[Bacula-users] Bacula status update + you can help
Hello, After having been totally frustrated chasing this kernel crash for the last few weeks (I really could not believe that it was not a Bacula bug), I have finally found a work around and at the same time, proven that it is a SuSE problem. Last night, I loaded the Fedora FC4 kernel from my server onto my SuSE machine, and rebooted, not quite expecting it to boot up. Well, it did and with the exception of AppArmor, which is SuSE specific, and mtx, which seems to have been modified by SuSE, everything works -- i.e. a SuSE 10.1 system with a Fedora kernel. (Note, even though I have switched from Fedora to SuSE because of too many bugs on the Fedora releases, I have never experienced a Fedora kernel bug -- probably due to the presense of Alan Cox!). Results of Bacula tests: with the *identical* binaries, everything works fine. Running my tape regression script (minus the regressions needing mtx), no crash. Conclusion: SuSE has seriously broken their SCSI driver. What you can do: I don't ususally ask Bacula users to do something, but this is one time when you can potentially be a big help to the project, and this is especially important for SuSE users (in my opinion): please go to the following link: https://bugzilla.novell.com/show_bug.cgi?id=208782 and if you don't have an account with their buzilla, please create one, then add your name to the CC list for this bug, and cast a vote for this bug (Arno gave me the idea). If we can get a good number of people interested in this bug, I think Novell will stand up and take notice, and possibly improve its kernel support. My purpose is in a polite way to show them that this bug is important so that they put the appropriate resources on it. Best regards, Kern - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Moving Files and updating catalog
Dear listusers, we are currently backing up 5 servers as a pilot before final bacula-installation with storage unit and bg server ;-) We configured expiration and pruning to suit our needs and everything's fine. What we are experiencing is, that we need to insert another harddrive to handle the upcoming data. So, what I am planning to do and where I need some advice: Plugin another harddrive, partition and format it and mount it etc. Then I wanna tell bacula in his config, to backup 2 servers to this new partition. I need to configure another StorageDevice which uses this new partition. Then I want to move the affected container files, which we're already written, from old to new partition. Then I need to tell bacula, that these container files moved, and that is my main question: How do I do that? Stop bacula and then I hope that there's a command that tells bacula, that these files are now on another place in the file system and that bacula doesn't look for them in the old but in the new partition from there on (should be important for restore ;-) ). Any help is appreciated. Greetings, Benni -- Benjamin Zeller Ing.-Büro Hohmann Bahnhofstr. 34 D-82515 Wolfratshausen Tel.: +49 (0)8171 347 88 12 Mobil: +49 (0)160 99 11 55 23 Fax: +49 (0)8171 910 778 mailto: [EMAIL PROTECTED] www.ibh-wor.de pgp7h5nyyej9q.pgp Description: PGP signature - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] problem after upgrade from 1.38.9 to 1.38.11
Hi guys, thanks your help, I've been added SD_USER=bacula SD_GROUP=tape in bacula-ctl-sd and everything it's fine Jea - Original Message - From: Martin Simmons [EMAIL PROTECTED] To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED]; bacula-users@lists.sourceforge.net Sent: Thursday, September 28, 2006 8:41 PM Subject: Re: [Bacula-users] problem after upgrade from 1.38.9 to 1.38.11 On Thu, 28 Sep 2006 13:12:26 +0200 (CEST), lordjea said: yes, and the bacula user belong to tape group too [EMAIL PROTECTED]:/usr/local/bacula/var/bacula$ ls -l /dev/nst0 crw-rw 1 root tape 9, 128 2005-02-26 07:38 /dev/nst0 [EMAIL PROTECTED]:/usr/local/bacula/var/bacula$ id uid=3D1001(bacula) gid=3D1001(bacula) grupos=3D1001(bacula),26(tape) [EMAIL PROTECTED]:/usr/local/bacula/var/bacula$ Bacula 1.38.11 ignores the user's groups, so you need to pass -g tape when starting bacula-sd. This is the same in 1.38.9, so maybe you changed something? __Martin __ LLama Gratis a cualquier PC del Mundo. Llamadas a fijos y móviles desde 1 céntimo por minuto. http://es.voice.yahoo.com - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Moving Files and updating catalog
Hi, On 9/29/2006 11:19 AM, Benjamin Zeller wrote: Dear listusers, we are currently backing up 5 servers as a pilot before final bacula-installation with storage unit and bg server ;-) We configured expiration and pruning to suit our needs and everything's fine. What we are experiencing is, that we need to insert another harddrive to handle the upcoming data. So, what I am planning to do and where I need some advice: Plugin another harddrive, partition and format it and mount it etc. Then I wanna tell bacula in his config, to backup 2 servers to this new partition. I need to configure another StorageDevice which uses this new partition. Then I want to move the affected container files, which we're already written, from old to new partition. Then I need to tell bacula, that these container files moved, and that is my main question: How do I do that? Stop bacula and then I hope that there's a command that tells bacula, that these files are now on another place in the file system and that bacula doesn't look for them in the old but in the new partition from there on (should be important for restore ;-) ). Hmm. I don't think you need to shutdown Bacula for that. Rather I'd suggest the following: First, make sure you have a good understanding of the reationship between MediaType of the volume files and the storage devices. Depending on your version of Bacula you might find you're not able to easily restore from from any volume using any storage device. Then, after moving the volume files, simply change their media types. I'd do this using a SQL query, either through bconsole or using your database front-end. Also, change the volumes pool association. Any help is appreciated. What I'd _really_ do, though, is thinking about your backend storage setup. I wouldn't consider storing backups on single hard drives a good idea. Rather use a standalone disk system where you can add drives to a RAID array on the fly. Start wih a RAID5 with 3 disks, and make sure you know how to add disks later to increase the overall capacity. The same result can be achieved by using LVM/dm in linux, but I'd prefer a stand-alone disk subsystem for better maintenance. Arno Greetings, Benni - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users -- IT-Service Lehmann[EMAIL PROTECTED] Arno Lehmann http://www.its-lehmann.de - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] what's the best way to do backup for a logical volume (LVM)
Hi gentlemen, I tried before to do things that could not be the best way to
backup my logical volume.
The problem:
I have many logical volumes from machines that could not be sttoped, then I
had two idea
The alternatives:
1- to do a backup for whole RAW partition making a snapshot
(lvcreate --snapshot -L 1Gb --name backup-root vg-uml-008/root)
2- to do a snapshot, then mount the snapshot and make a backup from the
mounted point (to do this I'm trying to use a RunScript in Job resource**)
3-somebody have any idea reliable?
I submit the question because you have a lot of experience that 3 weeks that
I have
** but have a mistake:
29-sep 11:59 doom-dir: ERROR in parse_conf.c:853 Config error: Keyword
RunScript not permitted in this resource.
Perhaps you left the trailing brace off of the previous resource.
: line 480, col 12 of file /usr/local/bacula/etc/bacula-dir.conf
RunScript {
my job it's:
Client {
Name = wrack-fd
Address = wrack.mydomain.com
Catalog = Catalogo
Password = defwllkjh3g7834kjgetk4w84tn24f67g
}
Job {
Name = wrack
Client = wrack-fd
#FileSet = Full UML
FileSet = RawPartition
Write Bootstrap = /usr/local/bacula/var/bacula/working/wrack.bsr
#JobDefs = Default Servers
JobDefs = Raw Servers
RunScript {
Command = /home/uml-homes/uml-base/scripts/uml_baculaINI.sh
RunsOnClient = yes
RunsWhen = Before
AbortJobOnError = Yes
}
# RunScript {
# RunsOnClient = yes
# RunsWhen = After
# RunsOnSuccess = Yes
# Command = /home/uml-homes/uml-base/scripts/uml_baculaFIN.sh
# }
}
thanks again
lordjea
__
LLama Gratis a cualquier PC del Mundo.
Llamadas a fijos y móviles desde 1 céntimo por minuto.
http://es.voice.yahoo.com
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
On Fri, 29 Sep 2006, Kern Sibbald wrote: After having been totally frustrated chasing this kernel crash for the last few weeks (I really could not believe that it was not a Bacula bug), I have finally found a work around and at the same time, proven that it is a SuSE problem. One of the reason we dumped SLES on our production machines in favour of RHEL was that SUSE was consistently shipping with mismatching dynamic and static library versions - and would not fix it even when notified. SuSE may be great for home systems but having endured it (and SuSE's so-called support desk) for 4 years, I do not believe it is suitable for enterprise or business production use. Novell (SuSE's owners) management in the UK even tried to intervene on our behalf and were completely stonewalled by SuSE. If a company is this dysfunctional internally, then I don't hold out much hope for getting any problems fixed at all, let alone in a reasonable timeframe. AB - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Rif: what's the best way to do backup for a logical volume (LVM)
I use your point 2, and it works very well.
A Client Run Before Job stops the application, makes the snapshot,
restarts the application and mounts it on a different mount point.
A Client Run After Job unmounts the snapshot and removes it.
Obviously the File section specifies the snapshot mount point.
--
Ferdinando Pasqualetti
G.T.Dati srl
Tel. 0557310862 - 3356172731 - Fax 055720143
[EMAIL PROTECTED]
Inviato da: [EMAIL PROTECTED]
29/09/2006 12.17
Per
bacula-users@lists.sourceforge.net
CC
Oggetto
[Bacula-users] what's the best way to do backup for a logical volume
(LVM)
Hi gentlemen, I tried before to do things that could not be the best way
to
backup my logical volume.
The problem:
I have many logical volumes from machines that could not be sttoped, then
I
had two idea
The alternatives:
1- to do a backup for whole RAW partition making a snapshot
(lvcreate --snapshot -L 1Gb --name backup-root vg-uml-008/root)
2- to do a snapshot, then mount the snapshot and make a backup from the
mounted point (to do this I'm trying to use a RunScript in Job resource**)
3-somebody have any idea reliable?
I submit the question because you have a lot of experience that 3 weeks
that
I have
** but have a mistake:
29-sep 11:59 doom-dir: ERROR in parse_conf.c:853 Config error: Keyword
RunScript not permitted in this resource.
Perhaps you left the trailing brace off of the previous resource.
: line 480, col 12 of file
/usr/local/bacula/etc/bacula-dir.conf
RunScript {
my job it's:
Client {
Name = wrack-fd
Address = wrack.mydomain.com
Catalog = Catalogo
Password = defwllkjh3g7834kjgetk4w84tn24f67g
}
Job {
Name = wrack
Client = wrack-fd
#FileSet = Full UML
FileSet = RawPartition
Write Bootstrap = /usr/local/bacula/var/bacula/working/wrack.bsr
#JobDefs = Default Servers
JobDefs = Raw Servers
RunScript {
Command = /home/uml-homes/uml-base/scripts/uml_baculaINI.sh
RunsOnClient = yes
RunsWhen = Before
AbortJobOnError = Yes
}
# RunScript {
# RunsOnClient = yes
# RunsWhen = After
# RunsOnSuccess = Yes
# Command = /home/uml-homes/uml-base/scripts/uml_baculaFIN.sh
# }
}
thanks again
lordjea
__
LLama Gratis a cualquier PC del Mundo.
Llamadas a fijos y móviles desde 1 céntimo por minuto.
http://es.voice.yahoo.com
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share
your
opinions on IT business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
On Friday 29 September 2006 12:31, Alan Brown wrote: On Fri, 29 Sep 2006, Kern Sibbald wrote: After having been totally frustrated chasing this kernel crash for the last few weeks (I really could not believe that it was not a Bacula bug), I have finally found a work around and at the same time, proven that it is a SuSE problem. One of the reason we dumped SLES on our production machines in favour of RHEL was that SUSE was consistently shipping with mismatching dynamic and static library versions - and would not fix it even when notified. SuSE may be great for home systems but having endured it (and SuSE's so-called support desk) for 4 years, I do not believe it is suitable for enterprise or business production use. Well, with the exception of three things, I have found the installation and stability 10x better than Fedora. RHEL stability is exceptional, so I cannot comment, but the SuSE installer is far superior to the RHEL installer. However, I cannot afford to be on RHEL, and at least for the moment, would prefer not to be on one of the clones. One is the terribly slow speed for Yast2 doing rpm updates. That I have resolved by switching to yumex, which was what I preferred under Fedora anyway. The second problem is that they don't take enough care to make sure that their updates have all dependencies resolved. With Yast it is a catastrophe because of the slowness, with yumex, one just excludes a few updates and away you go. And finally, what is really disturbing me is this kernel oops. It really killed me -- for two weeks I beat tried everything (lots of work) thinking it was a Bacula bug. In the end, I had to reluctantly admit it was either a compiler or a kernel bug -- I've now proved it to be a kernel bug -- very frustrating. Novell (SuSE's owners) management in the UK even tried to intervene on our behalf and were completely stonewalled by SuSE. If a company is this dysfunctional internally, then I don't hold out much hope for getting any problems fixed at all, let alone in a reasonable timeframe. I suspect that has changed and evolve even more in the future. They seem to be taking it seriously and were quite polite in their response -- them taking it seriously is surely in part due to those of you who quickly responded to my call -- there are now 13 votes for fixing the bug. Thanks :-) If anyone can spare a few minutes to create a login and vote and/or signup for a CC on the bug, please do so. https://bugzilla.novell.com/show_bug.cgi?id=208782 Regards, Kern PS: I'm going to test it against their SuSE 10.2 kernel (to be released in Dec if I remember right) and if it fails, I'll file a blocker, which will ensure that it is fixed. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
On Fri, 29 Sep 2006, Kern Sibbald wrote: One of the reason we dumped SLES on our production machines in favour of RHEL was that SUSE was consistently shipping with mismatching dynamic and static library versions - and would not fix it even when notified. SuSE may be great for home systems but having endured it (and SuSE's so-called support desk) for 4 years, I do not believe it is suitable for enterprise or business production use. Well, with the exception of three things, I have found the installation and stability 10x better than Fedora. Fedora is by definition bleeding edge and we've frequently found that Fedora won't even install on new hardware while RHEL will. RHEL stability is exceptional, so I cannot comment, but the SuSE installer is far superior to the RHEL installer. However, I cannot afford to be on RHEL, and at least for the moment, would prefer not to be on one of the clones. Centos is _very_ stable. RHEL can be licensed quite cheaply if you don't buy the support package (about US$10/machine) The second problem is that they don't take enough care to make sure that their updates have all dependencies resolved. Yes. This, plus their refusal to deal with people pointing it out even if they have paid for support, plus the refusal to even talk to Novell management when we escalated it) gives the impression of a bunch of surly teenagers operating out of bedrooms rather than a professional software company. And finally, what is really disturbing me is this kernel oops. It really killed me -- for two weeks I beat tried everything (lots of work) thinking it was a Bacula bug. In the end, I had to reluctantly admit it was either a compiler or a kernel bug -- I've now proved it to be a kernel bug -- very frustrating. Your experiemce is not unique. Novell (SuSE's owners) management in the UK even tried to intervene on our behalf and were completely stonewalled by SuSE. If a company is this dysfunctional internally, then I don't hold out much hope for getting any problems fixed at all, let alone in a reasonable timeframe. I suspect that has changed and evolve even more in the future. This was current as of June 2006. PS: I'm going to test it against their SuSE 10.2 kernel (to be released in Dec if I remember right) and if it fails, I'll file a blocker, which will ensure that it is fixed. OpenSuse is similar to fedora - bleeding edge. SLES is supposedly a more stable animal - and at US$1500 per machine per year, I'd expect professional behaviour and responses, instead of refusal to respond when serious deficiencies are uncovered. AB - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
On Friday 29 September 2006 13:43, Alan Brown wrote: On Fri, 29 Sep 2006, Kern Sibbald wrote: One of the reason we dumped SLES on our production machines in favour of RHEL was that SUSE was consistently shipping with mismatching dynamic and static library versions - and would not fix it even when notified. SuSE may be great for home systems but having endured it (and SuSE's so-called support desk) for 4 years, I do not believe it is suitable for enterprise or business production use. Well, with the exception of three things, I have found the installation and stability 10x better than Fedora. Fedora is by definition bleeding edge and we've frequently found that Fedora won't even install on new hardware while RHEL will. Yes, I like to be on current software but not the bleeding edge. RHEL is sometimes not as current as I would like, though they are excellent for their security updates. RHEL stability is exceptional, so I cannot comment, but the SuSE installer is far superior to the RHEL installer. However, I cannot afford to be on RHEL, and at least for the moment, would prefer not to be on one of the clones. Centos is _very_ stable. RHEL can be licensed quite cheaply if you don't buy the support package (about US$10/machine) The last time I looked (some time ago), it was over $200/machine. That is too much for me. For a company or someone serious about servers, that's OK and quite far given their security updates. The second problem is that they don't take enough care to make sure that their updates have all dependencies resolved. Yes. This, plus their refusal to deal with people pointing it out even if they have paid for support, plus the refusal to even talk to Novell management when we escalated it) gives the impression of a bunch of surly teenagers operating out of bedrooms rather than a professional software company. Fortunately, I haven't seen that, and I hope it doesn't happen. And finally, what is really disturbing me is this kernel oops. It really killed me -- for two weeks I beat tried everything (lots of work) thinking it was a Bacula bug. In the end, I had to reluctantly admit it was either a compiler or a kernel bug -- I've now proved it to be a kernel bug -- very frustrating. Your experiemce is not unique. Novell (SuSE's owners) management in the UK even tried to intervene on our behalf and were completely stonewalled by SuSE. If a company is this dysfunctional internally, then I don't hold out much hope for getting any problems fixed at all, let alone in a reasonable timeframe. I suspect that has changed and evolve even more in the future. This was current as of June 2006. Hmmm. PS: I'm going to test it against their SuSE 10.2 kernel (to be released in Dec if I remember right) and if it fails, I'll file a blocker, which will ensure that it is fixed. OpenSuse is similar to fedora - bleeding edge. Well, up to today, I have found SuSE 10.1 very stable, and as far as I know they are not trying to put out a new version every 6 months (which IMO is the main cause of problems with Fedora). Also as I said, without going into all the gory details, the installation is at least 10x better than anything I have ever seen (in short after loading the first of 4 CDs, my external CDROM drive for my laptop died, and I was left with a system that didn't even have a root login, but with little effort I was able to complete the installation via the network, it picked up where it left off -- no other distro can do that! In addition, after getting all the right packages loaded, it even automatically reconfigured the screen to the correct driver, ...). SLES is supposedly a more stable animal - and at US$1500 per machine per year, I'd expect professional behaviour and responses, instead of refusal to respond when serious deficiencies are uncovered. AB - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] what's the best way to do backup for a logical volume (LVM)
Hi, I'm trying as you told me. The bacula it's backing up whole logical
volume
with the config:
Client {
Name = wrack-fd
Address = wrack.mydomain.com
Catalog = Catalogo
Password = seknmit5fwllkjh3g78klwuethñkw4nk645tig
}
Job {
Name = wrack
Client = wrack-fd
FileSet = RawPartition
Write Bootstrap = /usr/local/bacula/var/bacula/working/wrack.bsr
JobDefs = Raw Servers
Client Run Before Job =
/home/uml-homes/uml-base/scripts/scriptBacula/uml-008_baculaIni.sh
Client Run After Job =
/home/uml-homes/uml-base/scripts/scriptBacula/uml-008_baculaFin.sh
}
and FileSet, telling explicit to bacula the mount point to backup
FileSet {
Name = RawPartition
Include {
Options { signature=MD5; sparse=yes }
File = /home/uml-homes/uml-008/mnt
File = /home/uml-homes/uml-008/mnt/var
File = /home/uml-homes/uml-008/mnt/usr
}
}
Thanks again guys
lordjea
- Original Message -
From: MaxxAtWork [EMAIL PROTECTED]
To: [EMAIL PROTECTED]; bacula-users@lists.sourceforge.net
Sent: Friday, September 29, 2006 12:53 PM
Subject: Re: [Bacula-users] what's the best way to do backup for a logical
volume (LVM)
On 9/29/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
** but have a mistake:
29-sep 11:59 doom-dir: ERROR in parse_conf.c:853 Config error: Keyword
RunScript not permitted in this resource.
Perhaps you left the trailing brace off of the previous resource.
: line 480, col 12 of file
/usr/local/bacula/etc/bacula-dir.conf
RunScript {
my job it's:
Client {
Name = wrack-fd
Address = wrack.mydomain.com
Catalog = Catalogo
Password = defwllkjh3g7834kjgetk4w84tn24f67g
}
Job {
Name = wrack
Client = wrack-fd
#FileSet = Full UML
FileSet = RawPartition
Write Bootstrap = /usr/local/bacula/var/bacula/working/wrack.bsr
#JobDefs = Default Servers
JobDefs = Raw Servers
RunScript {
Command = /home/uml-homes/uml-base/scripts/uml_baculaINI.sh
RunsOnClient = yes
RunsWhen = Before
AbortJobOnError = Yes
}
# RunScript {
# RunsOnClient = yes
# RunsWhen = After
# RunsOnSuccess = Yes
# Command = /home/uml-homes/uml-base/scripts/uml_baculaFIN.sh
# }
}
I think what you need is a Client Run Before Job with a script to
create the snapshot,
then the regular Bacula backup of the snapshot, then a Client Run
After Job to unmount the snapshot.
--
Maxx
__
LLama Gratis a cualquier PC del Mundo.
Llamadas a fijos y móviles desde 1 céntimo por minuto.
http://es.voice.yahoo.com
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
I'll chime in with my endorsement of CentOS as well. I use it specifically for compatibility testing as a stand-in for RHEL as well as for commercial apps that only officially support RedHat and have never had a problem. The CentOS network also provides very timely security updates at no charge. --PLB Centos is _very_ stable. RHEL can be licensed quite cheaply if you don't buy the support package (about US$10/machine) The last time I looked (some time ago), it was over $200/machine. That is too much for me. For a company or someone serious about servers, that's OK and quite far given their security updates. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
On Fri, 29 Sep 2006, Kern Sibbald wrote: Centos is _very_ stable. RHEL can be licensed quite cheaply if you don't buy the support package (about US$10/machine) The last time I looked (some time ago), it was over $200/machine. That is too much for me. For a company or someone serious about servers, that's OK and quite far given their security updates. Redhat offer some discounts for developers, etc. As a Bacula feature request is now in their system and has been requested by at least 25 different customers, they may well be interested in giving you a free license for development purposes. I don't disagree with your assessment of installations - I use suse at home, but I have serious issues with their level of professionalism in the commercially supported products. (Having said that, RHEL installation is also very straoghtforward) AB - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
On Friday 29 September 2006 14:39, Alan Brown wrote: On Fri, 29 Sep 2006, Kern Sibbald wrote: Centos is _very_ stable. RHEL can be licensed quite cheaply if you don't buy the support package (about US$10/machine) The last time I looked (some time ago), it was over $200/machine. That is too much for me. For a company or someone serious about servers, that's OK and quite far given their security updates. Redhat offer some discounts for developers, etc. That is interesting. As a Bacula feature request is now in their system and has been requested by at least 25 different customers, they may well be interested in giving you a free license for development purposes. Hmmm. That is even more interesting. I actually have RHEL and access to their network, but that is because I administer a machine, where the organization (MercyShips) has a global RedHat license. That said, other than having the CDs for recovery purposes, which unfortunately I needed recently, I cannot load them on my machines. I don't disagree with your assessment of installations - I use suse at home, but I have serious issues with their level of professionalism in the commercially supported products. Well, distros are a bit of a religious thing and very personal. I look for leading edge software, good update/security service, and stability. RedHat is excellent for that, but now that they are commercial, too expensive. Fedora as you say and as I experienced is too bleeding edge. I asked them to use a 9 month release cycle, and they sent me a very kind reply giving their reasons for a 6 month cycle. I then looked at a lot of distros: debian, kubantu, ubantu, madrivia, ... However, most of them wouldn't even install on a leading edge Dell (debian, ubantu), others (kubantu) are for users that don't know Unix or the distro is a one man show without a significant organization, or rely on other distros for security patches, ... For me, for the moment, with the exception of this SCSI bug, SuSE has been great (as I say, for me). One good thing from the time I wasted on this bug is that I learned that within certain restrictions (SeLinux, AppArmor, ...), unlike rpms, I can mix and match kernels from different distros as I want. (Having said that, RHEL installation is also very straoghtforward) Yes, but if *anything* goes wrong, it simply dies. SuSE has a vga exception handler that takes over (sort of like a rescue disk) that allows you in many cases to get out of trouble -- e.g. switch where the source CDs are coming from, really quite cool. AB - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Email bounces
Hello, Someone on this list appears to be bouncing email back causing it to get delivered several times to the bacula-devel list. This seems to happen only for email sent to both bacula-users and bacula-devel at the same time. If you have recently subscribed or changed your email setup please check it. The culprit seems to be in GMT+2 timezone, using qmail, and is using i-netpartner.net as an ISP, relay or smtp host. Best regards, Kern - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Moving Files and updating catalog
On Friday 29 September 2006 12:02, Arno Lehmann wrote: Hi, Hi Arno, On 9/29/2006 11:19 AM, Benjamin Zeller wrote: Dear listusers, we are currently backing up 5 servers as a pilot before final bacula-installation with storage unit and bg server ;-) We configured expiration and pruning to suit our needs and everything's fine. What we are experiencing is, that we need to insert another harddrive to handle the upcoming data. So, what I am planning to do and where I need some advice: Plugin another harddrive, partition and format it and mount it etc. Then I wanna tell bacula in his config, to backup 2 servers to this new partition. I need to configure another StorageDevice which uses this new partition. Then I want to move the affected container files, which we're already written, from old to new partition. Then I need to tell bacula, that these container files moved, and that is my main question: How do I do that? Stop bacula and then I hope that there's a command that tells bacula, that these files are now on another place in the file system and that bacula doesn't look for them in the old but in the new partition from there on (should be important for restore ;-) ). Hmm. I don't think you need to shutdown Bacula for that. Rather I'd suggest the following: First, make sure you have a good understanding of the reationship between MediaType of the volume files and the storage devices. Depending on your version of Bacula you might find you're not able to easily restore from from any volume using any storage device. Well, to be more specific: atm, backups to hdd are situated in /backup. I want to plug in a 2nd harddrive and mount it at /backup1 (e.g.). Then I want to move backupfiles of Server1 and Server2 from /backup to /backup1. Then, after moving the volume files, simply change their media types. Why change their mediatypes? They are both MediaType=File, aren't they? I'd do this using a SQL query, either through bconsole or using your database front-end. Also, change the volumes pool association. I thought that there was a command in bacula (update in bconsole whatever) to tell bacula, that files from Server1 and Server2 have moved from /backup to /backup1. I wouldn't be able to tell, what changes are needed in SQL. Any help is appreciated. What I'd _really_ do, though, is thinking about your backend storage setup. I wouldn't consider storing backups on single hard drives a good idea. So won't I ;-) Rather use a standalone disk system where you can add drives to a RAID array on the fly. Start wih a RAID5 with 3 disks, and make sure you know how to add disks later to increase the overall capacity. The same result can be achieved by using LVM/dm in linux, but I'd prefer a stand-alone disk subsystem for better maintenance. This is the plan for the final server, as the current setup only displays a test, where given hardware was planned to be used. In the end, we will introduce a StorageUnit with plenty of space, raid and the possibility to increase capacity by simply plugging in another drive, be sure that we made this reflections,too. Nevertheless, thanks a lot for your suggestions. Arno Benni Greetings, Benni - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users -- Benjamin Zeller Ing.-Büro Hohmann Bahnhofstr. 34 D-82515 Wolfratshausen Tel.: +49 (0)8171 347 88 12 Mobil: +49 (0)160 99 11 55 23 Fax: +49 (0)8171 910 778 mailto: [EMAIL PROTECTED] www.ibh-wor.de pgpq5UOPOpc1v.pgp Description: PGP signature - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
I'm catching up with this thread with interest after just coming into work this morning. Kern, my sympathy. I can imagine how frustrated you are, but it just goes to show how good a programmer you are, that it wasn't Bacula after all! After reading the merits weaknesses of RHEL, SUSE, Centos, I have only one question. Have you considered or tried Debian lately? I settled on Debian several years ago after becoming disenchanted with each of the major commercial distributions and haven't regreted it for a moment. (My servers are humming along happily with Debian AMD64-Sarge Bacula 1.36 as we speak!) Three Cheers for Bacula! cmr On Friday 29 September 2006 08:00, Kern Sibbald wrote: On Friday 29 September 2006 14:39, Alan Brown wrote: On Fri, 29 Sep 2006, Kern Sibbald wrote: Centos is _very_ stable. RHEL can be licensed quite cheaply if you don't buy the support package (about US$10/machine) The last time I looked (some time ago), it was over $200/machine. That is too much for me. For a company or someone serious about servers, that's OK and quite far given their security updates. Redhat offer some discounts for developers, etc. That is interesting. As a Bacula feature request is now in their system and has been requested by at least 25 different customers, they may well be interested in giving you a free license for development purposes. Hmmm. That is even more interesting. I actually have RHEL and access to their network, but that is because I administer a machine, where the organization (MercyShips) has a global RedHat license. That said, other than having the CDs for recovery purposes, which unfortunately I needed recently, I cannot load them on my machines. I don't disagree with your assessment of installations - I use suse at home, but I have serious issues with their level of professionalism in the commercially supported products. Well, distros are a bit of a religious thing and very personal. I look for leading edge software, good update/security service, and stability. RedHat is excellent for that, but now that they are commercial, too expensive. Fedora as you say and as I experienced is too bleeding edge. I asked them to use a 9 month release cycle, and they sent me a very kind reply giving their reasons for a 6 month cycle. I then looked at a lot of distros: debian, kubantu, ubantu, madrivia, ... However, most of them wouldn't even install on a leading edge Dell (debian, ubantu), others (kubantu) are for users that don't know Unix or the distro is a one man show without a significant organization, or rely on other distros for security patches, ... For me, for the moment, with the exception of this SCSI bug, SuSE has been great (as I say, for me). One good thing from the time I wasted on this bug is that I learned that within certain restrictions (SeLinux, AppArmor, ...), unlike rpms, I can mix and match kernels from different distros as I want. (Having said that, RHEL installation is also very straoghtforward) Yes, but if *anything* goes wrong, it simply dies. SuSE has a vga exception handler that takes over (sort of like a rescue disk) that allows you in many cases to get out of trouble -- e.g. switch where the source CDs are coming from, really quite cool. AB - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users -- Debian 'Sarge': Registered Linux User #241964 More laws, less justice. -- Marcus Tullius Ciceroca, 42 BC - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Job not found??
Hello. Last night , one of my backup jobs started, but for some reason a problem occour. Here is part of the status dir command: Running Jobs: JobId Level Name Status == 1304 FullJob.SCD03.2006-09-28_23.05.09 is waiting for Client SCD03-fd to connect to Storage FileUSERS 1305 FullJob.BIBLIOTECA01.2006-09-28_23.05.10 is waiting for its start time 1315 FullJob.jtgv.gsi.2006-09-29_09.50.00 is waiting for its start time Now when I try to cancel the job, i get a job not found: *cancel Select Job: 1: JobId=1304 Job=Job.SCD03.2006-09-28_23.05.09 2: JobId=1305 Job=Job.BIBLIOTECA01.2006-09-28_23.05.10 3: JobId=1315 Job=Job.jtgv.gsi.2006-09-29_09.50.00 Choose Job to cancel (1-3): 1 2001 Job Job.SCD03.2006-09-28_23.05.09 marked to be canceled. 3904 Job Job.SCD03.2006-09-28_23.05.09 not found. How do I kill it? It will probaby work in i restart bacula-dir, but then I would have to start the other jobs manually. Thanks. -- Jaime Ventura [Infra-estruturas e Comunicações] Rua Dr. António Bernardino de Almeida, 431 4200 - 072 Porto Telef: +351 22 834 05 00 (04) - ext. 1641 Fax: +351 22 832 11 59 e-mail: [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] url:www.isep.ipp.pt http://www.isep.ipp.pt d - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
On Friday 29 September 2006 16:53, Mike Reinehr wrote: I'm catching up with this thread with interest after just coming into work this morning. Kern, my sympathy. I can imagine how frustrated you are, but it just goes to show how good a programmer you are, that it wasn't Bacula after all! Thanks. :-) After reading the merits weaknesses of RHEL, SUSE, Centos, I have only one question. Have you considered or tried Debian lately? I settled on Debian several years ago after becoming disenchanted with each of the major commercial distributions and haven't regreted it for a moment. (My servers are humming along happily with Debian AMD64-Sarge Bacula 1.36 as we speak!) Yes, I tried Debian (3.1 if I remember right), but where I could get it to install, it came with a 2.4 kernel, and it won't install on most systems I want because it cannot recognize either my graphics card or my ethernet card. Without the ethernet card, I either go through a painful process of writing files to a CD or give up, which is what I did. Yes, Debian is very stable, and they are good about applying security fixes to their production system, but getting it running is a problem, they don't yet (at least as of a few months ago) provide security updates for testing, and they lag very far behind where I want to be on my development machine. I don't want to sound in the least like I am knocking Debian, it just doesn't at the moment quite fit what I am looking for. I see that they are discussing/changing a lot of things (possibly hiring developers!), perhaps some of that comes from Ubantu, so I'm lingering in the sidelines waiting to see what happens. I don't exclude Debian for my server, which is currently on FC4 -- except that I don't think that Debian either has SELinux (as my server does) or AppArmor as SuSE does. Three Cheers for Bacula! Thanks, especially for the words of encouragement. Kern PS: Thanks to all of you who voted for the bug report I filed -- the last time I looked there were 18 votes! The good news as of a couple of minutes ago is that their 10.2 kernel 2.6.18-rc5-git6-2-bigsmp does not have the bug, so at least I have an all SuSE solution :-) cmr On Friday 29 September 2006 08:00, Kern Sibbald wrote: On Friday 29 September 2006 14:39, Alan Brown wrote: On Fri, 29 Sep 2006, Kern Sibbald wrote: Centos is _very_ stable. RHEL can be licensed quite cheaply if you don't buy the support package (about US$10/machine) The last time I looked (some time ago), it was over $200/machine. That is too much for me. For a company or someone serious about servers, that's OK and quite far given their security updates. Redhat offer some discounts for developers, etc. That is interesting. As a Bacula feature request is now in their system and has been requested by at least 25 different customers, they may well be interested in giving you a free license for development purposes. Hmmm. That is even more interesting. I actually have RHEL and access to their network, but that is because I administer a machine, where the organization (MercyShips) has a global RedHat license. That said, other than having the CDs for recovery purposes, which unfortunately I needed recently, I cannot load them on my machines. I don't disagree with your assessment of installations - I use suse at home, but I have serious issues with their level of professionalism in the commercially supported products. Well, distros are a bit of a religious thing and very personal. I look for leading edge software, good update/security service, and stability. RedHat is excellent for that, but now that they are commercial, too expensive. Fedora as you say and as I experienced is too bleeding edge. I asked them to use a 9 month release cycle, and they sent me a very kind reply giving their reasons for a 6 month cycle. I then looked at a lot of distros: debian, kubantu, ubantu, madrivia, ... However, most of them wouldn't even install on a leading edge Dell (debian, ubantu), others (kubantu) are for users that don't know Unix or the distro is a one man show without a significant organization, or rely on other distros for security patches, ... For me, for the moment, with the exception of this SCSI bug, SuSE has been great (as I say, for me). One good thing from the time I wasted on this bug is that I learned that within certain restrictions (SeLinux, AppArmor, ...), unlike rpms, I can mix and match kernels from different distros as I want. (Having said that, RHEL installation is also very straoghtforward) Yes, but if *anything* goes wrong, it simply dies. SuSE has a vga exception handler that takes over (sort of like a rescue disk) that allows you in many cases to get out of trouble -- e.g. switch where the source CDs are coming from, really quite cool.
[Bacula-users] about manual and rejected volumes
hi list,
1st there's a recurrent bug in the manual, you can't
use latex commands (\bf) inside a verbatim section.
i think there's an alternative section type where you
can do that.
that gives lines like these:
===
(in the Console program)
*{\bf restore}
First you select one or more JobIds that contain files
===
about the Rejected Volumes After a Crash tip, i
think bacula changed in between. here's my bconsole
session:
===
*m
29-Sep 16:19 server-sd: Please mount Volume Daily2
on Storage Device hpsw for Job
NightlySave.2006-09-29_16.18.55
*update
Update choice:
1: Volume parameters
2: Pool from resource
3: Slots from autochanger
Choose catalog item to update (1-3): 1
Defined Pools:
1: Daily
2: Weekly
3: Monthly
Select the Pool (1-3): 1
+-++---+--+--+--+-+--+---+---+-+
| MediaId | VolumeName | VolStatus | VolBytes |
VolFiles | VolRetention | Recycle | Slot | InChanger |
MediaType | LastWritten |
+-++---+--+--+--+-+--+---+---+-+
| 1 | Daily1 | Error | 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 2006-09-29 15:35:44 |
| 2 | Daily2 | Append| 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 0 |
| 3 | Daily3 | Append| 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 0 |
| 4 | Daily4 | Append| 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 0 |
| 5 | Daily5 | Append| 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 0
|
+-++---+--+--+--+-+--+---+---+-+
Enter MediaId or Volume name: 1
Updating Volume Daily1
Parameters to modify:
1: Volume Status
2: Volume Retention Period
3: Volume Use Duration
4: Maximum Volume Jobs
5: Maximum Volume Files
6: Maximum Volume Bytes
7: Recycle Flag
8: Slot
9: InChanger Flag
10: Volume Files
11: Pool
12: Volume from Pool
13: All Volumes from Pool
14: Done
Select parameter to modify (1-14): 10
Warning changing Volume Files can result
in loss of data on your Volume
Current Volume Files is: 0
Enter new number of Files for Volume: 9
Normally, you should only increase Volume Files by
one!
Continue? (yes/no): yes
New Volume Files is: 9
===
and yes i had a 9/0 difference:
===
29-Sep 01:30 server-dir: No prior Full backup Job
record found.
29-Sep 01:30 server-dir: No prior or suitable Full
backup found. Doing FULL backup.
29-Sep 01:30 server-dir: Start Backup JobId 1,
Job=NightlySave.2006-09-29_01.30.00
29-Sep 01:30 server-sd: Volume Daily1 previously
written, moving to end of data.
29-Sep 01:30 server-sd:
NightlySave.2006-09-29_01.30.00 Error: I canot write
on Volume Daily1 because:
The number of files mismatch! Volume=9 Catalog=0
29-Sep 01:30 server-sd: Marking Volume Daily1 in
Error in Catalog.
===
regards,
mlistus
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula status update + you can help
Kern, On Friday 29 September 2006 10:28, Kern Sibbald wrote: On Friday 29 September 2006 16:53, Mike Reinehr wrote: I'm catching up with this thread with interest after just coming into work this morning. Kern, my sympathy. I can imagine how frustrated you are, but it just goes to show how good a programmer you are, that it wasn't Bacula after all! Thanks. :-) After reading the merits weaknesses of RHEL, SUSE, Centos, I have only one question. Have you considered or tried Debian lately? I settled on Debian several years ago after becoming disenchanted with each of the major commercial distributions and haven't regreted it for a moment. (My servers are humming along happily with Debian AMD64-Sarge Bacula 1.36 as we speak!) Yes, I tried Debian (3.1 if I remember right), but where I could get it to install, it came with a 2.4 kernel, and it won't install on most systems I want because it cannot recognize either my graphics card or my ethernet card. Without the ethernet card, I either go through a painful process of writing files to a CD or give up, which is what I did. I understand. The Sarge (3.1r3) installer also comes with a 2.6 kernel, but in my case, even that wasn't new enough to support my SATA controller. I think I ended up using a Knoppix disk with a newer kernel to bootstrap my installation. The new Etch installer (beta 3) includes a 2.6.16 kernel that solves that problem, but if you want Sarge you either have to obtain a custom install disk, or do a basic Etch install then downgrade. Yes, Debian is very stable, and they are good about applying security fixes to their production system, but getting it running is a problem, they don't yet (at least as of a few months ago) provide security updates for testing, and they lag very far behind where I want to be on my development machine. FWIW the security team has begun providing at least some security updates for Etch. I don't want to sound in the least like I am knocking Debian, it just doesn't at the moment quite fit what I am looking for. I see that they are discussing/changing a lot of things (possibly hiring developers!), perhaps some of that comes from Ubantu, so I'm lingering in the sidelines waiting to see what happens. I don't exclude Debian for my server, which is currently on FC4 -- except that I don't think that Debian either has SELinux (as my server does) or AppArmor as SuSE does. Actually, SELinux does seem now to be available both for Sarge Etch, but I don't know anything about it. Oh well, I just couldn't resist putting in a plug for Debian, but I think I rather would prefer discussing religion or politics than debating the merits of Linux distributions! ;-) Cheers! cmr PS Bearing in mind your message about email bounces, I received three copies of your reply--one addressed to me, personally, and two addressed to bacula-users. PPS I'm a user of apcupsd, too! Three Cheers for Bacula! Thanks, especially for the words of encouragement. Kern PS: Thanks to all of you who voted for the bug report I filed -- the last time I looked there were 18 votes! The good news as of a couple of minutes ago is that their 10.2 kernel 2.6.18-rc5-git6-2-bigsmp does not have the bug, so at least I have an all SuSE solution :-) cmr On Friday 29 September 2006 08:00, Kern Sibbald wrote: On Friday 29 September 2006 14:39, Alan Brown wrote: On Fri, 29 Sep 2006, Kern Sibbald wrote: Centos is _very_ stable. RHEL can be licensed quite cheaply if you don't buy the support package (about US$10/machine) The last time I looked (some time ago), it was over $200/machine. That is too much for me. For a company or someone serious about servers, that's OK and quite far given their security updates. Redhat offer some discounts for developers, etc. That is interesting. As a Bacula feature request is now in their system and has been requested by at least 25 different customers, they may well be interested in giving you a free license for development purposes. Hmmm. That is even more interesting. I actually have RHEL and access to their network, but that is because I administer a machine, where the organization (MercyShips) has a global RedHat license. That said, other than having the CDs for recovery purposes, which unfortunately I needed recently, I cannot load them on my machines. I don't disagree with your assessment of installations - I use suse at home, but I have serious issues with their level of professionalism in the commercially supported products. Well, distros are a bit of a religious thing and very personal. I look for leading edge software, good update/security service, and stability. RedHat is excellent for that, but now that they are
Re: [Bacula-users] about manual and rejected volumes
On Friday 29 September 2006 18:43, m listus wrote:
hi list,
1st there's a recurrent bug in the manual, you can't
use latex commands (\bf) inside a verbatim section.
i think there's an alternative section type where you
can do that.
There should not be any {\bf ...} inside a verbatim. If there is, it is
probably because we forgot an \end{verbatim} or made a typo in entering it.
If you point me to the exact file where there are problems (the first
occurrence), I will be happy to fix it.
that gives lines like these:
===
(in the Console program)
*{\bf restore}
First you select one or more JobIds that contain files
===
about the Rejected Volumes After a Crash tip, i
think bacula changed in between. here's my bconsole
session:
Yes, this is quite possible (Bacula changed). The manual is so big that I
cannot possibly check the whole thing each time. If you give the the exact
file where the problem exists, I can fix it.
===
*m
29-Sep 16:19 server-sd: Please mount Volume Daily2
on Storage Device hpsw for Job
NightlySave.2006-09-29_16.18.55
*update
Update choice:
1: Volume parameters
2: Pool from resource
3: Slots from autochanger
Choose catalog item to update (1-3): 1
Defined Pools:
1: Daily
2: Weekly
3: Monthly
Select the Pool (1-3): 1
+-++---+--+--+--+-+--+---+---+-+
| MediaId | VolumeName | VolStatus | VolBytes |
VolFiles | VolRetention | Recycle | Slot | InChanger |
MediaType | LastWritten |
+-++---+--+--+--+-+--+---+---+-+
| 1 | Daily1 | Error | 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 2006-09-29 15:35:44 |
| 2 | Daily2 | Append| 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 0 |
| 3 | Daily3 | Append| 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 0 |
| 4 | Daily4 | Append| 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 0 |
| 5 | Daily5 | Append| 0| 0
| 864000 | 1 | 0| 0 | DDS-4
| 0
|
+-++---+--+--+--+-+--+---+---+-+
Enter MediaId or Volume name: 1
Updating Volume Daily1
Parameters to modify:
1: Volume Status
2: Volume Retention Period
3: Volume Use Duration
4: Maximum Volume Jobs
5: Maximum Volume Files
6: Maximum Volume Bytes
7: Recycle Flag
8: Slot
9: InChanger Flag
10: Volume Files
11: Pool
12: Volume from Pool
13: All Volumes from Pool
14: Done
Select parameter to modify (1-14): 10
Warning changing Volume Files can result
in loss of data on your Volume
Current Volume Files is: 0
Enter new number of Files for Volume: 9
Normally, you should only increase Volume Files by
one!
Continue? (yes/no): yes
New Volume Files is: 9
===
and yes i had a 9/0 difference:
===
29-Sep 01:30 server-dir: No prior Full backup Job
record found.
29-Sep 01:30 server-dir: No prior or suitable Full
backup found. Doing FULL backup.
29-Sep 01:30 server-dir: Start Backup JobId 1,
Job=NightlySave.2006-09-29_01.30.00
29-Sep 01:30 server-sd: Volume Daily1 previously
written, moving to end of data.
29-Sep 01:30 server-sd:
NightlySave.2006-09-29_01.30.00 Error: I canot write
on Volume Daily1 because:
The number of files mismatch! Volume=9 Catalog=0
29-Sep 01:30 server-sd: Marking Volume Daily1 in
Error in Catalog.
===
regards,
mlistus
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
Re: [Bacula-users] Bacula status update + you can help
On Fri, 29 Sep 2006, Les Mikesell wrote: CentOS has some additions to the stock RH version as well, like an optional kernel with firewire support and the reiserfs and xfs filesystems. It uses yum for updates and they generally stay within a few days of RH update releases. This is getting well outside the realm of Bacula itself, but I would really like to see the Enterprise volume management system (EVMS) in widespread use, as it makes disk hardware migration a painless operation while bringing all the various disk-related tools under one interface. Migrating storage has always been the achilles heel of almost every *nix, with downtime invariably necessary - and that tends to be hard to schedule in a 24*7 enterprise or research environment. AB - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] about manual and rejected volumes
If you point me to the exact file where there are
problems (the first
occurrence), I will be happy to fix it.
here's the diff -Naur for docs/manual/tips.tex taken
from your CVS (less than 1h ago).
i removed all {\bf inside verbatim sections, and
corrected 1 typo NFS,NSF.
if i have time, i could clean up the rest. (lynx
*.html |grep {\bf ...)
=
--- tips.tex2006-09-29 18:10:16.0 +
+++ tips.tex.orig 2006-09-29 17:28:57.0
+
@@ -302,7 +302,7 @@
Note, one disadvantage of writing to an NFS mounted
volume as I do is
that if the other machine goes down, the OS will wait
forever on the fopen()
call that Bacula makes. As a consequence, Bacula will
completely stall until
-the machine exporting the NFS mounts comes back up. A
possible solution to this
+the machine exporting the NSF mounts comes back up. A
possible solution to this
problem was provided by Andrew Hilborne, and consists
of using the {\bf soft}
option instead of the {\bf hard} option when mounting
the NFS volume, which is
typically done in {\bf /etc/fstab}/. The NFS
documentation explains these
@@ -325,7 +325,7 @@
\footnotesize
\begin{verbatim}
(in the Console program)
-*restore
+*{\bf restore}
First you select one or more JobIds that contain
files
to be restored. You will then be presented several
methods
of specifying the JobIds. Then you will be allowed to
@@ -337,12 +337,12 @@
4: Enter SQL list command
5: Select the most recent backup for a client
6: Cancel
-Select item: (1-6): 5
+Select item: (1-6): {\bf 5}
The defined Client resources are:
1: Minimatou
2: Rufus
3: Timmy
-Select Client (File daemon) resource (1-3): 2
+Select Client (File daemon) resource (1-3): {\bf 2}
The defined FileSet resources are:
1: Kerns Files
Item 1 selected automatically.
@@ -360,7 +360,7 @@
remove files to be restored. All files are initially
added.
Enter done to leave this mode.
cwd is: /
-$ done
+$ {\bf done}
84 files selected to restore.
Run Restore job
JobName:kernsrestore
@@ -370,10 +370,10 @@
Client: Rufus
Storage:File
JobId: *None*
-OK to run? (yes/mod/no): no
-quit
+OK to run? (yes/mod/no): {\bf no}
+{\bf quit}
(in a shell window)
-cp ../working/restore.bsr
/mnt/deuter/files/backup/rufus.bsr
+{\bf cp ../working/restore.bsr
/mnt/deuter/files/backup/rufus.bsr}
\end{verbatim}
\normalsize
@@ -452,12 +452,12 @@
\footnotesize
\begin{verbatim}
-update
+{\bf update}
Update choice:
1: Volume parameters
2: Pool from resource
3: Slots from autochanger
-Choose catalog item to update (1-3): 1
+Choose catalog item to update (1-3): {\bf 1}
Defined Pools:
1: Default
2: File
@@ -467,7 +467,7 @@
+---+-++-+---+--+--+--+-+
| 1 | test01 | DDS-4 | Error | 352427156 |
... | 31536000 | 1| 0 |
+---+-++-+---+--+--+--+-+
-Enter MediaId or Volume name: 1
+Enter MediaId or Volume name: {\bf 1}
\end{verbatim}
\normalsize
@@ -491,11 +491,11 @@
9: Volume Files
10: Pool
11: Done
-Select parameter to modify (1-11): 9
+Select parameter to modify (1-11): {\bf 9}
Warning changing Volume Files can result
in loss of data on your Volume
Current Volume Files is: 10
-Enter new number of Files for Volume: 11
+Enter new number of Files for Volume: {\bf 11}
New Volume Files is: 11
Updating Volume test01
Parameters to modify:
@@ -510,7 +510,7 @@
9: Volume Files
10: Pool
11: Done
-Select parameter to modify (1-10): 1
+Select parameter to modify (1-10): {\bf 1}
\end{verbatim}
\normalsize
@@ -528,7 +528,7 @@
4: Full
5: Used
6: Read-Only
-Choose new Volume Status (1-6): 1
+Choose new Volume Status (1-6): {\bf 1}
New Volume status is: Append
Updating Volume test01
Parameters to modify:
@@ -543,7 +543,7 @@
9: Volume Files
10: Pool
11: Done
-Select parameter to modify (1-11): 11
+Select parameter to modify (1-11): {\bf 11}
Selection done.
\end{verbatim}
\normalsize
@@ -721,7 +721,7 @@
\footnotesize
\begin{verbatim}
-list volumes
+{\bf list volumes}
Using default Catalog name=BackupDB DB=bacula
Pool: Default
=
Yes, this is quite possible (Bacula changed). The
manual is so big that I
cannot possibly check the whole thing each time. If
you give the the exact
file where the problem exists, I can fix it.
i'm using 1.36.3 on sarge, so my session itself could
be outdated. so i didn't want to change the diff...
here it is (starting docs/manual/tips.tex:482)
=
Parameters to modify:
1: Volume Status
2: Volume Retention Period
3: Volume Use Duration
4: Maximum Volume Jobs
5: Maximum Volume Files
6: Maximum Volume Bytes
7: Recycle Flag
8: Slot
9: InChanger Flag
10: Volume Files
11: Pool
12: Volume from Pool
13: All Volumes from Pool
14: Done
Select
[Bacula-users] Console messages and ACLs
I just logged into Bacula from a client machine that has tighly restricted access. It can only access its own pools, jobs, etc. When I logged in with wx-console I was surprised by a sudden flood of queued console messages for all jobs from last night. Those messages of course give quite a bit of detail that I do not want to be seen by the typical client. Offhand I don't see a good way in the Console resource to restrict message access. Is the best option for the moment simply to disable console message support and depend on email and the logs? -- -- Michael - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Console messages and ACLs
On Friday 29 September 2006 23:19, Michael Brennen wrote: I just logged into Bacula from a client machine that has tighly restricted access. It can only access its own pools, jobs, etc. When I logged in with wx-console I was surprised by a sudden flood of queued console messages for all jobs from last night. Those messages of course give quite a bit of detail that I do not want to be seen by the typical client. Offhand I don't see a good way in the Console resource to restrict message access. Is the best option for the moment simply to disable console message support and depend on email and the logs? Good question. Please try it and let me know if it works as you expect. The whole console interface needs to be redesigned. The current design handles only one console -- if you connect two at a time, I imagine who gets the console messages generated by jobs is most the first one that happens to be active at that instant, so any new messages probably will be more or less randomly distributed between the two consoles. Not too cool, but since no one has complained yet, I haven't looked at it :-) I suppose that at a minimum I wonder if it would be possible to hack the interface so that console messages never get sent to any restricted console. By the way, the current CVS has ACLs covering all the various restore options. However, someone complained that status output showed all jobs -- yes, that is correct, and if one doesn't want one's users to see everything that is in a status output, disable the status command, since I don't feel it is worth the effort to try to add finer grain ACL to the status output. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Console messages and ACLs
On Friday 29 September 2006 23:19, Michael Brennen wrote: I just logged into Bacula from a client machine that has tighly restricted access. It can only access its own pools, jobs, etc. When I logged in with wx-console I was surprised by a sudden flood of queued console messages for all jobs from last night. Those messages of course give quite a bit of detail that I do not want to be seen by the typical client. Offhand I don't see a good way in the Console resource to restrict message access. Is the best option for the moment simply to disable console message support and depend on email and the logs? I just looked at the code, and this problem is very likely *much* easier to solve than I thought, in fact, I would classify it as a bug since to receive messages, the restricted console should have a messages ACL, and if it does not, it should not receive any messages. This will be fixed shortly. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Console messages and ACLs
On Friday 29 September 2006 04:41 pm, Kern Sibbald wrote: On Friday 29 September 2006 23:19, Michael Brennen wrote: Offhand I don't see a good way in the Console resource to restrict message access. Is the best option for the moment simply to disable console message support and depend on email and the logs? I just looked at the code, and this problem is very likely *much* easier to solve than I thought, in fact, I would classify it as a bug since to receive messages, the restricted console should have a messages ACL, and if it does not, it should not receive any messages. This will be fixed shortly. I just disabled all console messages in the Messages resources in the director and for the moment that will do what I need. Good news on the source ACL fixes. :) -- -- Michael - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] CVS multiple drives (was Console messages and ACLs)
On Friday 29 September 2006 04:36 pm, Kern Sibbald wrote: By the way, the current CVS has ACLs covering all the various restore options. I noted today on current cvs that the last 20 jobs list in the wx console now gives a message about sql access denied or something like that. Thanks, that will help restrict information. I am glad you were finally able to resolve the kernel crash; that sort of non-self-induced bug is really frustrating and disfocusing. Just for information so these are not lost, I will note again a couple of cvs bugs that I reported a few days ago. One problem I still see in CVS is that only one drive is used on a four drive autochanger. A couple of days ago I did send the output of some jobs showing that drive 0 was always used. You mentioned a few days ago that I might need to supply some sort of regression script test to demonstrate that. I don't know what such a regression script would look like or where to find examples. If you can point me to a section in the docs I will be glad to supply what information need. Another problem that I reported recently is that I cannot restore from an ACL restricted windows wx console. After going through the file select process and submitting the job, a message is returned that access is denied to a pool other than the one that is selected for restore. Thanks again, -- -- Michael - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] CVS multiple drives (was Console messages and ACLs)
hi michael, You mentioned a few days ago that I might need to supply some sort of regression script test to demonstrate that. I don't know what such a regression script would look like or where to find examples. If you can point me to a section in the docs I will be glad to supply what information need. basically, you build a script that does the operation you experience problems with. then you run it on a working version of bacula, and then on the cvs version that shows the regression. i have come across bacula's regression tests, you can get them this way: cvs -d:pserver:[EMAIL PROTECTED]:/cvsroot/bacula login cvs -z3 -d:pserver:[EMAIL PROTECTED]:/cvsroot/bacula co -P regress if it helps... mlistus. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] CVS multiple drives (was Console messages and ACLs)
On Friday 29 September 2006 17:39, m listus wrote: You mentioned a few days ago that I might need to supply some sort of regression script test to demonstrate that. I don't know what such a regression script would look like or where to find examples. If you can point me to a section in the docs I will be glad to supply what information need. basically, you build a script that does the operation you experience problems with. then you run it on a working version of bacula, and then on the cvs version that shows the regression. Thanks, I checked out the regress cvs and will look at it. -- -- Michael - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] syslog and btape
hi, i'm using 1.36.2 on sarge. i get this kernel message on dmesg on a per job basis i think: st0: MTSETDRVBUFFER only allowed for root. i read the Syslog Errors section in the tape testing chapter, but it's too vague: . In general, the message can be ignored providing you are sure that your OS parameters are properly configured as described earlier in this manual. i read the chapter, and i'm still not sure the OS params are right. that said it works. but i want to know if i need to set some st0 param at boot time in an init script. that seems the most appropriate. i have an 8mb internal tape buffer, and i even don't know which size is set by the kernel. 2nd point, i've been trying to get the 2nd part of tape testing chapter to work, that is: fill w/ 2 tapes. all tests work (including fill w/ 1 tape). the problem happens when using the 2nd tape. this might be a problem becaus i'm using the 'daily, weekly, monthly scheme' that uses a daily tape until it fills, and the it requests the next one. thanks, mlistus __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
