need the debian package files to build my own. Dan Rich http://www.employees.org/~drich/ "Step up to red alert!" "Are you sure, sir? It means changing the bulb in the sign..." - Red Dwarf (BBC) On Dec 16, 2020, 23:17 -0800, Sven Hartge , wrote: > On 17.12.20 07:22, Daniel Rich wrote: > > > Does anyone know where I can find the packaging files for bacula? I > > cloned the git repo, but while it has the source for building the > > software it doesn’t have any of the packaging files (i.e. the debian > > directory and assorted files). > > > > Since I can’t install the current binary packages without removing > > MariaDB and installing MySQL, it looks like building my own from source > > is the only way to go. And I really don’t want to have to run a manual > > install if I can build debian packages instead. > > > > Debian provides up-to-date Bacula packages working with MariaDB via > Buster-Backports. > > The only downside of those package is they don't support S3 and the > Docker-Plugin. > > Grüße, > Sven. > > > ___ > Bacula-users mailing list > Bacula-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bacula-users ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Deb packages for mariadb?
Does anyone know where I can find the packaging files for bacula? I cloned the git repo, but while it has the source for building the software it doesn’t have any of the packaging files (i.e. the debian directory and assorted files). Since I can’t install the current binary packages without removing MariaDB and installing MySQL, it looks like building my own from source is the only way to go. And I really don’t want to have to run a manual install if I can build debian packages instead. Dan Rich http://www.employees.org/~drich/ "Step up to red alert!" "Are you sure, sir? It means changing the bulb in the sign..." - Red Dwarf (BBC) On Dec 9, 2020, 16:18 -0800, Daniel Rich , wrote: > Alternatives could be used to specify what database to use, but I don’t know > that it can override dependencies that are hard-coded into the package spec. > The only solutions I know of is for either bacula-mysql to not depend on the > mysql packages, or to have a separate bacula-mariadb with the dependencies on > the MariaDB packages. > > I may take a stab at building my own packages this evening so I can move my > server over to a new box that is running Ubuntu 20.04 and MariaDB. > > Dan Rich > http://www.employees.org/~drich/ > "Step up to red alert!" "Are you sure, sir? > It means changing the bulb in the sign..." > - Red Dwarf (BBC) > On Dec 9, 2020, 11:31 -0800, Phil Stracchino , wrote: > > On 12/9/20 1:58 PM, Daniel Rich wrote: > > > This might be a better question for bacula-devel, but are there plans to > > > provide MariaDB packages in addition to the existing MySQL packages? The > > > current bacula-mysql package fails to install on my Ubuntu 20.04 system > > > with MariaDB due to the embedded mysql dependencies. I was hoping that > > > there were some compatibility packages that would take care of this, but > > > I haven’t been able to track any down. > > > > > > The following packages have unmet dependencies: > > > bacula-mysql : Depends: mysql-client but it is not going to be installed > > > Depends: libmysqlclient20 (>= 5.7.11) but it is not > > > installable > > > > > > It seems to me like this is a problem that ought to be overcome by the > > Debian alternatives system. > > > > > > > > -- > > Phil Stracchino > > Babylon Communications > > ph...@caerllewys.net > > p...@co.ordinate.org > > Landline: +1.603.293.8485 > > Mobile: +1.603.998.6958 > > > > > > ___ > > Bacula-users mailing list > > Bacula-users@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/bacula-users > ___ > Bacula-users mailing list > Bacula-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bacula-users ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Deb packages for mariadb?
Alternatives could be used to specify what database to use, but I don’t know that it can override dependencies that are hard-coded into the package spec. The only solutions I know of is for either bacula-mysql to not depend on the mysql packages, or to have a separate bacula-mariadb with the dependencies on the MariaDB packages. I may take a stab at building my own packages this evening so I can move my server over to a new box that is running Ubuntu 20.04 and MariaDB. Dan Rich http://www.employees.org/~drich/ "Step up to red alert!" "Are you sure, sir? It means changing the bulb in the sign..." - Red Dwarf (BBC) On Dec 9, 2020, 11:31 -0800, Phil Stracchino , wrote: > On 12/9/20 1:58 PM, Daniel Rich wrote: > > This might be a better question for bacula-devel, but are there plans to > > provide MariaDB packages in addition to the existing MySQL packages? The > > current bacula-mysql package fails to install on my Ubuntu 20.04 system > > with MariaDB due to the embedded mysql dependencies. I was hoping that > > there were some compatibility packages that would take care of this, but > > I haven’t been able to track any down. > > > > The following packages have unmet dependencies: > > bacula-mysql : Depends: mysql-client but it is not going to be installed > > Depends: libmysqlclient20 (>= 5.7.11) but it is not > > installable > > > It seems to me like this is a problem that ought to be overcome by the > Debian alternatives system. > > > > -- > Phil Stracchino > Babylon Communications > ph...@caerllewys.net > p...@co.ordinate.org > Landline: +1.603.293.8485 > Mobile: +1.603.998.6958 > > > ___ > Bacula-users mailing list > Bacula-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bacula-users ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Deb packages for mariadb?
This might be a better question for bacula-devel, but are there plans to provide MariaDB packages in addition to the existing MySQL packages? The current bacula-mysql package fails to install on my Ubuntu 20.04 system with MariaDB due to the embedded mysql dependencies. I was hoping that there were some compatibility packages that would take care of this, but I haven’t been able to track any down. The following packages have unmet dependencies: bacula-mysql : Depends: mysql-client but it is not going to be installed Depends: libmysqlclient20 (>= 5.7.11) but it is not installable Dan Rich http://www.employees.org/~drich/ "Step up to red alert!" "Are you sure, sir? It means changing the bulb in the sign..." - Red Dwarf (BBC) ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Configuring Bacula File Daemon on High-availability Replicated Servers
Hello, I have a system where the Bacula File Daemon must run on a high-availability pair of servers. I have some services managed by Pacemaker, but I have not configured this yet for the Bacula services. If I configure bacula-fd identically and start it on both servers in the HA pair, will there be a problem? In other words, does bacula-fd attempt to connect to bacula-dir, or does bacula-dir only connect to bacula-fd when needed? If it's the former scenario, then it is necessary for me to configure Pacemaker to start and stop bacula-fd on the HA servers so that only one has it running at a time. If it is the latter scenario, then I can setup init to start bacula-fd on both servers and simply configure bacula-dir to find its clients at the HA "floating" IP address that is managed by Pacemaker. Thanks for your help. Regards, Rich Otero Director, Technical Support and Professional Services EditShare rot...@editshare.com 617-782-0479 -- Go from Idea to Many App Stores Faster with Intel(R) XDK Give your users amazing mobile app experiences with Intel(R) XDK. Use one codebase in this all-in-one HTML5 development environment. Design, debug & build mobile apps & 2D/3D high-impact games for multiple OSs. http://pubads.g.doubleclick.net/gampad/clk?id=254741911=/4140___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Configuring Bacula File Daemon on High-availability Replicated Servers
Thanks, Dimitri. I am backing up shared data. I've done some testing with independent bacula-fd instances running at the same time with identical configuration on both nodes. Bacula-dir is configured to connect to the client at the floating IP address. It seems to work well. I have failed over the floating IP address with Pacemaker in both direction and Bacula seems to keep working as expected. I just wanted to make sure that I'm not doing something that is bad for Bacula that is going to hurt me later. Thanks for your input. Regards, Rich Otero Director, Technical Support and Professional Services EditShare rot...@editshare.com 617-782-0479 On Wed, Dec 2, 2015 at 1:56 PM, Dimitri Maziuk <dmaz...@bmrb.wisc.edu> wrote: > On 12/02/2015 12:22 PM, Rich Otero wrote: > ... > > configure Pacemaker to start and stop bacula-fd on the HA servers so that > > only one has it running at a time. If it is the latter scenario, then I > can > > setup init to start bacula-fd on both servers and simply configure > > bacula-dir to find its clients at the HA "floating" IP address that is > > managed by Pacemaker. > > What are you backing up? I run 3 bacula clients for 2 HA nodes. I have > drbd pairs where I'm backing up /etc on each node *and* the drbd > filesystem on the floating ip. > > For the latter, fd runs on a different port and is started by heartbeat. > Bacula-dir looks for it at the floating ip. If you aren't backing up any > shared data, you don't need that. > > For the former: stock setup starting from init using fixed ips. If you > aren't backing up any local data you don't need that (but you probably > want to back up your etcs and crontabs). > > HTH > -- > Dimitri Maziuk > Programmer/sysadmin > BioMagResBank, UW-Madison -- http://www.bmrb.wisc.edu > > > > -- > Go from Idea to Many App Stores Faster with Intel(R) XDK > Give your users amazing mobile app experiences with Intel(R) XDK. > Use one codebase in this all-in-one HTML5 development environment. > Design, debug & build mobile apps & 2D/3D high-impact games for multiple > OSs. > http://pubads.g.doubleclick.net/gampad/clk?id=254741911=/4140 > ___ > Bacula-users mailing list > Bacula-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bacula-users > > -- Go from Idea to Many App Stores Faster with Intel(R) XDK Give your users amazing mobile app experiences with Intel(R) XDK. Use one codebase in this all-in-one HTML5 development environment. Design, debug & build mobile apps & 2D/3D high-impact games for multiple OSs. http://pubads.g.doubleclick.net/gampad/clk?id=254741911=/4140___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bconsole Reload Sometimes Doesn't Load New Job Resource Definitions
Ben, thanks for the input. We used to have that problem too: We had a bug in our code that would result in a lot of old bconsoles being left open and we would hit the concurrent jobs limit. We fixed that some time ago, but your message and someone else's yesterday are directing me to the message. I haven't looked at those yet. Regards, Rich Otero Director, Technical Support and Professional Services EditShare rot...@editshare.com 617-782-0479 On Tue, Oct 20, 2015 at 2:47 AM, Roberts, Ben <ben.robe...@gsacapital.com> wrote: > > On a particular customer system, we find that the new job resource has > not been loaded after the `reload` command has completed. > > > > My workflow relies on the reload command heavily (puppet drops in updated > config files and executes the reload command afterwards; this normally > works very well. Occasionally I have problems and this is always evidenced > by “Error: Too many open reload requests. Request ignored” in the messages > output. The normal cause for this in my experience is too many connected > bconsoles (which appear to count against the director’s concurrent job > limit?) Killing off the old bconsole sessions that people have spawned > under screen weeks before and forgotten usually fixes this. I rarely have > to restart the director service more than once or twice a year and that’s > normally for other issues. > > > > Regards, > > Ben Roberts > > > -- > This email and any files transmitted with it contain confidential and > proprietary information and is solely for the use of the intended > recipient. If you are not the intended recipient please return the email to > the sender and delete it from your computer and you must not use, disclose, > distribute, copy, print or rely on this email or its contents. This > communication is for informational purposes only. It is not intended as an > offer or solicitation for the purchase or sale of any financial instrument > or as an official confirmation of any transaction. Any comments or > statements made herein do not necessarily reflect those of GSA Capital. GSA > Capital Partners LLP is authorised and regulated by the Financial Conduct > Authority and is registered in England and Wales at Stratton House, 5 > Stratton Street, London W1J 8LA, number OC309261. GSA Capital Services > Limited is registered in England and Wales at the same address, number > 5320529. > > -- ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Bconsole Reload Sometimes Doesn't Load New Job Resource Definitions
Hello, We make a backup product that uses Bacula as the backend. Our systems run Ubuntu v12.04 (kernel v3.12.17) with Bacula v5.2.12. Our product sets up the job resource definitions based on the user's choices, it validates the configuration with `bacula-dir -t`, and then it issues the bconsole `reload` command to load the new job resource. On a particular customer system, we find that the new job resource has not been loaded after the `reload` command has completed. If we give the `run` command, the new job is not in the list of jobs that can be executed. If we restart bacula-dir, then the job resource is loaded and it can be run. We haven't been able to figure out why `reload` isn't doing what we expect, so I hope that you can offer some advice. I have attached the Director configuration file from the affected system (bacula-dir.conf) and the file where we put the job resource definitions (jobs.conf), which is imported by the main configuration file. I've found a few relevant points in researching this problem online. The bconsole documentation seems to discourage you from using `reload`: "While it is possible to reload the Director's configuration on the fly, even while jobs are executing, this is a complex operation and not without side effects. Accordingly, if you have to reload the Director's configuration while Bacula is running, it is advisable to restart the Director at the next convenient opportunity." http://www.bacula.org/5.2.x-manuals/en/console/console/Bacula_Console.html Indeed, I have found that restarting the Director causes the new job resource to be loaded, but this is not a good solution for us because we often have many Bacula jobs running simultaneously; the "convenient" times when the Director could be restarted will be too infrequent. I also found a closed bug ticket from a user who describes practically the same problem: http://bugs.bacula.org/view.php?id=1573 Thank you in advance for your help. Regards, Rich Otero Director, Technical Support and Professional Services EditShare rot...@editshare.com 617-782-0479 jobs.conf Description: Binary data bacula-dir.conf Description: Binary data -- ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] products based on bacula
On 2009.01.19. 22:11, C M Reinehr wrote: On Mon 19 January 2009 13:57, Mag Gam wrote: It comes in the night and sucks the essence from your computers. My manager would freak if she sees this... Can't help you with that! (You'll have to take that up with Kern Sibbald. ) It is, of course, open source, so I suppose you could repackage it and call it anything you like. :-) i'm sure kern would love to provide a differently named package for a double price (like backup ueber alles or similar =) ) cmr On Mon, Jan 19, 2009 at 2:02 PM, C M Reinehr c...@amsent.com wrote: On Mon 19 January 2009 12:08, Mag Gam wrote: We are planning to implement bacula at one of our major manufacturing plants. Is there a professional version of this product or something similar to it with a different name. Our storage engineering team is in the process of writing proposals for various solutions but they are hesitant to use Bacula because of its name :-(. Any ideas if something like this exists? TIA Yes, it exists--but it still is called Bacula. ;-) http://www.baculasystems.com/eng/About-us/Recent-news/Bacula-Systems-laun ches Cheers! cmr -- Debian 'Etch' - Registered Linux User #241964 More laws, less justice. -- Marcus Tullius Ciceroca, 42 BC -- Rich -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] bacula and MySQL license
On 2008.11.07. 16:21, Silver Salonen wrote: On Friday 07 November 2008 15:57:52 LeJav wrote: Hello, I am preparing a NFS server (Linux) which is a part of a global project for a customer. This NFS server will be backuped with bacula; I use MySQL for the database. Could someone explain me if I am right with MySQL licencing ? MySQL is only used for the bacula data. My global project is not GPL; but my NFS Server and bacula are GPL. Can I install bacula and MySQL on this server with respect to GPL licencing ? Thx for your answers. Yes, you're OK. You're only affected by GPL licence if you want to change the code of either MySQL or Bacula. Otherwise it doesn't matter. actually, you can change code as hell. gpl will only kick in if you will _distribute_ said code. -- Silver -- Rich - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] bacula and MySQL license
On 2008.11.07. 18:39, Jose Ildefonso Camargo Tolosa wrote: On Sat, Nov 8, 2008 at 9:52 AM, Dan Langille [EMAIL PROTECTED] wrote: On Nov 7, 2008, at 8:57 AM, LeJav wrote: Hello, I am preparing a NFS server (Linux) which is a part of a global project for a customer. This NFS server will be backuped with bacula; I use MySQL for the database. Could someone explain me if I am right with MySQL licencing ? MySQL is only used for the bacula data. My global project is not GPL; but my NFS Server and bacula are GPL. Can I install bacula and MySQL on this server with respect to GPL licencing ? I cannot imagine any circumstances under which this would not be allowed. Yes, there are: say that the project is a propiertary backup solution, and that the company wants to use bacula and/or mysql code for their non-gpl backup solution: they can't do that. please, please, don't do this. they can change the code until nobody can recognise it or whatever. they can molest, abuse, ridicule that code. as long as it stays inhouse, it's their choice. now, if they modify and distribute the modified code... yes, now gpl kicks in and requires publishing changes (actually, that's not true either, they have to provide source to the receivers of the modifications). -- Dan Langille http://langille.org/ -- Rich - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] http://bugs.bacula.org/ does not need a login for viewing
On 2008.02.14. 05:03, Dan Langille wrote: Michael Short wrote: Thanks, this makes it easier to find existing bugs. I had no idea there was an anonymous account (so I was forced to make one...). FYI, this account was documented on the front page of the website. but some users followed buglinks from the mailing list, thus missed that and only got the login form ;) thanks for the change, though i had already created the account as well. This link is Google's cache which will soon expire I'm sure: http://64.233.169.104/search?q=cache:GnTceWz_n1gJ:bugs.bacula.org/+bugs.bacula.orghl=enct=clnkcd=1gl=us -- Rich - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Restore from command line
unmark $(cat file) ? if i understood the question :) On 2008.02.13. 12:14, le dahut wrote: Hello, I'm launching 'restore' from command line with : echo -e restore all\n5\ncd var\ncd lib\nunmark ldap\ndone\n2\nyes\n|bconsole -c /etc/bacula/bconsole.conf As you can see, I unmark some files, but I've more files to unmark and this way looks very ugly to me. Is there a way to specify all those arguments in a file and only launch one command in bconsole to run the restore job ? -- Rich - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Restore from command line
On 2008.02.13. 12:53, le dahut wrote: but : 'unmark /home/myfolder' 'unmark home/myfolder' don't work that's unfortunate, if this is the latest version of bacula, it would be a welcome improvement 'cd home' 'unmark myfolder' works so 'unmark $(cat file)' doesn't work (I've tested it to be sure and confirm). you could try cd home\nunmark whatever in the included file - but you'll probably have to be careful with quoting/escaping. creating such a file might be quite a hassle, so in that case a wrapper script could take simple list and construct a file for inclusion (using basename/dirname) - but that's a workaround for a workaround, so bconsole taking full paths would be ven more welcome ;) ... -- Rich - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Run Before Job
On 2008.01.22. 16:58, Adrián Ribao Martínez wrote: Hello, I need to make a copy of all the db in a remote computer. I saw how bacula makes a backup of the catalog, and I tried to do it in a similar way for the remote computer, but I've realized that [Run Before Job] executes the script locally ( the computer from where the director is running ), so as I need to export the databases to make a backup, I'm thinking about create a cron job in the remote computer. Is there any other way? Can bacula handle this situation (execute a script in a remote computer)? see the manual - ClientRunBeforeJob Thank you. -- Rich - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Some bacula-dir.conf doubts
On 2008.01.07. 17:55, Reynier Perez Mira wrote:
Hi every:
After install Bacula I need to configure bacula-dir to test if it works or
not. Reading the manual left by Mathieu [1] I have a lot of questions. In the
section: Server (Director) Configuration (bacula-dir.conf) I see:
### DEFINE DIRECTOR #
Director
{
Name = backup-dir #or whatever name you want
Description = Bacula Director - $MY_LOCATION # Bacula Director - Main
Office
DIRport = 9101
QueryFile = /etc/bacula/query.sql
WorkingDirectory = /var/bacula
PidDirectory = /var/run
Maximum Concurrent Jobs = 2
FDConnectTimeout = 1min
SDConnectTimeout = 1min
Password = $CONSOLE_PASSWORD # gsdfgfdsg44
Messages = Daemon
}
The Password parameter is generated automatically? Because I have a large and
strong password here that I never wrote before I think.
yes
Now in this other section:
### DEFINE CLIENTS ##
Client {
Name = $SOME_IMPORTANT_CLIENT-fd # accountant-fd
Address = $FQDN_OR_IP # accountant.mydomain.local
FDPort = 9102
Catalog = GiantCatalog
Password = $CLIENT_ACCESS_PASSWORD # strong_password
File Retention = 6 days
Job Retention = 6 days
AutoPrune = yes
Maximum Concurrent Jobs = 2
}
I have only two doubts: In Address parameter I can write the IP address and
not the DNS name?
of course :)
it even says ..._OR_IP ;)
The Password Parameter is the administrator account in the workstation in
wich I have Bacula Client installed, no?
hmm. no, it is the same as defined in bacula fd configuration file for
that machine
This are for now
Cheers, thanks in advance and waiting for answers
[1]
http://gentoo-wiki.com/HOWTO_Encrypted_Backup_System_Using_Bacula_and_GnuPG
Ing. Reynier Pérez Mira
-
This SF.net email is sponsored by: Microsoft
yay =)
--
Rich
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] auto labeling with python
On 2007.11.30. 18:10, Jason Joines wrote:
I'm using Bacula 2.2.5 and have tried auto labeling with
LabelFormat but haven't had much luck. I don't know python but since
LabelFormat is deprecated I thought I'd go ahead and give it a try. My
well, bad news for you. python seems to be the one who will go away.
labelformat variable expansion also is supposed to disappear at some
point, but in the end it was suggested to simply use labelformat for now
and deal with any potential changes in future.
goal is to get volume names ( and the corresponding file name on disk,
I'm backing up to disk ) to be formatted like jobname-date-two digit
number of volumes for this job. I've also set the max volume size to 5
GB. So if I had a job named sivadaily run at 0300 and backup 9 GB I
would like the resulting two volumes to be sivadaily-20071130fr0300-01
and sivadaily-20071130fr0300-02.
currently i am using the following labelformat value :
Label Format =
${Pool}_${Job}-${Level}-${Year}.${Month:p/2/0/r}.${Day:p/2/0/r}-${Hour:p/2/0/r}.${Minute:p/2/0/r}
it should be possible to get out of it and bacula manual the desired
format for you.
I started with the sample director startup script and the
documentation at
http://www.bacula.org/dev-manual/Python_Scripting.html#SECTION00356.
I defined jobname = job.JobName and modified job.VolumeName in
NewVolume() to be 'job.VolumeName=%s-%d % (jobname, numvol)'. Then I
ran a test job, cas-testdaily, and got volume autolabeled as 'Labeled
new Volume jobname-44'.
So I guess definition of jobname didn't work. Also, I'm assuming
the %s and %d are for string variable and integer variable although a
bit of googling and python manual reading hasn't verified that.
Any suggestions for how to accomplish what I'm after and
recommendations for learning the python basics demonstrated in the
sample script?
Jason Joines
=
--
Rich
-
SF.Net email is sponsored by: The Future of Linux Business White Paper
from Novell. From the desktop to the data center, Linux is going
mainstream. Let it simplify your IT future.
http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] All the problems with Bacula
On 2007.11.29. 23:53, John Stoffel wrote: mark == mark bergman [EMAIL PROTECTED] writes: mark = I've got a plan to sit down and sketch out more consistent and clear mark = set of commands for manipulating bacula from the command line. Anyone mark = else like the 'bcli' name? mark = mark No! No! This will only exacerbate the problem (if one defines mark the problem as multiple tools/methods, with varying mark documentation and different levels of support for each mark tool). My goal is to *replace* bconsole with a better tool, which does all that bconsole does but in a more consistent manner. And which offers cleaner interface to the underlying bacula commands and processes. isn't it possible to extend and improve bconsole instead of a complete rewrite ? If you look at the source code to bconsole, the parser is very very simplistic and doesn't have any notion of command completion, or even the ability to give help at each level. i have wanted better cli (maybe using readline or what is it everybody else is using) for, like YEARS :) ... Why do we have N different commands (cancel, disable, enable, run, show, wait?) for handling jobs? It's a pain to figure out/remember which one to use. I think it should be broken down more like: job cancel|disable|enable|run|status|* jobid|jobname good idea. this reminds me somewhat about lvm umbrella program that was introduced with lvm2 or so. instead. It shrinks down the help screen, it makes it obvious that all job related commands are grouped together, etc. ... -- Rich - SF.Net email is sponsored by: The Future of Linux Business White Paper from Novell. From the desktop to the data center, Linux is going mainstream. Let it simplify your IT future. http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Multiple commands in RunScript possible?
resending to list, too
--
On 2007.11.14. 16:13, Vladimirs Vecgailis wrote:
Hi all!
Following question - is it possible to run multiple commands with RunScript
on client side?
I.e.
RunScript {
RunsWhen = Before
FailJobOnError = Yes
Command = /etc/init.d/someservice stop; lvcreate -s -n
myservicebackup /dev/mylv/servicelv; /etc/init.d/someservice start
i would suggest using a script (i was doing that in a very similar
scenario).
script would also take care of any errors and could return nicely
desired code for bacula to decide whether backup should proceed or not.
it would also make sure that, if snapshot creation fails, service is
started again (which might not happen in the second configuration).
obviously, don't forget runafter job which removes the snapshot (unless
for some reason it will be kept)
RunsOnClient = yes
}
I tested such Definition, but only the first commans was executed -
someservice stop.
All other have been ignored (?).
What i done now is, i defined multiple RunScript-Definitions and arranged
them in the order i need:
RunScript {
RunsWhen = Before
FailJobOnError = Yes
Command = /etc/init.d/someservice stop
RunsOnClient = yes
}
RunScript {
RunsWhen = Before
FailJobOnError = Yes
Command = lvcreate -s -n myservicebackup /dev/mylv/servicelv
RunsOnClient = yes
}
RunScript {
RunsWhen = Before
FailJobOnError = Yes
Command = /etc/init.d/someservice start
RunsOnClient = yes
}
1. Question - is it possible to define multiple commands?
2. Question - is such hack - definition of multiple RunScripts ok?
...
--
Rich
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] python support removal
you probably wanted to send this to list ;) On 2007.11.05. 13:01, Marek Simon wrote: I think some scripting support is good for various hacking and unusual solutions, but I think python is kind of exotical language. Very few users know it, very few projects is written in it and on universities it's tought very rarely. May be some other language would be appriciate. it's not that exotic, actually. you can even write openoffice.org macros in it :) as for the language itself, i can't comment. i don't code, and if i can configure it for my liking once, that's all i want. oh, and i also want it not to break on upgrades =) MArek ... -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] [hugin-ptx] Re: Hugin version numbering
On 2007.11.02. 18:25, Peter Crowley wrote: On Nov 1, 5:25 pm, Yuv [EMAIL PROTECTED] wrote: I would suggest using an even minor for production releases and an odd minor for test releases, like the Linux kernel does. Yuv If we do evens and odds then would we have two branches, one stable and another unstable and switch back and forth between them? With evens and odds you could conceivably have the scenario of 7.6 with new features followed by 6.4 as a bug fix to the stable version without the new features. first number is rarely used for that, it's usually second one. it would mean using 0.8.z as a release branch. as soon as development get's ready to start releasing alphas/betas/whatever, those are not released as 0.9 beta 18, but as 0.9.z. next stable release thus is 0.10.0 this does not require supporting both branches separately, but gives an easy way to do so, if desired. as for actual numbering, modeling it after a disfunctional installer doesn't seem a good idea. I'm very comfortable with a single development chain. If we continue with a single chain then the evens and odds mean we release 8.0 and immediately begin development on 9.1. Other than burning through the numbers faster I see no real downside. But I'm also not sure what the benefit would be. -Peter -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Fatal Python error: Interpreter not initialized
regarding issue http://bugs.bacula.org/view.php?id=998 i can't attach a comment to it - maybe i have to reopen it first ? i don't want to do that unless python is kept, just wanted to note that only single python version is installed on the machine, no additional modules are installed and bacula was compiled on the same machine. i have no idea about python, but attempting to import sys and datetime in a separate script works without errors. -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] That Doh! moment - FD to SD communication
On 2007.10.31. 03:01, David Legg wrote: I've spent the last two weeks compiling, downloading, installing, building and configuring. Just when I think I'm nearly there a 200 pound gorilla jumped out and ruined everything ;-) I have a set of remote servers I'd like to keep backed up over the internet by using Bacula to save files to a local server behind a router. All was well as I finally managed to get the director daemon on the local server to connect to the file daemon on a remote machine through port 9102. Then, when I actually attempted to run a backup job I noticed: - 30-Oct 23:56 frenzy-fd JobId 6: Fatal error: Failed to connect to Storage daemon: optimus:9103 That's when I had my Doh! moment. I think that when the file daemon (on the remote server) gets its marching orders from the director (on the local machine) the file daemon attempts to contact the storage server (on the local machine) using information supplied to it by the director. Of course this is doomed to failure because the director and the storage server are on one and the same machine and in any case the IP address is dynamically assigned by the DHCP server and is different to what the File daemon would see because of NAT. somehow i don't think it is a good idea to have servers using dhcp, and a dynamical at that. the most sane solution - set the server to use static ip. next solution - set dhcp server to issue single ip to that server. possible next solution - use ddns i'd go with the first :) The only practical way I can see to break this problem is to reverse the direction of the connection so that the director tells the storage daemon to contact the file daemon and not the other way around. This would work because the remote client machines have a FQDN and are thus easier to reach. Don't suppose this is possible is it ? I can't believe I've come so close only to be thwarted at the last hurdle. Best Regards, David Legg -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula and Day saving time
On 2007.10.30. 21:53, Arno Lehmann wrote: ... That said, I'd expect Bacula will skip the backup, given its behavior in this case. One way to solve this problem, I guess, would be to never start backups between 2 and 3 AM. yup. Very reasonable, but IIRC a better solution was implemented in Bacula, probably since 2.0. I couldn't find this with a grep in ReleaseNotes, but I recall that being discussed a year or two ago on the mailing lists. good to hear. if anybody knows more, it would be very nice to have this behaviour clearly documented. so, is there some dst intelligence in latest versions of bacula ? if not, maybe a comment, warning about this should be added to default configuration file schedules :) No, they weren't, but depending on where you live, there was no problem this weekend. :) See above - I believe there is the necessary intelligence in Bacula. Arno How do other packages handle this? I believe cron is one of the more intelligent programs when dealing with this. Perhaps anacron? depends (c) if i remember correctly, vixie cron was doing all kinds of automagic detection, dcron was leaving this all up to the admin. most important thing - having this all well documented, -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] python support removal
see http://bugs.bacula.org/view.php?id=998 After version 2.2.x Python will no longer be support -- there are an insufficient number of users. which kinda worries me a bit. what about Label Format directive ? if python was supposed to replace it, but is being removed itself... will this directive stay supported ? arno, i guess you will not have to devote time for python scripts anymore ;) -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] python support
On 2007.10.29. 23:38, Arno Lehmann wrote: ... import datetime and this is where it breaks. just adding that line to the script prevents director from starting at all : Fatal Python error: Interpreter not initialized (version mismatch?) That looks like a python problem... or something related to the way python is embedded into Bacula. You should consider a bug report at bugs.bacula.org, and point out that this might be a configure or linking problem, or even one of finding the right python module path. done as http://bugs.bacula.org/view.php?id=998 ... actually, i would prefer to error out in such a case - if volume exists, something's broken. Well, relying onto the number of volumes in a pool is not safe, for example. Imagine you have volumes V1, V2, V4 in a pool and label with 'V'+number. The next volume will be the fourth one, so you'd end up with V4 a second time. oh, no, i am using single volume (file) per job, and each is named with date-hour-minute. if the same job is executed twice in a single minite (especially as i have only one concurrent job configured), something's not right :) As this can happen easily when you delete volumes, precautions are important. looking over the example, i could not find a way to cancel the job - how would i do that ? I don't know - perhaps return the null value from the method. You'll have to try :-) nope, returning 0 still allows to proceed :) ... Arno -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] problems when not running director as root
it seems there are some problems if director is not run as root. using bacula 2.2.5 on slackware 12. 1. important environment variables are not set if bacula is configured and compiled with --with-dir-user and --with-dir-group, director is started with correct -u parameter, and runs as the correct user - but environment variable HOME is not set to user home, it is set to /root. obviously, make_catalog_backup can't access /root/.my.cnf, and it also doesn't know anything about the correct location for .my.cnf. this way it is not possible to use (without modifying scripts themselves) suggestion from http://www.bacula.org/dev-manual/Catalog_Maintenance.html#SECTION002413 this might also apply to other daemons. would this qualify as a valid bug report ? 2. emails are not sent in default configuration if bacula is configured and compiled with --with-dir-user and --with-dir-group, /sbin/bsmtp is still owned by root.root, and it's permissions are set to rwxr-xr--. this prevents director from sending out mails in the default configuration. maybe it would be better (if user and group are set when compiling) to make permissions 755 or set owning group to one the director is using ? would this qualify as a valid bug report ? -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Problem backing up catalog
On 2007.10.24. 16:53, Dan Langille wrote:
On 24 Oct 2007 at 14:39, Simon Barrett wrote:
...
On this matter; adding the password to the RunBeforeJob line causes my
database password to appear on the status emails:
24-Oct 13:09 fs01-dir: BeforeJob: run command
/etc/bacula/make_catalog_backup
bacula bacula MyPasswordHere
Status emails are sent in clear text across our network. Is there a
recommended solution to include sensitive variables in the config files
without exposing them like this?
http://www.bacula.org/dev-manual/Catalog_Maintenance.html
Click on Security considerations
using bacula 2.2.5, there seems to be several problems with these
suggestions.
one (when not running director as root) i am outlining in another email,
problems when not running director as root.
another one is that even if i hardcode password in the
make_catalog_backup scripts, something prints the password to messages,
if python scripting is enabled.
the line looks a bit weird, as it contains what seems to be another line
at the end.
30-Oct 11:40 backup_1-dir JobId 52: ('bacula', None, 'bacula',
'baculapassword', None, 0, 'MySQL')30-Oct 11:40 backup_1-dir JobId 52:
python script is slightly modified default example, and looking over it,
i can't see any report line which could output anything like that.
what could be the reason for this line ?
--
Rich
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Problem backing up catalog
On 2007.10.30. 12:05, Arno Lehmann wrote:
...
the line looks a bit weird, as it contains what seems to be another line
at the end.
30-Oct 11:40 backup_1-dir JobId 52: ('bacula', None, 'bacula',
'baculapassword', None, 0, 'MySQL')30-Oct 11:40 backup_1-dir JobId 52:
python script is slightly modified default example, and looking over it,
i can't see any report line which could output anything like that.
The above looks like a printout of the CatalogRes provided as an
attribute by Baculas job object.
ah, indeed - i was looking for report lines, missed print lines...
print job.CatalogRes
seems that default example script suffers from some problems :)
Arno
what could be the reason for this line ?
--
Rich
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Encryption and compression
On 2007.10.30. 13:21, Eric Böse-Wolf wrote: I did a few test and maybe someone is interested in the results. I did a backup of the linux kernel documentation, quite a lot of text (du -sh . gives 22M in the directory), under different conditions, each of those backups I wrote to a new file volume and here are the sizes of the file volumes: Keylength Compression(GZIP) Resulting Mediasize (Bytes) noneno 17299177 noneyes 6638157 1024no 18840845 2048no 19751779 4096no 21519151 1024yes 8179802 2048yes 9091106 4096yes10858478 If you compress and you have a damaged tape, you will lose more date as if you did not use compression, but for encryption the tape is wothless with only a single failure to, so I think if you use encryption you can also use compression, if you have the CPU power for it. (The time a backup needed increased significantly with the keylength, but I did not measure exact times here.) maybe this should be added to the wiki ? Yours sincerely, Eric Böse-Wolf -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula and Day saving time
On 2007.10.30. 17:51, Chris Hoogendyk wrote: Ryan Novosielski wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Are you certain that the DST dates did not move in your area? I'm not sure which zones expanded DST and which did not. I realize that is not your question, but it is something I wanted to mention, since my timezone does not switch until Nov 4th. That said, I'd expect Bacula will skip the backup, given its behavior in this case. One way to solve this problem, I guess, would be to never start backups between 2 and 3 AM. yup. Daylight Savings time laws changed affective spring 2007 in the U.S. and Australia. I have no clue about the Czech Republic or others. There were patches to be applied according to your operating system. I applied Solaris 9 patches. Mac OS X and Windows systems got theirs by automatic updates. One of our labs had an older Sun OS 5.6 system for which there were no patches. After messing with it for a while, they finally punted and said it didn't really matter. ... i don't think timezone changes was the question here :) actual question would be - how does bacula actually handle these times. i also had one job executed twice, but given that it's still 1.36, i didn't bother much. so, is there some dst intelligence in latest versions of bacula ? if not, maybe a comment, warning about this should be added to default configuration file schedules :) -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] python support
On 2007.10.27. 17:27, Arno Lehmann wrote:
...
2. if Scripts Directory is specified in director configuration file,
director can't be stopped with ctrl+c, when run in foreground;
I don't know if this qualifies as a bug - because, in normal
circumstances, you should only run the DIR in foreground under the
debugger (Kern says ;-) - but I don't think this is serious.
yeah, but consistency and obeying ctrl+c in foreground are good ;)
...
1. the example is using both lines with spacing and without spacing
around equal sign (a=b and a = b). which would be a better style choice ?
I prefer the spaces around operators because it creates much better
readability. Others disagree.
actually, i don't care that much about the method used as for it being
consistent in the example script :)
...
4. does somebody have example on writing out datetime string that would
reproduce behaviour that label format currently produces for
${Year}.${Month:p/2/0/r}.${Day:p/2/0/r}-${Hour:p/2/0/r}.${Minute:p/2/0/r}
?
Again, no examples, manuals, or even working code, but something like
import datetime
and this is where it breaks.
just adding that line to the script prevents director from starting at all :
Fatal Python error: Interpreter not initialized (version mismatch?)
Kaboom! bacula-dir, backup_1-dir got signal 6 - IOT trap. Attempting
traceback.
Kaboom! exepath=/sbin/
Calling: /sbin/btraceback /sbin/bacula-dir 32340
execv: /sbin/btraceback failed: ERR=No such file or directory
Traceback complete, attempting cleanup ...
Orphaned buffer: backup_1-dir 16 bytes buf=820b150 allocated at
watchdog.c:86
Orphaned buffer: backup_1-dir 16 bytes buf=82330a8 allocated at
watchdog.c:87
on #python, it was suggested that :
that error is caused by importing a module compiled for a different
version of Python, on Windows.
but i don't understand how could that be possible
where could the problem be ?
...
5. a (slightly modified) line reads :
job.JobReport = Exists=%d %s % (job.DoesVolumeExist(Vol), Vol)
if i understand correctly, it would only add a line to the report and
proceed, right ?
Yup.
it seems that in such a case job simply stucks, waiting for a manual
volume specification.
can i make a check that would error out if such a volume already exists ?
Sure. Check the volume name you just created using DoesVolumeExist,
and if it already exists, append or modify a suffix and loop.
Otherwise, continue.
actually, i would prefer to error out in such a case - if volume exists,
something's broken.
looking over the example, i could not find a way to cancel the job - how
would i do that ?
Hope this helps,
it did a lot, huge thanks.
i hope that the result of my pain will help somebody else :)
Arno
--
Rich
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] python support
On 2007.10.29. 19:09, Rich wrote: ... import datetime and this is where it breaks. just adding that line to the script prevents director from starting at all : Fatal Python error: Interpreter not initialized (version mismatch?) Kaboom! bacula-dir, backup_1-dir got signal 6 - IOT trap. Attempting traceback. Kaboom! exepath=/sbin/ Calling: /sbin/btraceback /sbin/bacula-dir 32340 execv: /sbin/btraceback failed: ERR=No such file or directory Traceback complete, attempting cleanup ... Orphaned buffer: backup_1-dir 16 bytes buf=820b150 allocated at watchdog.c:86 Orphaned buffer: backup_1-dir 16 bytes buf=82330a8 allocated at watchdog.c:87 on #python, it was suggested that : that error is caused by importing a module compiled for a different version of Python, on Windows. a bit more on this :) You have to make aboslutely sure that what Bacula links against, at runtime, is the right version. You also have to make sure that the modules you end up importing at runtime are for the same Python version, and also that this Bacula program properly initializes the Python it embeds. as i have no idea how to proceed, i would appreciate any help in debugging and resolving python scripting in bacula. ... -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] python support
i've recompiled bacula 2.2.5 with python. configure output claims it is enabled : Python support: yes but even starting director with -f -d90 doesn't produce any messages - manual claims that a message should be printed : 09-Jun 15:14 bacula-dir: ERROR in pythonlib.c:131 Could not import Python script /etc/bacula/scripts/DirStartUp. Python disabled. i did not do 'make clean' after enabling python - was that required ? -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] python support
On 2007.10.26. 14:04, Rich wrote:
i've recompiled bacula 2.2.5 with python. configure output claims it is
enabled :
Python support: yes
but even starting director with -f -d90 doesn't produce any messages -
manual claims that a message should be printed :
09-Jun 15:14 bacula-dir: ERROR in pythonlib.c:131 Could not import
Python script /etc/bacula/scripts/DirStartUp. Python disabled.
i did not do 'make clean' after enabling python - was that required ?
ok, this didn't help. even after a clean compile bacula dos not attempt
to use scripts.
i even tried with strace -e file, not even a single attempt to load any
python script.
turns out, i must explicitly add Scripts Directory parameter... from
manual, i understood that bacula would be looking in /etc/bacula/scripts
by default - it says normal default /etc/bacula/scripts.
so, some suggestions/problems i noticed :)
1. (normal default /etc/bacula/scripts, - unclosed parenthesis in the
manual.
also, this paragraph probably should note that the directive has to be
added and no actual default is enabled.
2. if Scripts Directory is specified in director configuration file,
director can't be stopped with ctrl+c, when run in foreground;
3. if bacula is compiled with python support, maybe it is worth adding
this parameter to the default director configuration file ?
--
now, for the actual scripts...
1. the example is using both lines with spacing and without spacing
around equal sign (a=b and a = b). which would be a better style choice ?
2. it seems that i got it partially working by using the following lines :
Vol = %s_%s-%s-%d % (job.Pool, job.Job, job.Level, numvol)
job.JobReport = Exists=%d %s % (job.DoesVolumeExist(Vol), Vol)
job.VolumeName=Vol
are there any huge mistakes ?
3. can i somehow avoid this %d, %s thing and specifying variables after
that ? like, writing someting like {job.Pool)_{job.Job}-{job.Level} etc.
4. does somebody have example on writing out datetime string that would
reproduce behaviour that label format currently produces for
${Year}.${Month:p/2/0/r}.${Day:p/2/0/r}-${Hour:p/2/0/r}.${Minute:p/2/0/r}
?
5. a (slightly modified) line reads :
job.JobReport = Exists=%d %s % (job.DoesVolumeExist(Vol), Vol)
if i understand correctly, it would only add a line to the report and
proceed, right ?
it seems that in such a case job simply stucks, waiting for a manual
volume specification.
can i make a check that would error out if such a volume already exists ?
--
Rich
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Problems with 'Automatic Volume Labeling'
On 2007.10.23. 22:02, Arno Lehmann wrote:
...
That requirement, by the way, is not very problematic. The sample
given in the manual should almost work out of the box, and python is
easier to learn than Baculas variable substitution language :-)
i do not agree with that ;)
This is described in the manual, for example
http://www.bacula.org/dev-manual/Python_Scripting.html#SECTION00356
...
i'm now trying to understand at least something in all this...
1. the example has a section that is not simply stating noop, but is
preceded by it :
def JobInit(self, job):
noop = 1
if (job.JobId 2):
startid = job.run(run kernsave)
job.JobReport = Python started new Job: jobid=%d\n % startid
print name=%s version=%s conf=%s working=%s % (bacula.Name,
bacula.Version, bacula.ConfigFile, bacula.WorkingDir)
manual says If you do not want a particular event, simply replace the
existing code with a noop = 1.
what does this section do then ? is preceding it with noop = 1 also
disabling it ?
2. the volume label itself.
i guess i should leave at least parts of def JobStart, right ?
it creates JobEvents class, which in turn includes what seems to be the
correct section - def NewVolume.
i suppose this line should be modified :
Vol = TestA-%d % numvol
later two other lines have same string :
job.JobReport = Exists=%d TestA-%d % (job.DoesVolumeExist(Vol), numvol)
job.VolumeName=TestA-%d % numvol
must i replace all of these strings ? if yes, can't they just be
referenced from the first string, Vol ?
now, as for constructing the volume labal string... it seems that most
variables map to something job. - like job.Pool, job.Job, job.Level etc.
how would i use these variables to define the label ?
i am trying to imitate a volume label like
${Pool}_${Job}-${Level}-${Year}.${Month:p/2/0/r}.${Day:p/2/0/r}-${Hour:p/2/0/r}.${Minute:p/2/0/r}
which would also require variables like year, month etc. how can i
include those ?
i'm really lost with all this, but this probably is the best time to do
this as i have set up a test system before attempting a large upgrade,
so postponing the change can come back to me later...
...
Arno
--
Rich
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Problems with 'Automatic Volume Labeling'
On 2007.10.23. 11:40, Arno Lehmann wrote: Hello, ... My only solution for this right now is to estimate how much space the Backup will use, and label the Volumes before the Job runs, but as you can imagine thats annoying ... Yes. is there anything i'm missing that this happens ? thanks for any hint ... I'd first suggest to drop the LabelFormat directive. It's deprecated and will vanish some day. Instead, use a python event to provide names for new volumes. umh. i must have missed this... will it really be dropped ? as i am using it, i don't fancy learning python just to replicate its functionality. i am using this parameter to create single file per backup job and give them meaningful names - would it indeed at some point get replaced by a requirement to script in python ? This is described in the manual, for example http://www.bacula.org/dev-manual/Python_Scripting.html#SECTION00356 ... -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] make_catalog_backup problem with 2.2.5
on a slackware 12 box, bacula is creating make_catalog_backup file with
the following line :
${BINDIR}/mysqldump -u $2$MYSQLPASSWORD$MYSQLHOST -f --opt $1 $1.sql
obviously, this is not working.
is this a known problem ?
--
Rich
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] make_catalog_backup problem with 2.2.5
On 2007.10.23. 19:17, Dan Langille wrote:
On 23 Oct 2007 at 19:04, Rich wrote:
on a slackware 12 box, bacula is creating make_catalog_backup file with
the following line :
${BINDIR}/mysqldump -u $2$MYSQLPASSWORD$MYSQLHOST -f --opt $1 $1.sql
obviously, this is not working.
is this a known problem ?
Sorry, I see you put that in the subject. It helps to repeat such
critical information in the body.
whoops, i usually do that, somehow missed this time :)
Yes, I can confirm that I can repeat the problem. There has been a
patch committed recently.
thanks for the information
In the meantime, add spaces:
yes, exactly what i did after looking at the line wondering how did it
end up like that :)
$2 $MYSQLPASSWORD $MYSQLHOST
--
Rich
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] mysql threadsafe client lib requirement
i stumbled upon the requirement for threadsafe mysql client library. the only way i found out what's wrong was by searching list archives :) it would be nice if requirements page at http://www.bacula.org/dev-manual/System_Requirements.html (and probably also for stable version) would explicitly state this. even better would be if configure would not succeed with non-threadsafe client lib and spit out a meaningful message, hinting at threadsafe requirement (currently configure succeeds, but make fails in a quite cryptic way) -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Backup to file - delete on monthly basis
not yet (well, not without using some heavy sql magic). check out 'projects' at bacula page (it's like a feature request list) - this feature is number 5, if i remember correctly On 2007.09.25. 19:10, sgraves66 wrote: I have a full backup scheduled monthly and incremental daily. Since space is limited, every month during the full backup, I'd like to completely delete the prior months backups. At the moment, I'm doing this manually prior to the monthly full. Is there a way to automate this process? -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] How to purge files from a volume?
On 2007.10.05. 17:30, Foo Bar wrote: ... What I would like to do is to keep for example 2 full backups, and as soon as the third is made the oldest one should be deleted from disk (not just the job/volume entries in the catalog). it seems that you might be interested in feature request #5 : http://www.bacula.org/?page=projects 'Remove Volume After' is of no use since I want to keep the volume, just control its contents. actually, if you would limit each volume to a single job, it would nicely achieve the goal, i think Whether 'Volume Data Retention' is useful depends on how it is implemented. If it's tied to a time period then it's again not finegrained enough (depending on the timing of individual (manual) jobs it could delete the previous full backup before the next one is made, or keep superfluous files around too long). If it's tied to an action like an explicit command (purge all files before previous full backup or similar), great. i think the only way to implement is the second way. everything else would be useless :) I guess I could work around this by using multiple volumes and 'Maximum Volume Files' or 'Maximum Volume Duration', but these have the same drawbacks as 'Volume Data Retention': if I make extra (incremental) backups in the meantime or am late with the full backup the next volume will have useless incremental stuff on it, thus wasting space, and involving extra work finding a complete restore. well, ideally, for on-disk backups, there would be a storage that is finegrained enough (to allow easy syncing changes only to a remote system) and that can really use the benefits of the disk based solution. base jobs would be one step further to such a situation. in the end, a system/format that would back up only files that change and operate on a file level at the server end would be ideal - and do that with putting as little load on the server as possible (base + consolidation jobs would soon become quite resource intensive). such a system should be able to easily and without much overhead update and remove files that are backed up, and provide a control over the backups at file level - i think, in tivoli there were options to keep x amount of last versions for a file that still exists and y copies of a file that has been deleted. -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] backward compability
On 2007.10.02. 18:57, Marek Simon wrote: Hi, I have director 1.38.11 from stable debian release. I tried to get windows client of the same version, I searched for it very long, but I did not succeed. I tried a new windows client (2.2.4), but it does not work (as said in manual). Can I dig winbacula-1.38.11 installation program somewhere in the world (or in software heaven or software hell)? Or do I need to upgrade all clients, storages and director to 2.x.x? hmm. i am using 36.1 director/sd and 2.2.something agents. works fine, also restoration ;) Thanks. Marek ... -- Rich - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] cluster e redundancy of Bacula
i submitted a feature suggestion for that some time ago. it didn't go in the 'projects' list, though :) On 2007.10.03. 11:01, MasterBrian wrote: Greetings, anyone here have experience in clustering bacula to have geographical redundancy? I've looked up into the manual and google, but I've not find anything usefull. Thank you -- Rich - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] cluster e redundancy of Bacula
On 2007.10.03. 11:17, MasterBrian wrote: Hi, Are you doing any self-made clustering police while waiting? :) well, no. simple syncing at file level with manual failover ;) it's still highly problematic as all clients also would have to be reconfigured. Rich ha scritto: i submitted a feature suggestion for that some time ago. it didn't go in the 'projects' list, though :) On 2007.10.03. 11:01, MasterBrian wrote: Greetings, anyone here have experience in clustering bacula to have geographical redundancy? I've looked up into the manual and google, but I've not find anything usefull. Thank you -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Restoring /etc/hosts on rhel4
On 2007.10.01. 21:41, Arno Lehmann wrote: ... When I finish the restore, no files are restored, except the etc/sysconfig/networking/profiles/default directory structure. Now, on the real filesystem, there is a hosts file in that directory, as well as /etc/hosts. But they are not symlinked. ... the two files are not symlinked but hard links to only one set of data. It seems like Bacula is getting confused about which hosts file I actually want restored. It happens if I set the restore client to the same machine, or a different one. I think that, in case these files are hard links, Bacula should restore the one you explicitly selected, i.e. /etc/hosts. What it should do the other hard links is something *I* wouldn't want to decide... so the files were hardlinked. but, from reading the above, it seems that _none_ of the files was actually restored... reading the manual, i would expect even with hardlinks=no one file to be restored, and only the correct one :) also, the manual could be a bit more clear. it says : In such a case, set hardlinks=no and hard links will not be backed up. Note, using this option will most likely backup more data and on a restore the file system will not be restored identically to the original. i think last sentence refers to 'hardlinks=no' setting, but at first i thought it was referring to '=yes' :) as for what to do - i would expect all hardlinked files to be restored. if not, the system would not be restored as it was. also, what happens if you restore one hardlinked file (and others are not restored) - it has to be unlinked (and should only happen if hardlinks=no is explicitly set) Arno Perhaps this is just my inexperience with the software, but can someone offer a suggestion as to what the problem might be? Maybe someone else can confirm this happens with them? Thanks Jason ... -- Rich - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] multiple messages records
i am reading the manual and experimenting with some configurations, but can't get multiple 'messages' directives referenced from a single job (or jobdef in this case). specifying multiple Messages = in a jobdef resulted in director complaining and not starting up. is this possible in some way ? -- Rich - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Strange backups
On 2007.08.23. 14:51, Angel Mieres wrote: I will try to answer all things that you said friends. - I don't use compression method to do a backup. - The database is postgres and during backup isn't taking all CPU or memory, only a 10% CPU. Process like bacula-fd/sd/dir exactly the same. what about io-based database load ? how high are iowait values during the backup ? - Im not backing up to a tape, the two backups are from: - /usr to a volume file stored on /tmp (~7 files, 2Gb) - /root to a volume file stored on /tmp ( ~100 files, 1,5Gb) - The way to tar or gzip the files is not possible, i need stimate to need a backup from 15 diferent servers and this could be crazy. Thanks to all. ... -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Problems linking MySQL Bacula
On 2007.08.17. 11:35, Aitor wrote: Hello, I have never used Bacula, and now I'm configuring it under Fedora 7. I've installed all the packages with yum I've configured the conf files, although, now when I try to link mysql with bacula I have some problems. In order to link the user-manual explains to following command: ./configure --with-mysql=mysql-directory you most probably have installed binary bacula package - so you don't have to compile it anymore. However I don't know in what path to do it, I installed bacula with yum and It is in /etc/bacula, /usr/libexec/bacula Any idea? Thanks. Aitor -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Where do we go after Bacula 2.2.0?
On 2007.08.20. 15:37, Kern Sibbald wrote: On Monday 20 August 2007 11:42, Alan Brown wrote: My observation: Item 1: Accurate restoration of renamed/deleted files Item 3: Merge multiple backups (Synthetic Backup or Consolidation) ... Item 1 is a very complex problem that has serious performance implications depending on how it is implemented particularly for the FD, and is a major addition to the current code. Probably the best solution that scales is to push the work out to the client (FD). However, doing so risks to overrun the capacities of the FD. The project involves sending a full and accurate state of the Client as known in the Bacula catalog to the client, which would then reference this information (potentially very large) when backing up files. i suppose checking only modified directories (and working on from there) is considered not safe enough, right ? in most cases this would probably result in less work, but some implementations that do not update directory mtime might break backups badly... This project has certain aspects in common with Item 7 Implement Base jobs, which also must have a full and accurate state of the catalog at the disposal of the Client. Regards, Kern -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] can't load library: libmysqlclient_r.so.15
On 2007.08.10. 01:29, Jarrod Lewis-Peacock wrote: ... /etc/bacula/bin/bacula-dir: error while loading shared libraries: libmysqlclient_r.so.15: cannot open shared object file: No such file or directory how did you install bacula (that is, did you compile it yourself or from packages) ? do you have a file named libmysqlclient_r.so.15 anywhere at all ? I searched the archives manual to find the solution of setting the LD_LIBRARY_PATH to the appropriate mysql library, and also hard coding this in /etc/ld.so.conf. LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/mysql/lib/ export LD_LIBRARY_PATH I did this and still get the linking problem. I verified that /usr/local/mysql/lib/ is accessible which it is, and it has the file libmysqlclient_r.a in there, which I assume is what is needed for libmysqlclient_r.so.15. Any further suggestions? Thanks, Jarrod -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] configuration weirdness
first, generic problem - --enable-client-only parameter for configure. as far as i understand, this switch is supposed to disable building of director sd. it's pretty unintuitive, as it would have been much better if all three components had individual switches with identical behaviour (each switch would enable that particular component). second - bacula 2.2.0. running configure with --enable-client-only shows at the end of the configuration : client-only:yes build-dird: yes build-stored: yes it seems that either the parameter is incorrectly named or not working ? -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bat advertisement
On 2007.08.08. 16:59, Kern Sibbald wrote: ... You can see an overview of the features at: http://www.bacula.org/dev-manual/GUI_Programs.html Is bat going to be considered stable for this release? I am sorry if I missed that. I consider it stable, but given that it is a *very* large body of brand new code, we can expect a number of bugs to crop up when it becomes heavily used under varying conditions. I don't expect any of the bugs to be a show stopper. I have used it exclusively for at least four months now. if there was a restrictd console that would only allow any access only to few hosts, would bat sensibly allow accessing director and, for example, restoring data ? or would it freak out because of being unable to access some information ? if that was possible, it would make some admins pretty happy as they could restore their server data with a clickthrough tool themselves :) John -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] backing up MySQL databases
On 2007.08.01. 14:37, Dimitrios wrote: Based on my research, i found out that backing up MySQL is quite un-efficient. On a small database, you can just dump the contents of MySQL into a file and then backup that file. On a large database (several Gigs) dumping to a file should be avoided, for example, on a hosting service the dump could fill your hosting space. In addition, its time consuming and allows for database badness where one table which is already backed up changes while the current one is still locked. This can be avoided if you dump the database with a system-wide lock but that means your web sites or application will be offline for the duration of the backup process (and since we are dumping whole gigs of data, it can be very time consuming). using innodb, it should be possible to dump all tables as a single transaction. Running a replicated database is not ideal for hosting servers, again that will double your hosting space and/or entire hosting service (colocation, etc). Based on my research, it seems the 'best' solution for really big databases is to use the Binary Log (mysql 4.1.3 or newer) and do incremental backups of the database. Thus, you do a full backup at first and then you only backup the Binary Log based on date ranges or snapshop points. Ofcourse this is process is badly documented and i couldn't find any scripts that can help me do this. binary logs also tend to eat up a LOT of space. other method might be feeding mysqldump output to pipe, where data is directly picked up by bacula client (i think, this was implemented somewhere around 2.0 bacula). has the advantage of using no additional diskspace on client, but can be tricky to correctly set up and restore. I'd appriciate your thoughts on this. Thank you. -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] backing up MySQL databases
On 2007.08.01. 14:51, Rich wrote: On 2007.08.01. 14:37, Dimitrios wrote: Based on my research, i found out that backing up MySQL is quite un-efficient. On a small database, you can just dump the contents of MySQL into a file and then backup that file. On a large database (several Gigs) dumping to a file should be avoided, for example, on a hosting service the dump could fill your hosting space. In addition, its time consuming and allows for database badness where one table which is already backed up changes while the current one is still locked. This can be avoided if you dump the database with a system-wide lock but that means your web sites or application will be offline for the duration of the backup process (and since we are dumping whole gigs of data, it can be very time consuming). using innodb, it should be possible to dump all tables as a single transaction. Running a replicated database is not ideal for hosting servers, again that will double your hosting space and/or entire hosting service (colocation, etc). Based on my research, it seems the 'best' solution for really big databases is to use the Binary Log (mysql 4.1.3 or newer) and do incremental backups of the database. Thus, you do a full backup at first and then you only backup the Binary Log based on date ranges or snapshop points. Ofcourse this is process is badly documented and i couldn't find any scripts that can help me do this. binary logs also tend to eat up a LOT of space. other method might be feeding mysqldump output to pipe, where data is directly picked up by bacula client (i think, this was implemented somewhere around 2.0 bacula). has the advantage of using no additional diskspace on client, but can be tricky to correctly set up and restore. oh, another method i forgot to mention (mostly because it would not be possible in most hosted environments :) ) you could lock all tables in mysql, flush all data, then create lvm snapshot. do this in runbeforejob. then, bacula could back up mysql files and remove snapshot in runafterjob. i still prefer dumps, they are more portable, easier to restore and compress better. I'd appriciate your thoughts on this. Thank you. -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] backing up MySQL databases
On 2007.08.01. 15:33, Dimitrios wrote: On Wed, 01 Aug 2007 15:20:50 +0300 Rich [EMAIL PROTECTED] wrote: i still prefer dumps, they are more portable, easier to restore and compress better. but what about incremental backups? as far as i know, you'd have to backup the entire database over and over again and for a very large database its something we'd like to avoid. yes, that would be problematic. maybe some hackish method of taking diff from two dumps would work, but that would require even more diskspace on the client. if you really need incremental backups, i don't know of any other method than using binlogs -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] backing up MySQL databases
On 2007.08.01. 15:31, Dimitrios wrote: On Wed, 01 Aug 2007 14:51:43 +0300 Rich [EMAIL PROTECTED] wrote: ... other method might be feeding mysqldump output to pipe, where data is directly picked up by bacula client (i think, this was implemented somewhere around 2.0 bacula). has the advantage of using no additional diskspace on client, but can be tricky to correctly set up and restore. wow, thats possible? its definetly interesting and i'll try googling... just look into http://www.bacula.org/rel-manual/FileSet_Resource.html - readfifo parameter :) -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] backing up MySQL databases
On 2007.08.01. 16:16, Drew Bentley wrote: ... if you really need incremental backups, i don't know of any other method than using binlogs -- Rich Yes, in order to do incrementals, you'll need binlogs enabled and you'll need to back these up. But also consider they take or can take a huge amount of space, so log rotating is necessary along with a possible performance hit as well, depending on the database demands, types of transactions, etc, going on. looking at bacula projects page, i noticed one item that might allow incremental backups of db dumps, if ever implemented : Item: 8 Directive/mode to backup only file changes, not entire file ... -drew -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] space issues
On 2007.08.01. 17:10, Robert LeBlanc wrote: Not without losing all your back-ups. With disk, it is best to set a volume limit so that it will create multiple back-up files. These will look like tapes and bacula will be able to prune and recycle these, freeing up disk space the size of the back-up file. You may be able to use bcopy to extract the back-up into another set of files, but I'm not sure and it would require more disk space. or use single file for each backup, as i am doing. makes it much easier to synchronise them, too. deletion would still have to be performed manually, at least for now - i hope project 5 gets some love :) Item: 5 Deletion of Disk-Based Bacula Volumes so if bacula was able to delete expired disk volumes, that would take some burden off admins (and responsibility, as currently it must be determined which is the first full volume before desired deletion date for each job) Robert LeBlanc College of Life Sciences Computer Support Brigham Young University [EMAIL PROTECTED] (801)422-1882 ... -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] volumes not reducing in size
On 2007.08.01. 17:22, [EMAIL PROTECTED] wrote: I have an ongoing problem, when my backup jobs fill my hard drive to 100%. Many times I have gone in and pruned jobs, and files, but the volume size stays the same, and will not run any new jobs. Is there a fix for this? this question is being asked quite often, and actually, another thread is going in parallel right now :) unfortunately, sf archive seems to be down since jul 27, so, if you are subscribed, check out thread space issues -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula Status + professional services initiative
On 2007.07.16. 19:40, Kern Sibbald wrote: Hello, ... 4. In the near future the Bacula project will no longer be providing binaries. They will be available for free to individuals, contributors, and charitable organizations through the professional web site. i hope i haven't missed an answer to my question as i am looking through several hundreds of messages on this list. would binary packages provided by contributors be discouraged within the project ? i would hope not, but would like to know for sure :) ... -- Rich - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Restore question
On 2007.07.12. 14:16, Frank Sweetser wrote: Ivan Adzhubey wrote: Hi, I have been using bacula for 5 years now and finally comes the first time I need to do a major restore from our tapes. I read the tutorial, tried the instructions and it seemingly worked as designed (I haven't run actual restore yet) but I am not sure which exactly files it was going to restore. I have two pools of tapes: one for full backups (run monthly) and another one for incremental (run every night). The 'restore all' command, when instructed to find JobIds of the most recent backup for a client, correctly located and listed all relevant full backups plus all incremental ones. It then proceeded creating directory tree and in the process went through both latest full backup and all incrementals since. So I assume this tree contains the latest snapshot? Or should I actually run two restores to get all the latest files: one for last full set and another one for all incrementals? I guess this is a stupid question but I'd really appreciate if someone could enlighten me since going through 50,000+ files manually to check the versions restored is really not an option. Bacula will handle making sure that only the latest version of each file is restored, so you only have to do it once. it probably is a good time to remind that files deleted/moved at some point inbetween backups also will be restored. -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] 2.0.3 agents with 1.36 backend
On 2007.07.11. 21:19, Rich wrote: is it possible to use 2.0.3 (or even 2.2) agents with 1.36 director/sd ? i can report that a test backup and restore were performed succesfully with 1.36.1 director/sd and 2.0.3 fd. -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] wx-console compiling error
wxGTK-2.8.4, bacula 2.0.3 what could be the reason that compiling wx-console fails ? is it still maintained, with the advance of bat (i'd guess qt4 won't be everywhere for a while :) ) Compiling wxbrestorepanel.cpp wxbrestorepanel.cpp: In constructor ‘wxbRestorePanel::wxbRestorePanel(wxWindow*)’: wxbrestorepanel.cpp:219: error: invalid use of undefined type ‘struct wxImageList’ /usr/local/include/wx-2.8/wx/generic/listctrl.h:16: error: forward declaration of ‘struct wxImageList’ wxbrestorepanel.cpp:220: error: invalid use of undefined type ‘struct wxImageList’ /usr/local/include/wx-2.8/wx/generic/listctrl.h:16: error: forward declaration of ‘struct wxImageList’ wxbrestorepanel.cpp:221: error: invalid use of undefined type ‘struct wxImageList’ /usr/local/include/wx-2.8/wx/generic/listctrl.h:16: error: forward declaration of ‘struct wxImageList’ wxbrestorepanel.cpp:222: error: invalid use of undefined type ‘struct wxImageList’ /usr/local/include/wx-2.8/wx/generic/listctrl.h:16: error: forward declaration of ‘struct wxImageList’ wxbrestorepanel.cpp:281: error: ‘wxIMAGE_LIST_SMALL’ was not declared in this scope wxbrestorepanel.cpp: In destructor ‘virtual wxbRestorePanel::~wxbRestorePanel()’: wxbrestorepanel.cpp:394: warning: possible problem detected in invocation of delete operator: wxbrestorepanel.cpp:394: warning: invalid use of undefined type ‘struct wxImageList’ /usr/local/include/wx-2.8/wx/generic/listctrl.h:16: warning: forward declaration of ‘struct wxImageList’ wxbrestorepanel.cpp:394: note: neither the destructor nor the class-specific operator delete will be called, even if they are declared when the class is defined. make[1]: *** [wxbrestorepanel.o] Error 1 -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] 2.0.3 agents with 1.36 backend
is it possible to use 2.0.3 (or even 2.2) agents with 1.36 director/sd ? this is intended only as a temporary solution until serverside is upgraded (1.36 agent fails to compile on a recent system) -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] start/stop client on demand
On 2007.07.02. 21:52, [EMAIL PROTECTED] wrote: Ah... Some clarification: by clients, I meant the client application, the bacula-fd agent rather then the machines themselves, though in general if you can use RunBeforeJob with a completely shut down system and have no problems, doing so with just a downed agent should be no problem as well. is there any reason to keep bacula-fd shut down ? is it taking too much resources, or is this because of security reasons ? the client probably could be additionally firewalled, if this is because of the security. as for resource consumption, i don't think bacula-fd resource usage is noticeable. the only reason left is psychological wish to have it shut down, but that's another issue ;) Thanks, Andrew ... -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Can I use a remote network storage device?
On 2007.07.02. 18:56, nellyginrich wrote: Is it possible to use a remote network storage device as the backup medium with Bacula? My understanding from the documentation and the questions posed here, is that it isn't possible - because each storage device needs to have an attached machine on which the storage daemon is running. Am I right? I have a remote linux (Fedora Core 6) server running Bacula (which works nicely with the test configuration to a file in /tmp), but I want the backups done to my local network here. The Windows XP and 2000 pcs here are rather ancient, and not always on, so I am considering purchasing a network storage device (NSD), but cannot see how Bacula's daemon's running on the remote server will be able to contact this device. as has been already mentioned, one of the best solutions in yuor case would be a nas device that supports nfs. you mount this device on the machine where sd resides, and back things up to it. this will make clients connect to sd, and it will manage any further connections to nas. make sure to check out wsize/rsize/tcp/udp and other nfs mount parameters - these can significantly impact nfs mount performance and reliability. if you are worried about possible nas/network problems, it might also be a good idea to but some scripts in runbefore directive, that would check out network connections and bring them up, if necessary. ... -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] bacula 1.38.11 or bacula 2.0.3 error
are you sure other jobs, jobdefs and whatnot you have before this entry
in the file are correct ?
maybe you should try to strip the file to absolute minimum, and, if that
does not work, show the complete configuration.
On 2007.06.28. 09:48, Omer Faruk Sen wrote:
Ok then.
Here is the lines that are not working:
# List of files to be backed up
FileSet {
Name = Full Set
Include {
Options {
signature = MD5
}
#
# Put your list of files here, preceded by 'File =', one per line
#or include an external list with:
#
#File = file-name
#
# Note: / backs up everything on the root partition.
#if you have other partitons such as /usr or /home
#you will probably want to add them too.
#
# By default this is defined to point to the Bacula build
#directory to give a reasonable FileSet to backup to
#disk storage during initial testing.
#
File = /home/fs/rpmbuild/BUILD/bacula-1.38.11
}
#
# If you backup the root directory, the following two excluded
# files can be useful
#
Exclude {
File = /proc
File = /tmp
File = /.journal
File = /.fsck
}
}
And the error message:
Starting Bacula Director services: 28-Haz 09:55 bacula-dir: ERROR
TERMINATION at parse_conf.c:853
Config error: Keyword Include not permitted in this resource.
Perhaps you left the trailing brace off of the previous resource.
: line 76, col 10 of file /etc/bacula/bacula-dir.conf
Include {
It says I can't put Options{} in FileSet{} section
...
--
Rich
-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] The thing with restoring Moved and Deleted files
On 2007.06.25. 23:04, Ryan Novosielski wrote: ... Imagine your Sitation: The whole server goes down. You have a 10 Days old full backup and 9 incrementals. Now you bring the system back up but with inconsitent files and contents?! Try to explain that to you boss or the company you work for?! Since you seem to be so sure about the subject, I'd like to see you tell me which backup software does and does not have this problem. My suspicion is that many of the commercial packages have the same problem (at least that's what I remember finding out last time I looked into this). I can't tell right now from looking at the web, as this really isn't mentioned anyplace. How would you find out whether your choice of software package has this limitation or not? tivoli has an ability to keep a specified changed versions of files that are still on the disk and a specified amount of versions for files that have been deleted (that is, if you overwrite a file 3 times, only two latest revisions are kept. if you delete it, only single last revision is kept. or the other way around). this also includes the ability to keep particular files for a specified amount of days. together this allows for a pretty flexible and hdd space efficient solution. ... -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula won't start, please help
try running commands directly with console output, for example : bacula-dir -f if that produces nothing, try adding -dnumber and increase number until you get something useful. a wild guess : it's possible your config file permissions are incorrect :) On 2007.06.15. 17:41, ChowYunFett wrote: Hi, looking for a good backup solution that can handle multiple machines, I installed Bacula 2.3.0 with MySQL on openSUSE, every step according to the user guide. When I run ./bacula start I get no error messages. I found I couldn't connect to the Console daemon by ./bconsole so I checked the status by ./bacula status which gave me bacula-sd is stopped bacula-fd is stopped bacula-dir is stopped I am a linux noob and even more so with bacula so I don't know where to start looking for what causes the trouble. Any hints are very much appreciated. Thanks, Chow -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Has this list been having problems sending mail?
On 2007.06.15. 18:15, Russell Howe wrote: ... It could also be the original poster gets fed up waiting for me to find time to run through the queue, and so subscribes and then reposts. Then when I go through the queue, I release the original message and you get a dup. Sorry about that - I'm struggling to find the time to even go through the queue (nor can I even find the time to maintain my bacula install!) with delays approaching 4 and more days, would it make sense to disallow non-subscribed posting ? ... -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula and my MYSQL backup
On 2007.06.12. 11:27, Mostafa Itani wrote: Dear All, I am running bacula version “2.0.2” on Linux Redhat enterprise 4. The server is running moodle application. Basically a web application with MYSQL database. I want to make sure that the back up of MYSQL is done even when the database is being accessed. Meaning when users are modifying the database, and the backup is running at the same time, I just want to make sure about the result, and how bacula will handle this. As I read it takes a shadow or snapshot, backs it up and then delete the image, I just want a confirmation message, that the database will not be skipped during the back up. there are several ways to accomplish this. 1. if all (or most) of your tables are innodb, the best probably is to use mysqldump and dump databases in a single transaction. a) you can dump the data to files, which later get backed up by bacula; b) you can take db backup through a pipe 2. you can also place your db on lvm, and backup script can lock all tables for writing, take a volume snapshot, unlock tables and then back up the snapshot. could be less reliable for complex systems. Best regards, Mostafa Itani System Administrator American University of Beirut College Hall B2 P.O. Box 11-0236 Beirut, Lebanon Tel: +961 1 35 - Ext: 3518 Cell: +961 3 811972 Web: www.aub.edu.lb http://www.aub.edu.lb -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] store data on two storages at one time
On 2007.06.08. 11:15, Maik Derstappen wrote: ... hello Ryan, thank for your answer. I the 2 Storages are 2 Servers which stores the data on harddisks. Just i need an solution to store the same data on more than one storage. And i don't want run 2 job for this. Because the filedaemon then read the data twice for the same Backup. if i understood you correctly, you have two servers with hdd storage and would like to replicate backups on both without running them twice. which is also what i have :) this would be ideally handled by a clustered bacula servers (a feature request i made recently) - but that probably won't appear anytime soon. until then, your best bet probably would be to rsync changes between the servers. Any Idears for an nice solution? regards Maik ... -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] RunBeforeJob / RunRightBeforeWrite question
On 2007.06.08. 16:32, Steve Campbell wrote: I've been following the thread about waiting 40 minutes, and the suggestion to perform a RunBeforeJob seems to be the current solution. I have been running Mondo as a disaster recovery solution. It has a function that comes in handy here whereas it will perform a task just before writing it's ISO images. I am guessing that the RunBeforeJob in Bacula runs before even deciding what files it will back up. As I write to a Samba mount which need awaken before writing to it, and if Bacula takes a little bit of time to decide on and transfer files, is there a way to nudge the mount just before any writes to it? I realize this is i can't access archives right now, sf server gives me error 500... unless you found a better solution, quick and dirty hack would be a script which would be run before the job and would ls, touch or whatever you need every 10 or so seconds something on the mount. not the same situation, but I could buffer the writes to disk before writing to the mount. (I seem to recall there is this buffering option in Bacula). Thanks for any advice. Blackbeard -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula Status
On 2007.06.04. 15:46, Kern Sibbald wrote: i know this is a bad thing to ask... but are there any estimates when the next stable version could be expected ? i've been putting off touching a running system, so maybe i can drag it a bit longer and upgrade to the next stable (still running 1.36 ;) ) ... -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] disk array suggestions?
On 2007.06.05. 21:00, mikee wrote: I have approval to move my current bacula from one box to another, new box. The new box I need to purchase disk for the on-disk backups before the data reackes tape. I'm thinking about in the range of 2TB of RAID5. note that if you are interested in write performance (as you might be :) ), it might actually be cheaper... to go with raid10. it's all your decision, and i won't rehash arguments here - but http://www.baarf.com/ has some arguments :) Does anyone have any thoughts or recommendations for specific devices? have you already decided for hardware or software raid ? lately i'm mostly leaning towards software raid. some of the benefits are generic maintenance, individual disk access for smart monitoring (which also is supported by some hw raid controllers and latest smartmontools), much easier and safer recovery in case box goes down (you don't need another identical hw raid controller) - and there probably are more :) cpu overhead with raid5 for a backup machine with a modern cpu would not be significant (actually, my backup machines mostly have idle cpus), and probably non-existent with raid1[0] i was reading a page some time ago which summed up linux software raid benefits and drawbacks quite extensively (i think one of the developers wrote that), but i can't find it right now :) Mike -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] FW: Backup of a mysql dump of the database
On 2007.05.23. 11:53, Marc wrote:
Hi,
I've searched the archives, but I can not find any relevant information.
Therefor my question: is it possible to do a database by database dump and
backup these dump files? Because of database sizes, it would be very nice if
this can be done database by database. I mean first do the dump of database
1, move it to bacula, remove the dump, dump of database 2, move it to
bacula, etc...
Can this be done?
i didn't see whether you successfully resolved your problem, so here's a
quick crude script, which backs up each database as a separate file,
and does this for all of the databases.
script does not protect from simultaneous runs, does not compress dumps
(as that is done by bacula job) - though all this is very easy to add.
for bacula job, just add something like ;
ClientRunBeforeJob = /scripts/mysqlbackup create
ClientRunAfterJob = /scripts/mysqlbackup remove
note, the script probably has several problems, so feel free to correct
those ;)
-
#!/bin/bash
HOME=/root
DUMPDIR=/var/tmp/database_dump
MYSQLDUMP=/usr/local/mysql/bin/mysqldump
DUMPCOMMAND=$MYSQLDUMP --add-drop-database --add-drop-table --add-locks
--extended-insert\
--single-transaction --quick
fail() {
echo failure : $1
exit 1
}
create() {
if [ ! -d $DUMPDIR ]; then
mkdir -p $DUMPDIR || fail unable to create directory $DUMPDIR
fi
for i in `echo show databases | mysql -N`; do
$DUMPCOMMAND $i $DUMPDIR/$i || fail unable to dump
database $i
done
}
remove() {
rm $DUMPDIR/* || fail unable to remove db dumps
}
case $1 in
create)
create
;;
remove)
remove
;;
*)
fail pass either create or remove
esac
Kind regards,
Marc
--
Rich
-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Problem to run a first backup
On 2007.06.05. 15:48, Alexandre Bunn wrote:
Good morning
I´m with a problem with bacula configuration. My logs follow below
--- logs ---
05-Jun 09:33 bacula-dir: No prior Full backup Job record found.
05-Jun 09:33 bacula-dir: No prior or suitable Full backup found in
catalog. Doing FULL backup.
05-Jun 09:33 bacula-dir: Start Backup JobId 19,
Job=Birao.2007-06-05_09.33.40
05-Jun 09:33 bacula-dir: Birao.2007-06-05_09.33.40 Error: dev.c:126
Unable to stat device /backups: ERR=No such file or directory
05-Jun 09:33 bacula-dir: Birao.2007-06-05_09.33.40 Warning:
Device FileStorage requested by DIR could not be opened or does
not exist.
05-Jun 09:33 bacula-dir: Birao.2007-06-05_09.33.40 Error: dev.c:126
Unable to stat device /backups: ERR=No such file or directory
05-Jun 09:33 bacula-dir: Birao.2007-06-05_09.33.40 Warning:
Device FileStorage requested by DIR could not be opened or does
not exist.
05-Jun 09:33 bacula-dir: Birao.2007-06-05_09.33.40 Error: dev.c:126
Unable to stat device /backups: ERR=No such file or directory
05-Jun 09:33 bacula-dir: Birao.2007-06-05_09.33.40 Warning:
Device FileStorage requested by DIR could not be opened or does
not exist.
05-Jun 09:33 bacula-dir: Failed command: Jmsg
Job=Birao.2007-06-05_09.33.40 type=5 level=1181046823 bacula-dir:
Birao.2007-06-05_09.
33.40 Warning:
Device FileStorage requested by DIR could not be opened or does
not exist.
05-Jun 09:33 bacula-dir: Birao.2007-06-05_09.33.40 Fatal error:
Device FileStorage with MediaType File requested by DIR not
found in SD Device resources.
05-Jun 09:33 bacula-dir: Birao.2007-06-05_09.33.40 Fatal error:
Storage daemon didn't accept Device FileStorage because:
3924 Device FileStorage not in SD Device resources.
logs
See below a part of my bacula-sd.conf
Device {
Name = FileStorage
Media Type = File
Archive Device = /backups
LabelMedia = no; # lets Bacula label unlabeled media
Random Access = Yes;
AutomaticMount = yes; # when device opened, read it
RemovableMedia = no;
AlwaysOpen = yes;
}
do you have a directory named /backups that storage daemon has write
access to ?
also, unless you are planning to use some pre-made volume files only,
you probably should use LabelMedia = yes;
i also have AlwaysOpen = no; in my configuration, but i'm too lazy to
further read documentation right now ;)
My doubt is:
In definition FileStorage the Archive Device the /backups has that to be
a archive or a directory and what your permition?
Att,
Alexandre
--
Rich
-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Backup-Proxy-Solution needed
On 2007.05.28. 10:55, Stefan Wintermeyer wrote: Hi, we'd like to use Bacula as our one and only backup solution. But we have a problem with all home-office-installations. They normaly work with an asynchron DSL connection. It would take days to backup to our normal backup server in the internet. And normaly they only have a couple of hours during the night. Having a bright moment I thought: hmmm... why don't we install some sort of proxy at the home offices? People could backup their stuff to the proxy and the proxy syncs with the central server. Time wouldn't be such a big issue (even if it takes a couple of days to sync). well, maybe clustered bacula servers might be helpful here, if those ever get implemented ;) currently your best bet probably is to use rsync at some point, which would work nicely for syncing. Has anybody a solution for that problem? Stefan ... -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] voting (feature request : clustered server)
On 2007.05.25. 23:13, Arno Lehmann wrote: ... as for the method, what about using a wiki page for feature requests ? that way permanent links to versions could be used for voting results, but head could be used to add new ones, fix typos and so on. Yeah... well... difficult question. If you know a wiki system where you can do votes the way we always did them (or have good arguments for changing the voting process), I'm very open for suggestions :-) i was unable to subscribe to sourceforge lists for some time, but if i understand correctly, voting is done by sending a mail to voting address, right ? i would guess then a script calculates all the votes, and the result is put up on the webpage. making the result go in wiki (which would allow for easy updating), then refer to particular revision for voting, and just put up results in either a separate, or the same wiki page - that probably would not change current voting process much. all other suggested possibilities indeed seem to be a bit of an overkill to me (unless any of them gets implemented for other reasons, like bugzilla or something). ... -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] [Bacula-devel] feature request : clustered server
On 2007.05.24. 01:05, Arno Lehmann wrote: Hi, On 5/23/2007 3:12 PM, Kern Sibbald wrote: Hello, Thanks for your feature request. Unfortunately at the moment no one is collecting these, and so you will probably need to resubmit it later. Not exactly... I'm still thinking about how to do this properly, but in the meantime, I hope I won't forget any new feature requests. kern, arno, thanks for the responses :) i understand that clustered servers isn't exactly a basic feature or one to easy implement, but i hope that this still qualifies for a feature request, at least as #41 :) as for the method, what about using a wiki page for feature requests ? that way permanent links to versions could be used for voting results, but head could be used to add new ones, fix typos and so on. I noticed this one, at least :-) Arno -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] automatically removing disk based volumes
oops. resending to the list. On 2007.05.17. 23:56, Martin Simmons wrote: On Thu, 17 May 2007 13:32:54 +0300, Rich said: can bacula be configured to automatically remove disk based volumes, once the volume contains no data (after pruning db records) ? No, but it is feature request item 7. glad to hear that :) OTOH, it can't be configured to burn your old tapes either :-) i am keeping single volume per job, but date-based removal outside of bacula is cumbersome - bacula is the only one who has exact information on what has to be kept. for example, if i am taking a full backup once a month, this leaves some 30 or so day window when i can't safely remove volume files, as incremental ones without previous full aren't very useful. The usual way to configure Bacula it is to make it recycle (reuse) the existing volumes, but that is tricky if you must have control over exactly which volume is used for each job. yes, but currently i prefer to use single uniquely named volume/file for each job - this way it is easier to sync only changes to remote storage, and it also makes it easier to find the correct files in case of a real emergency. so, waiting for it to be implemented :) __Martin -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] clustered bacula servers
On 2007.05.17. 23:52, Florian Heigl wrote: Hi, 2007/5/17, Rich [EMAIL PROTECTED]: is there a functionality (or planned functionality) in bacula to allow two or more server instances work in a way similar to a cluster ? no, and maybe that would probably include two separate instances of director, storage daemon and database that would be able to share/sync information about jobs and job data. that's likewise an active-active cluster then, or at least if one gets that far there's not much reason to stop short of it. but, this is a feature not even most of the highend backup utilities i know easily provide. i haven't verified this, but i was told tivoli does this ;) file daemons would probably have two director access configured, with a way to set a preferred one, which would always contact particular client, except in cases when the preferred one is down. with an older version of bacula i have now, failed first server would require manual reconfiguring of all clients, and it also requires syncing data separately from bacula. in case of geographically distributed servers there also is no way to tell backup jobs to go to a particular server first, so there's increased network traffic as well. i've searched documentation and mailing list archives, but best i could find was about clustered services backup. if you search the archives you might be able to find an old mail inquiring about this; as far as i remember kern stated that at the moment this would require addition of a lot of table locking which isn't there. the database syncing itself is less of an issue, but the state information can't be shared at the moment. I'm taking a stake on using a cluster-aware filesystem beneath the bacula daemons for easy failover, but this is not able to deal with 'Geocluster'-Like scenarios either. quite useful feature also would be the ability to set preferred server for cases when services are geographically distributed alongside backup servers. i'll try to submit a correct feature request as martin suggested (even though this probably is worst time to do so :) ) Florian -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] feature request : clustered server
this probably is the worst time to submit a request, but given it's scope it probably won't have huge impact :) i am not currently subscribed to -devel - please, cc me or -users (or force me to subscribe, if the discussion would be limited to that list). ps. page http://www.bacula.org/?page=vote doesn't seem to have the latest results; page http://www.bacula.org/?page=projects has item description Implement a Bacula console, and management tools probably using Qt3 and C++. - from my reading of list archives, it seems qt4 is used ? the projects page also has some typos, that i would have fixed if that was in a wiki :) - Item 41: Clustered bacula server, including director, storage daemon and database Origin: Rihards, [EMAIL PROTECTED] Date: 18 May 2007 Status: What: It would be nice to run two (or more) separate backup servers that would synchronise backups and be able to take over each others clients in case any server dies. Clustered servers would each run a director, storage daemon and each would have a separate database. Why:Disk based backup solutions have become very popular, but proper offsite backups and distributed servers are cumbersome and requires too much of hacking and manual interventions. Currently, failed primary server would require manual reconfiguring of all clients (setting identical director names welcomes mistakes later on), and it also requires syncing data separately from bacula. Configuring and restoring from a secondary system is far from straightforward. In case of geographically distributed servers there also is no way to tell backup jobs to go to a particular server first, so there's increased network traffic as well. For example, if organisation has two locations that both have services to backup, each site could have a separate backup server with simple failover to the remote server. Notes: * File daemons would have access for all servers in the cluster; * It should be possible to set preferred server for each client; * How should sharing of the configuration be implemented ? Would it be enough to configure one server and get the changes propagated to all of them ? * Would it make sense to back up to one server only ? For example, some jobs could be configured to back up, but not sync to other servers. In case of the primary server failure secondary server would take the backup, but move it to the primary when it comes back. -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] clustered bacula servers
is there a functionality (or planned functionality) in bacula to allow two or more server instances work in a way similar to a cluster ? that would probably include two separate instances of director, storage daemon and database that would be able to share/sync information about jobs and job data. file daemons would probably have two director access configured, with a way to set a preferred one, which would always contact particular client, except in cases when the preferred one is down. with an older version of bacula i have now, failed first server would require manual reconfiguring of all clients, and it also requires syncing data separately from bacula. in case of geographically distributed servers there also is no way to tell backup jobs to go to a particular server first, so there's increased network traffic as well. i've searched documentation and mailing list archives, but best i could find was about clustered services backup. -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] automatically removing disk based volumes
can bacula be configured to automatically remove disk based volumes, once the volume contains no data (after pruning db records) ? i am keeping single volume per job, but date-based removal outside of bacula is cumbersome - bacula is the only one who has exact information on what has to be kept. for example, if i am taking a full backup once a month, this leaves some 30 or so day window when i can't safely remove volume files, as incremental ones without previous full aren't very useful. -- Rich - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: Compilation errors on Macintosh (was [Bacula-users] Bacula FD on Macintosh)
I really can't say enough good things about Darwin Ports. I had my bacula daemon up and running in under an hour, and most of that time was spent compiling it. http://bacula.darwinports.com/ has the basics on getting it and installing it. I'm trying to find the site I used that had almost step-by-step instructions on getting it going. Bill Moran wrote: On Sat, 8 Apr 2006 21:22:26 -0500 Erich Prinz [EMAIL PROTECTED] wrote: Once I figured out the appropriate configure options, it was a snap. ./configure \ --enable-smartalloc \ --sbindir=/usr/sbin \ --with-pid-dir=/var/run \ --with-subsys-dir=/var/run/subsys \ --with-mysql=/usr/local/mysql \ --with-working-dir=/etc/bacula/working \ --with-dir-password= \ --with-sd-password= \ --with-fd-password= \ [EMAIL PROTECTED] \ --with-smtp-host=server.domain.com \ --with-baseport=9101 Well. I'm having some problems and I hope you can help me. configure succeeds, although I used a slightly different set of options: ./configure \ --enable-smartalloc \ --sbindir=/usr/sbin \ --with-pid-dir=/var/run \ --with-subsys-dir=/var/run/subsys \ --with-working-dir=/etc/bacula/working \ --enable-client-only \ --with-dir-password= \ --with-sd-password= \ --with-fd-password= \ --with-job-email=[snip] \ --with-smtp-host=[snip] \ --with-baseport=9101 This machine has no database on it whatsoever, so I'd prefer not to bother with anything but the client (i.e. fd). DT is installed, and I created the directories you mentioned, but I get the following output during make: [no errors to this point] Make of lib is good ==Entering directory /Users/bill/bacula-1.38.7/src/findlib /usr/bin/g++ -c-I. -I.. -g -O2 -Wall find.c /usr/bin/g++ -c-I. -I.. -g -O2 -Wall match.c /usr/bin/g++ -c-I. -I.. -g -O2 -Wall find_one.c /usr/bin/g++ -c-I. -I.. -g -O2 -Wall attribs.c attribs.c: In function `bool set_attributes(JCR*, ATTR*, BFILE*)': attribs.c:351: error: `lchown' undeclared (first use this function) attribs.c:351: error: (Each undeclared identifier is reported only once for each function it appears in.) make[1]: *** [attribs.o] Error 1 == Error in /Users/bill/bacula-1.38.7/src/findlib == ==Entering directory /Users/bill/bacula-1.38.7/src/filed /usr/bin/g++ -c -I. -I.. -g -O2 -Wall filed.c /usr/bin/g++ -c -I. -I.. -g -O2 -Wall authenticate.c /usr/bin/g++ -c -I. -I.. -g -O2 -Wall acl.c acl.c:164:21: sys/acl.h: No such file or directory acl.c: In function `int bacl_get(JCR*, int)': acl.c:188: error: `acl_t' undeclared (first use this function) acl.c:188: error: (Each undeclared identifier is reported only once for each function it appears in.) acl.c:188: error: parse error before `;' token acl.c:190: error: `acl_type_t' undeclared (first use this function) acl.c:190: error: parse error before `;' token acl.c:193: error: `ostype' undeclared (first use this function) acl.c:193: error: `ACL_TYPE_DEFAULT' undeclared (first use this function) acl.c:193: error: `ACL_TYPE_ACCESS' undeclared (first use this function) acl.c:195: error: `acl' undeclared (first use this function) acl.c:195: error: `acl_get_file' undeclared (first use this function) acl.c:197: error: `acl_to_text' undeclared (first use this function) acl.c:199: error: `acl_free' undeclared (first use this function) acl.c: In function `int bacl_set(JCR*, int)': acl.c:217: error: parse error before `;' token acl.c:224: error: `acl_delete_def_file' undeclared (first use this function) acl.c:230: error: `acl_from_text' undeclared (first use this function) acl.c:240: error: `acl_valid' undeclared (first use this function) acl.c:246: error: `acl_set_file' undeclared (first use this function) make[1]: *** [acl.o] Error 1 == Error in /Users/bill/bacula-1.38.7/src/filed == ==Entering directory /Users/bill/bacula-1.38.7/src/console /usr/bin/g++ -c-I. -I.. -g -O2 -Wall console.c /usr/bin/g++ -c-I. -I.. -g -O2 -Wall console_conf.c /usr/bin/g++ -c-I. -I.. -g -O2 -Wall authenticate.c /usr/bin/g++ -c-I. -I.. -g -O2 -Wall conio.c /usr/bin/g++ -O -L../lib -L../cats -o bconsole console.o console_conf.o authent icate.o conio.o \ -ltermcap -lbac -lm -lpthread -lintl \ Make of console is good Does anyone have any suggestions on how to get the fd to compile? This is on Mac OS 10.4.6 -- Dan Rich [EMAIL PROTECTED] | http://www.employees.org/~drich/ | "Step up to red alert!" "Are you sure, sir? | It means changing the bulb in the sign..." | - Red Dwarf (BBC) signature.asc Description: OpenPGP digital signature
Re: [Bacula-users] Bacula FD on Macintosh
Bill Moran wrote: I've been deploying Bacula here for a few weeks now. All servers are being backed up from three locations. Very nice. Now it's time to start on the mobile machines. The one that has me concerned is the one Macintosh laptop that we have. Anyone running a Bacula FD on a Macintosh? Is it stable/reliable? Of course, couple that with the challenge of backing up a laptop and the whole thing is likely to be challenging. I'm interested to hear any an all stories and follow any and all links regarding using Bacula FD on a Mac, and/or laptop. Google was not terribly helpful, but I may not have been crafting the best search strings. TIA for any information. I run it on my Mac (laptop) at home and have run into a couple of issues. First of all, don't use the bacula-fd that comes from Fink. My experience was that it would hang almost nightly and cause any backups from then on to fail until I restarted it. I've had good luck with the version in Darwin Ports, it's been running for about a week now. The only issue there was my WiFi connection going away in the middle of the night. I'm not sure I can blame that on bacula (yet :) ), so it may just be a coincidence. The Darwin Ports version also comes with a launchd config file so you can have it start automatically on boot. -- Dan Rich [EMAIL PROTECTED] | http://www.employees.org/~drich/ | Step up to red alert! Are you sure, sir? | It means changing the bulb in the sign... | - Red Dwarf (BBC) --- 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 coding territory! http://sel.as-us.falkag.net/sel?cmd=lnkkid=110944bid=241720dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Database corruption problems
I've run into an odd problem where my database has become corrupted. I've decided that this is a good time to migrate my database from sqlite to mysql, but have a problem. Since the old database is corrupted sqlite3 won't read anything from it. Does anyone have tools to scan the existing volumes and rebuild the database? I can probably build some scripts to do part of the job, but I'm not sure what all of the table interdependencies are, so it's a bit risky. -- Dan Rich [EMAIL PROTECTED] | http://www.employees.org/~drich/ | Step up to red alert! Are you sure, sir? | It means changing the bulb in the sign... | - Red Dwarf (BBC) signature.asc Description: OpenPGP digital signature
[Bacula-users] Database corruption problems
I've run into an odd problem where my database has become corrupted. I've decided that this is a good time to migrate my database from sqlite to mysql, but have a problem. Since the old database is corrupted sqlite3 won't read anything from it. Does anyone have tools to scan the existing volumes and rebuild the database? I can probably build some scripts to do part of the job, but I'm not sure what all of the table interdependencies are, so it's a bit risky. -- Dan Rich [EMAIL PROTECTED] | http://www.employees.org/~drich/ | Step up to red alert! Are you sure, sir? | It means changing the bulb in the sign... | - Red Dwarf (BBC) signature.asc Description: OpenPGP digital signature
[Bacula-users] Database corruption problems
I've run into an odd problem where my database has become corrupted. I've decided that this is a good time to migrate my database from sqlite to mysql, but have a problem. Since the old database is corrupted sqlite3 won't read anything from it. Does anyone have tools to scan the existing volumes and rebuild the database? I can probably build some scripts to do part of the job, but I'm not sure what all of the table interdependencies are, so it's a bit risky. -- Dan Rich [EMAIL PROTECTED] | http://www.employees.org/~drich/ | Step up to red alert! Are you sure, sir? | It means changing the bulb in the sign... | - Red Dwarf (BBC) signature.asc Description: OpenPGP digital signature
Re: [Bacula-users] Database corruption problems
Dan Rich wrote: I've run into an odd problem where my database has become corrupted. I've decided that this is a good time to migrate my database from sqlite to mysql, but have a problem. Since the old database is corrupted sqlite3 won't read anything from it. Does anyone have tools to scan the existing volumes and rebuild the database? I can probably build some scripts to do part of the job, but I'm not sure what all of the table interdependencies are, so it's a bit risky. (Sorry about the multiple messages - my mail server refused the mail and then apparently sent it anyway). -- Dan Rich [EMAIL PROTECTED] | http://www.employees.org/~drich/ | Step up to red alert! Are you sure, sir? | It means changing the bulb in the sign... | - Red Dwarf (BBC) signature.asc Description: OpenPGP digital signature
