Re: [Bacula-users] Re: Bacula Multiple Tape Drives
Hi ! Now i`m on the list too, thanks for the nice explanation about the atimes tmpwatch ;) Kern Sibbald wrote: Yes, Bacula modifies atime for all files that it backs up. However, there is an option keepatime = yes, which will cause Bacula to reset the atime on files that it backs up. However, this option does not work until version 1.37.30 and later. This sounds very good - i`ve already setup the 1.37.30 (so no worries about atimes) on our server and all the initial tests look promising ... now i`m going to create a configuration for our site now ... and that means reading the documentation ... LOTS of reading :-) I suppose the /lib/tls warning is still valid ? We`re running RedHat Enterprise Linux 4 WS (Our company is very RedHat oriented). Best regards, -- Daniel HoltkampRiege Software International GmbH System Administration Mollsfeld 10 40670 Meerbusch, Germany Phone: +49-2159-9148-41 mail: [EMAIL PROTECTED] Fax: +49-2159-9148-11 --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Restore doesn't restore
Hi list, I'm happily using bacula 1.36.2 on a Debian sarge machine. The backup/restore was running great for weeks on a clone machine. Recently I changed to a HP server with the same Ultrium streamer. Again the backup is running great, but when I try to restore it appears to go okay but there aren't any files written. The output of the restore job is (cut) : 02-Aug 17:13 bacula-sd: Ready to read from volume Friday1 on device /dev/nst0. 02-Aug 17:13 bacula-sd: Forward spacing to file:block 6:0. server1-fd: -rwxrwxrwx 1 00 749056 2004-05-03 17:34:16 +/brs/E//Goldmine/relatie/Contact1.DBT server1-fd: -rwxrwxrwx 1 00 785408 2004-05-03 17:34:23 +/brs/E//Goldmine/relatie/Contact2.MDX server1-fd: -rwxrwxrwx 1 008327832 2004-05-03 17:34:23 +/brs/E//Goldmine/relatie/ContGrps.DBF server1-fd: -rwxrwxrwx 1 00 11201536 2004-05-03 17:34:25 +/brs/E//Goldmine/relatie/ContGrps.MDX 02-Aug 17:17 bacula-sd: End of Volume at file 6 on device /dev/nst0, Volume +Friday1 02-Aug 17:17 bacula-sd: End of all volumes. server1-fd: -rwxrwxrwx 1 00 10546 2004-05-03 17:35:13 +/brs/E//Goldmine/relatie/ContUDef.DBF server1-fd: -rwxrwxrwx 1 00 10240 2004-05-03 17:35:13 +/brs/E//Goldmine/relatie/ContUDef.MDX 02-Aug 17:18 bacula-dir: Bacula 1.36.2 (28Feb05): 02-Aug-2005 17:18:47 JobId: 119 Job:Restore-sv1.2005-08-02_17.09.22 Client: server1-fd Start time: 02-Aug-2005 17:09:24 End time: 02-Aug-2005 17:18:47 Files Expected: 17 Files Restored: 17 Bytes Restored: 242,580,387 Rate: 430.9 KB/s FD Errors: 0 FD termination status: OK SD termination status: OK Termination:Restore OK But after that no files show up in brs. I created the brs dir as a test. The restorejob defines /tmp/bacula-restore The /brs dir has system wide read/write rights. Also running bconsole as user or root doesn't matter. I backup with signature = MD5, the files come from a windows machine. Any idea what is going wrong ? Regards, Joost -- Things you don't want your SysAdmin to say -14. Sorry, the new equipment didn't get budgeted. --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] volume recycling
Martin Simmons wrote: On Tue, 2 Aug 2005 13:43:16 +0200, Kern Sibbald [EMAIL PROTECTED] said: Kern On Tuesday 02 August 2005 13:16, Martin Simmons wrote: On Tue, 02 Aug 2005 10:37:20 +0200, Rico [EMAIL PROTECTED] said: Rico it failed again. the admin job was ok. the volume was pruned and Rico recycled but thepig-job failed. The output looks normal to me, except for when the sd says Cannot find any appendable volumes. This could be a bug, so I suggest you report it at http://bugs.bacula.org/ including the latest console output. Kern I would suggest distilling the problem down to the very essence. From what I Kern saw (briefly) in the emails exchanged, I would have a hard time figuring out Kern what is not working correctly. I think this is the pertinent bit of the output. How can pe2800oracle-sd run between the Pruned and Recycled messages? 01-Aug 22:03 pe2800oracle-dir: Start Backup JobId 487, Job=thepig.2005-08-01_20.00.01 01-Aug 22:13 pe2800oracle-dir: Pruned 5 Jobs on Volume montag2 from catalog. 01-Aug 22:13 pe2800oracle-sd: Job thepig.2005-08-01_20.00.01 waiting. Cannot find any appendable volumes. Please use the label command to create a new Volume for: Storage: LTO Ultrium 2 Media type: LTO-G2 Pool: Default 01-Aug 23:13 pe2800oracle-dir: Recycled volume montag2 __Martin good morning could it be that automatic pruning takes too long time? i am using a sqlite-db and pruning of a used volume takes at least 30 minutes. if i prune the volume manual before schedule then the backup works. rico --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Re: Exabyte EZ-17 Loader Problem
On Wednesday 03 August 2005 01:09, George R. Kasica wrote: On Tue, 02 Aug 2005 13:20:13 -0500, you wrote: Hello: I'm still working on getting Bacula running here and have it working well with 2 of 3 tape drives (A Pair of HP DD3 Autochangers C1553A) however I'm having something of a problem with the 3rd device, an Exabyte EZ-17 unit. I can perform the btape test and auto perfectly, however I get in trouble when working through the test sequence on p.296 of the 1.36.3 manual as shown below, the unload fails. What do I need to do to make this function correctly? There is a tape in the drive from slot 1 which is why its not showing. [EMAIL PROTECTED] bin]# ./mtx-changer /dev/sg0 list 0 /dev/nst0 0 2: 3: 4: 5: 6: 7: [EMAIL PROTECTED] bin]# ./mtx-changer /dev/sg0 slots 0 /dev/nst0 0 7 [EMAIL PROTECTED] bin]# ./mtx-changer /dev/sg0 unload Unloading Data Transfer Element into Storage Element 1...mtx: Request Sense: Long Report=yes mtx: Request Sense: Valid Residual=no mtx: Request Sense: Error Code=70 (Current) mtx: Request Sense: Sense Key=Illegal Request mtx: Request Sense: FileMark=no mtx: Request Sense: EOM=no mtx: Request Sense: ILI=no mtx: Request Sense: Additional Sense Code = 3B mtx: Request Sense: Additional Sense Qualifier = 90 mtx: Request Sense: BPV=no mtx: Request Sense: Error in CDB=no mtx: Request Sense: SKSV=no mtx: MOVE MEDIUM from Element Address 82 to 1 Failed At this point the drive is unresponsive to further commands from mt for example mt -f /dev/nst0 offline fails as does anything else. I'm guessing the unit needs a power cycle to reset at this point however I'm not physically in front of it to tell at this time, it still reports the following mtx status: [EMAIL PROTECTED] ~]# mtx -f /dev/sg0 status Storage Changer /dev/sg0:1 Drives, 7 Slots ( 0 Import/Export ) Data Transfer Element 0:Full (Storage Element 1 Loaded) Storage Element 1:Empty Storage Element 2:Full Storage Element 3:Full Storage Element 4:Full Storage Element 5:Full Storage Element 6:Full Storage Element 7:Full tapeinfo -f /dev/sg0 Product Type: Medium Changer Vendor ID: 'EXABYTE ' Product ID: 'EXB-210 ' Revision: '1.06' Attached Changer: No SerialNumber: '38002496 ' SCSI ID: 0 SCSI LUN: 0 Ready: yes [EMAIL PROTECTED] ~]# loaderinfo -f /dev/sg0 Product Type: Medium Changer Vendor ID: 'EXABYTE ' Product ID: 'EXB-210 ' Revision: '1.06' Attached Changer: No Bar Code Reader: No EAAP: Yes Number of Medium Transport Elements: 1 Number of Storage Elements: 7 Number of Import/Export Element Elements: 0 Number of Data Transfer Elements: 1 Transport Geometry Descriptor Page: Yes Invertable: No Device Configuration Page: Yes Can Transfer: Yes [EMAIL PROTECTED] ~]# tapeinfo -f /dev/sg5 Product Type: Tape Drive Vendor ID: 'EXABYTE ' Product ID: 'EXB-89008E00012F' Revision: 'V39e' Attached Changer: No SerialNumber: '0060089150' MinBlock:1 MaxBlock:245760 SCSI ID: 6 SCSI LUN: 0 Ready: yes BufferedMode: yes Medium Type: 0xd2 Density Code: 0x27 BlockSize: 0 DataCompEnabled: yes DataCompCapable: yes DataDeCompEnabled: yes CompType: 0x10 DeCompType: 0x10 Block Position: 2 [EMAIL PROTECTED] ~]# [EMAIL PROTECTED] ~]# loaderinfo -f /dev/sg5 Product Type: Tape Drive Vendor ID: 'EXABYTE ' Product ID: 'EXB-89008E00012F' Revision: 'V39e' Attached Changer: No Bar Code Reader: No EAAP: No Transport Geometry Descriptor Page: No Device Configuration Page: No Well, I've partially solved my own problem but don't know how to get Bacula to do it automatically. The working sequence of load and eject is as follows: mtx -f /dev/sg0 load 1 mt -f /dev/nst0 rewind (OR WHATEVER YOU'RE DOING) mt -f /dev/nst0 offline mtx -f /dev/sh0 unload 1 The catch seems to be needing the offline command, but how do I get that into the bacula changer script and where??? If I am not mistaken, the script *shows* you exactly where it should go. Try looking at the script. -- Best regards, Kern ( /\ V_V --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] mtx-changer script with Solaris 10
I don't know if someone has already posted this but
I recently found that the mtx-changer script was
failing on my Solaris 10 machines because of the
seq command in line 45 not included in the Solaris 10
base system.
The workaround is quite easy and goes like this :
wait_for_drive() {
# for i in $(seq 180); do # Wait max 180 seconds
i=0
while [ $i -ne 180 ]; do
i=`expr $i + 1`
if mt -f $1 status | grep ONLINE /dev/null 21; then
break
fi
# echo Device $1 - not ready, retrying...
sleep 1
done
}
Maybe it could be nice to include it in comment,
like the one about BSD
--
-- Sam
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Automatically mount the tape for next backup
Hi,
I have installed Bacula 1.36.2 on Debian Sarge dealing with an LTO2 drive.
I have 7 Jobs to run each night and I ant to change the tape after each
full backup that occurs each night.
I use this option in order to mark the tape as Used when the backup is
finished :
Maximum Volume Jobs = 7
But, when I insert a new tape, I have to mount the tape so that the next
backup can run.
Is it possible to have a fully automatic configuration in order to avoid
the need to explicitly mount the tape through bconsole ?
Extract from bacula-sd.conf :
**
Device {
Name = LTO2
Media Type = LTO2
Archive Device = /dev/nst0AutomaticMount =
yes; AlwaysOpen = yes;
RemovableMedia = yes;
RandomAccess = no;
}
**
Thanks in advance,
Sébastien
hi
its easy.
make an admin job that runs first(priority).
in this admin job write this:
RunBeforeJob = /etc/bacula/before_backup
in /etc/bacula/before_backup(make executable) write this:
#! /bin/sh
bconsole -c /etc/bacula/bconsole.conf END_OF_DATA
@output /dev/null
mount storage=LTO-G2
quit
END_OF_DATA
exit 0
#EOF
or u can insert RunBeforeJob = /etc/bacula/before_backup in you first
backup-job.
It works !
Thanks again,
Sébastien
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77alloc_id492op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] umount blocks further usage
On Fri, 29 Jul 2005, Jesse Keating wrote: On Fri, 2005-07-29 at 21:49 +0100, Martin Simmons wrote: Oh, I thought the blockage and mount command is per storage device, not per tape. It does block the storage device. It prevents the storage device from automounting the next usable tape. I have automated marking the last tape as 'Full', and I want to further automate unmounting it and unloading it from the tape drive so that when I come in on Monday I can just eject the magazine and remove the used tapes, shuffle in tapes, reinsert the magazine and update slots. Run update slots as an admin job. it will (currently) park the tapes. As this is disruptive if there are jobs in progress, I believe that it won't park them in future versions. As it stands, it looks like I can automate the umounting and unloading, but I'll manually have to load a drive and mount it from bconsole to clear the block flag. No. Once the drive is cleared and the right tape is available, a mount command will cause Bacula to load the appropriate tapae from the magazine. That's how I do this by hand at present. --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] deleted database
On Mon, 1 Aug 2005, Ferdinando Pasqualetti wrote: I think you should recreate an empty database (using the scripts in /etc/bacula) and the bscan on all your backups tha should reconstruct your jobs. Once the database is reconstructed you should be able to restore the last database backup Running bscan again after that point would find jobs/files saved since the database was last backed up. --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Show next tape
On Tue, 2 Aug 2005, Georg Lutz wrote: 1. We life in a part of Germany where there are (fourtunately ? ;-)) a lot of holidays. We moved our main backup and tape change days to mid-week to deal with this problem AB --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Re: Exabyte EZ-17 Loader Problem
On Tue, 02 Aug 2005 13:20:13 -0500, you wrote: Hello: I'm still working on getting Bacula running here and have it working well with 2 of 3 tape drives (A Pair of HP DD3 Autochangers C1553A) however I'm having something of a problem with the 3rd device, an Exabyte EZ-17 unit. I can perform the btape test and auto perfectly, however I get in trouble when working through the test sequence on p.296 of the 1.36.3 manual as shown below, the unload fails. What do I need to do to make this function correctly? There is a tape in the drive from slot 1 which is why its not showing. [EMAIL PROTECTED] bin]# ./mtx-changer /dev/sg0 list 0 /dev/nst0 0 2: 3: 4: 5: 6: 7: [EMAIL PROTECTED] bin]# ./mtx-changer /dev/sg0 slots 0 /dev/nst0 0 7 [EMAIL PROTECTED] bin]# ./mtx-changer /dev/sg0 unload Unloading Data Transfer Element into Storage Element 1...mtx: Request Sense: Long Report=yes mtx: Request Sense: Valid Residual=no mtx: Request Sense: Error Code=70 (Current) mtx: Request Sense: Sense Key=Illegal Request mtx: Request Sense: FileMark=no mtx: Request Sense: EOM=no mtx: Request Sense: ILI=no mtx: Request Sense: Additional Sense Code = 3B mtx: Request Sense: Additional Sense Qualifier = 90 mtx: Request Sense: BPV=no mtx: Request Sense: Error in CDB=no mtx: Request Sense: SKSV=no mtx: MOVE MEDIUM from Element Address 82 to 1 Failed At this point the drive is unresponsive to further commands from mt for example mt -f /dev/nst0 offline fails as does anything else. I'm guessing the unit needs a power cycle to reset at this point however I'm not physically in front of it to tell at this time, it still reports the following mtx status: [EMAIL PROTECTED] ~]# mtx -f /dev/sg0 status Storage Changer /dev/sg0:1 Drives, 7 Slots ( 0 Import/Export ) Data Transfer Element 0:Full (Storage Element 1 Loaded) Storage Element 1:Empty Storage Element 2:Full Storage Element 3:Full Storage Element 4:Full Storage Element 5:Full Storage Element 6:Full Storage Element 7:Full tapeinfo -f /dev/sg0 Product Type: Medium Changer Vendor ID: 'EXABYTE ' Product ID: 'EXB-210 ' Revision: '1.06' Attached Changer: No SerialNumber: '38002496 ' SCSI ID: 0 SCSI LUN: 0 Ready: yes [EMAIL PROTECTED] ~]# loaderinfo -f /dev/sg0 Product Type: Medium Changer Vendor ID: 'EXABYTE ' Product ID: 'EXB-210 ' Revision: '1.06' Attached Changer: No Bar Code Reader: No EAAP: Yes Number of Medium Transport Elements: 1 Number of Storage Elements: 7 Number of Import/Export Element Elements: 0 Number of Data Transfer Elements: 1 Transport Geometry Descriptor Page: Yes Invertable: No Device Configuration Page: Yes Can Transfer: Yes [EMAIL PROTECTED] ~]# tapeinfo -f /dev/sg5 Product Type: Tape Drive Vendor ID: 'EXABYTE ' Product ID: 'EXB-89008E00012F' Revision: 'V39e' Attached Changer: No SerialNumber: '0060089150' MinBlock:1 MaxBlock:245760 SCSI ID: 6 SCSI LUN: 0 Ready: yes BufferedMode: yes Medium Type: 0xd2 Density Code: 0x27 BlockSize: 0 DataCompEnabled: yes DataCompCapable: yes DataDeCompEnabled: yes CompType: 0x10 DeCompType: 0x10 Block Position: 2 [EMAIL PROTECTED] ~]# [EMAIL PROTECTED] ~]# loaderinfo -f /dev/sg5 Product Type: Tape Drive Vendor ID: 'EXABYTE ' Product ID: 'EXB-89008E00012F' Revision: 'V39e' Attached Changer: No Bar Code Reader: No EAAP: No Transport Geometry Descriptor Page: No Device Configuration Page: No Well, I've partially solved my own problem but don't know how to get Bacula to do it automatically. The working sequence of load and eject is as follows: mtx -f /dev/sg0 load 1 mt -f /dev/nst0 rewind (OR WHATEVER YOU'RE DOING) mt -f /dev/nst0 offline mtx -f /dev/sh0 unload 1 The catch seems to be needing the offline command, but how do I get that into the bacula changer script and where??? George --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Restore doesn't restore files
Hi list, I'm happily using bacula 1.36.2 on a Debian sarge machine. The backup/restore was running great for weeks on a clone machine. Recently I changed to a HP server with the same Ultrium streamer. Again the backup is running great, but when I try to restore it appears to go okay but there aren't any files written. The output of the restore job is (cut) : 02-Aug 17:13 bacula-sd: Ready to read from volume Friday1 on device /dev/nst0. 02-Aug 17:13 bacula-sd: Forward spacing to file:block 6:0. server1-fd: -rwxrwxrwx 1 00 749056 2004-05-03 17:34:16 +/brs/E//Goldmine/relatie/Contact1.DBT server1-fd: -rwxrwxrwx 1 00 785408 2004-05-03 17:34:23 +/brs/E//Goldmine/relatie/Contact2.MDX server1-fd: -rwxrwxrwx 1 008327832 2004-05-03 17:34:23 +/brs/E//Goldmine/relatie/ContGrps.DBF server1-fd: -rwxrwxrwx 1 00 11201536 2004-05-03 17:34:25 +/brs/E//Goldmine/relatie/ContGrps.MDX 02-Aug 17:17 bacula-sd: End of Volume at file 6 on device /dev/nst0, Volume +Friday1 02-Aug 17:17 bacula-sd: End of all volumes. server1-fd: -rwxrwxrwx 1 00 10546 2004-05-03 17:35:13 +/brs/E//Goldmine/relatie/ContUDef.DBF server1-fd: -rwxrwxrwx 1 00 10240 2004-05-03 17:35:13 +/brs/E//Goldmine/relatie/ContUDef.MDX 02-Aug 17:18 bacula-dir: Bacula 1.36.2 (28Feb05): 02-Aug-2005 17:18:47 JobId: 119 Job:Restore-sv1.2005-08-02_17.09.22 Client: server1-fd Start time: 02-Aug-2005 17:09:24 End time: 02-Aug-2005 17:18:47 Files Expected: 17 Files Restored: 17 Bytes Restored: 242,580,387 Rate: 430.9 KB/s FD Errors: 0 FD termination status: OK SD termination status: OK Termination:Restore OK But after that no files show up in brs. I created the brs dir as a test. The restorejob defines /tmp/bacula-restore The /brs dir has system wide read/write rights. Also running bconsole as user or root doesn't matter. I backup with signature = MD5, the files come from a windows machine. Any idea what is going wrong ? Regards, Joost -- Things you don't want your SysAdmin to say -32. Why is my rm -R * taking so long? --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] umount blocks further usage
On Monday 01 August 2005 12:26, Alan Brown wrote: On Fri, 29 Jul 2005, Jesse Keating wrote: SO really, all I need to do after marking the tape as full is issue a update slots command. Easy sleasy. Of course if Kern ever changes how this works I may be in trouble as this is not the expected use of update slots (: It pays to issue mount commands too, just in case Update slots functionality is likely to change. This is why I've been asking for a separate command to park tapes. Yes, it has not changed yet, but it *is* likely to change. If you want a particular command before it is changed, then please send a RFC to the bacula-devel list with your command specification -- please keep in mind an Autochanger may have multiple drives, and the user may not have names in his Director's conf file that can refer to each of those drives, so in any RFC this issue must be addressed. --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users -- Best regards, Kern ( /\ V_V --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] mtx-changer script with Solaris 10
On Wednesday 03 August 2005 11:51, Samuel Olampi wrote:
I don't know if someone has already posted this but
I recently found that the mtx-changer script was
failing on my Solaris 10 machines because of the
seq command in line 45 not included in the Solaris 10
base system.
The workaround is quite easy and goes like this :
wait_for_drive() {
# for i in $(seq 180); do # Wait max 180 seconds
i=0
while [ $i -ne 180 ]; do
i=`expr $i + 1`
if mt -f $1 status | grep ONLINE /dev/null 21; then
break
fi
# echo Device $1 - not ready, retrying...
sleep 1
done
}
Maybe it could be nice to include it in comment,
like the one about BSD
It seems to me that this was changed quite some time ago in the sense you
indicate. You might take a look at the one in the CVS to be sure so I can
get it right for the 1.38 release.
--
Best regards,
Kern
(
/\
V_V
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Running storage daemon and director daemon on different machines
On Wednesday 03 August 2005 14:31, [EMAIL PROTECTED] wrote: Hi, I already have bacula installed on a two machine system and am very pleased with it's operation and performance - great job! I'd like to use it in different installation where the director and storage daemons reside on different machines. I'm using Redhat Enterpise Linux AS 2.1 and 3 (mixed, will eventually all be 3). However the existing RPM structure doesn't really allow for separate director / storage daemon installations. I realise the configuration files need to change but has anyone performed this on other systems before? Have they come across any difficulties I should be aware of? Is there already a way of repackaging the RPMS so that installation could be made easier, or should I look at modifying the SPEC files myself and then re-contribute them to the group? Before doing anything yourself, you probably should discuss this with Scott Barninger who listens on the bacula-devel list, and is also at the copy address above. Thanks, Best Regards, Brett Delle Grazie -- Best regards, Kern ( /\ V_V --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] status director and pruning
Kern Sibbald wrote:
On Wednesday 03 August 2005 14:18, Russell Howe wrote:
Should it be possible (via 'AutoPrune = no' options in all the right
places, together with 'Prune {Job,Volume,File}s = no') to stop status
director from doing a prune operation when it lists which tapes it
thinks it will need for upcoming jobs?
In principle, yes. If you can find a case where it goes off and prune (quite
possible since it is not a job as such), then I'll be happy to fix it.
It seems to prune every time for me. I'll take a look at the database
and check what the code's doing... IIRC it was a fairly simple sequence
of steps that ran when status director was issued. I started looking
before, but got sidetracked...
--
Russell Howe
[EMAIL PROTECTED]
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Re: Exabyte EZ-17 Loader Problem
On Wednesday 03 August 2005 00:32, Kern Sibbald wrote: On Wednesday 03 August 2005 01:09, George R. Kasica wrote: Well, I've partially solved my own problem but don't know how to get Bacula to do it automatically. The working sequence of load and eject is as follows: mtx -f /dev/sg0 load 1 mt -f /dev/nst0 rewind (OR WHATEVER YOU'RE DOING) mt -f /dev/nst0 offline mtx -f /dev/sh0 unload 1 The catch seems to be needing the offline command, but how do I get that into the bacula changer script and where??? If I am not mistaken, the script *shows* you exactly where it should go. Try looking at the script. At line 102 of mtx-changer (version 1.11.2.1, Bacula 1.36.1) there is a pair of lines that say: # enable the following line if you need to eject the cartridge mt -f $device offline I have it enabled in mine because I need the offline command. You should be able to add the rewind command right above it. Although, the offline command should perform a rewind, so I'm not sure if it is needed. j- k- -- Joshua Kugler CDE System Administrator http://distance.uaf.edu/ --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Setup question
Samuel Olampi wrote:
Thanks for your answer, Phil
Phil Stracchino wrote:
1) can Bacula do one part of the backup on tape (lev.0 + lev.1)
and the other on disks (lev.2) or do all the jobs need to be run on the
same storage device ?
This is fine.
Can you explain to me how do you split your backup between
tapes (l.0+1) and disks (l.2) ??
Do you do it in the schedule / pool section ?
Any help is appreciated !
This is simple. Create two pools, one of disk volumes, one of tape
volumes; we'll call the pools DISK and TAPE for clarity. Then, in your
job definitions, do something this:
JobDefs {
Name = Split Media
Type = Backup
...
Full Backup Pool = TAPE
Differential Backup Pool = TAPE
Incremental Backup Pool = DISK
...
}
This, as far as I know, should work just fine, so long as the DISK and
TAPE pools are controlled by the same storage daemon. (There is no way,
to my knowledge, to specify different storage daemons for different levels.)
Kern, how hard would it be to add a {Full,Differential,Incremental}
Backup Storage override directive to allow different Storage daemons
for different job levels, as well as different Pools?
--
Phil Stracchino [EMAIL PROTECTED]
Renaissance Man, Unix generalist, Perl hacker
Mobile: 603-216-7037 Landline: 603-886-3518
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Pool overrides
I would think a pool specified on a Schedule resource's Run line should
override the pool specified in the Job resource. The problem is that
only Pool= can be specified as a Run option. So when a Run line
specifies a pool and the job level is upgraded to full from differential
or incremental, the pool specified on the Run line still overrides the
Job resource specification. I think this is incorrect. For a job started
by a Run line with Level=Differential, when the level is upgraded to
Full, any Pool= specified on the Run line should be considered to be a
DifferentialPool= override, not a Pool= override.
For example, given:
Job {
...
Pool = full-pool
}
Schedule {
Run = Level=Full 1st fri at 01:01
Run = Level=Incremental Pool=incremental-pool daily at 04:01
}
A job started by the second Run line should do a incremental backup to
pool incremental-pool. If no existing full backup is found in the
catalog then, with the current pool selection scheme, a full backup will
be written to pool incremental-pool. I would rather see the pool
selection start over from the beginning, taking the new run level into
consideration and also considering any Pool= specified on the Run line
to override only the DifferentialPool= or IncrementalPool= specified in
the Job resource, not the Pool=.
A pool selection scheme might be:
1. If Pool is not specified then set Pool = default pool
2. If DifferentialPool is not specified then set DifferentialPool = Pool
3. If IncrementalPool is not specified then set IncrementalPool = Pool
4. If FullPool is not specified then set FullPool = Pool
5. If Run line has no Pool= override then goto 9
6. If Run has Level=Full then set FullPool = pool from Run's Pool= override
7. If Run has Level=Differential then set DifferentialPool = pool from
Run's Pool= override
8. If Run has Level=Incremental then set IncrementalPool = pool from
Run's Pool= override
9. Set pool to value of FullPool, DifferentialPool, or IncrementalPool
based on actual current run level
Josh Fisher
Kern Sibbald wrote:
Hello,
From time to time, I've heard some complaints about the way Bacula selects
pools, but it has never been specific enough for me to do something. Now, I
have an example where I have defined an Incremental Pool, and in using the
run command, even though the job is an Incremental job, the pool show is
the Full Pool.
In looking at the code, I see the following:
1. When (if) a job is scheduled, the Run pool overrides are applied to the
pool.
2. If you manually run a job, it starts here.
3. When the backup is initialized, the Job pool specifications are applied,
if any.
4. If the backup is Diff or Inc, and no Full backup is found in the catalog,
the job is upgraded to Full.
Now, I haven't looked at this in detail, but it seems to me that at least in
1.37, the Job pool specifications are not working as I intended, since step 4
should logically be done before step 3.
Also, I haven't checked the documentation, but it should clearly indicate that
Job Pool specifications (other than the Pool = ) override any run overrides.
Comments would be appreciated as I would like to fix this problem.
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Adding a Red Hat box to be backed up
On Wednesday 03 August 2005 11:42 am, Stack Stack wrote: Thanks, That at least fixed the problem on the Debian box. However, still the same error on the Red Hat box. I reinstalled the fd client on the Red Hat box. I also went through every bacula file that I could find on both boxes, and verified that the password was set properly. Still the same error on the Red Hat box. Any suggestions? I am beginning to think that the password isn't the problem, and that I have some other setting messed up. Thanks for the help, ~Stack~ On 8/3/05, Danie Theron [EMAIL PROTECTED] wrote: Stack Stack wrote: Hey Guys, Well I have Bacula running on my Debian box now. I have been toying with it for a few hours, and have figured out a lot. I got different folders, files, and drives being backed up. So I decided to test out adding another box. I am attempting to add a Red Hat 9 box. I installed the Client rpm, and setup the config files on both machines as I understand they should be according to the tutorial file. When I run a backup from the Debian box, I get this error: ERROR in authenticate.c:217 Unable to authenticate console *UserAgent* at client:/*IP addy*/:36131. Not certain what to do on that. When I run ./bconsole from the Red Hat box I get this error: Connecting to Director /*My IP*/:9102 Director authorization problem. Most likely the passwords do not agree. Please see http://www.bacula.org/html-manual/faq.html#AuthorizationErrors for help. ERR=1999 Authorization failed. I checked the passwords, they are correct. I checked the port numbers, they are correct. So I went to the website, and its a dead link. I couldn't find any errors listed as 1999, so I am quite baffled as to what is going on. If I left something out, let me know. I appreciate any help. Thanks Guys, ~Stack~ You might want to add the RH box to your hosts file /etc/hosts. Hope this helps If the problem was solved by adding the RH system name to your /etc/hosts file on the Debian system, then the problem was a DNS resolver problem. Have you tried adding the Debian system name to the /etc/hosts file on your RH system? You're likely having the same DNS resolver problem there, as well. cmr -- Debian 'Sarge': Registered Linux User #241964 More laws, less justice. -- Marcus Tullius Ciceroca, 42 BC --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Adding a Red Hat box to be backed up
Once I saw that it worked on one system(Debian), I made sure that it was implemented on the other (Red Hat). I am going to attempt a client install on another system. Hopefully this will point out which box is having the problem. Thanks for the suggestion, ~Stack~ On 8/3/05, Mike Reinehr [EMAIL PROTECTED] wrote: On Wednesday 03 August 2005 11:42 am, Stack Stack wrote: Thanks, That at least fixed the problem on the Debian box. However, still the same error on the Red Hat box. I reinstalled the fd client on the Red Hat box. I also went through every bacula file that I could find on both boxes, and verified that the password was set properly. Still the same error on the Red Hat box. Any suggestions? I am beginning to think that the password isn't the problem, and that I have some other setting messed up. Thanks for the help, ~Stack~ On 8/3/05, Danie Theron [EMAIL PROTECTED] wrote: Stack Stack wrote: Hey Guys, Well I have Bacula running on my Debian box now. I have been toying with it for a few hours, and have figured out a lot. I got different folders, files, and drives being backed up. So I decided to test out adding another box. I am attempting to add a Red Hat 9 box. I installed the Client rpm, and setup the config files on both machines as I understand they should be according to the tutorial file. When I run a backup from the Debian box, I get this error: ERROR in authenticate.c:217 Unable to authenticate console *UserAgent* at client:/*IP addy*/:36131. Not certain what to do on that. When I run ./bconsole from the Red Hat box I get this error: Connecting to Director /*My IP*/:9102 Director authorization problem. Most likely the passwords do not agree. Please see http://www.bacula.org/html-manual/faq.html#AuthorizationErrors for help. ERR=1999 Authorization failed. I checked the passwords, they are correct. I checked the port numbers, they are correct. So I went to the website, and its a dead link. I couldn't find any errors listed as 1999, so I am quite baffled as to what is going on. If I left something out, let me know. I appreciate any help. Thanks Guys, ~Stack~ You might want to add the RH box to your hosts file /etc/hosts. Hope this helps If the problem was solved by adding the RH system name to your /etc/hosts file on the Debian system, then the problem was a DNS resolver problem. Have you tried adding the Debian system name to the /etc/hosts file on your RH system? You're likely having the same DNS resolver problem there, as well. cmr -- Debian 'Sarge': Registered Linux User #241964 More laws, less justice. -- Marcus Tullius Ciceroca, 42 BC --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Adding a Red Hat box to be backed up
On Wednesday 03 August 2005 01:47 pm, Stack Stack wrote: Once I saw that it worked on one system(Debian), I made sure that it was implemented on the other (Red Hat). I am going to attempt a client install on another system. Hopefully this will point out which box is having the problem. Thanks for the suggestion, ~Stack~ Ok, but, while we're on the subject of networking have you tried the other obvious things like pinging the respective systems, one from the other, just to verify basic connectivity. Also, do you have any firewall software running on either system, which might be blocking ports? I did a new Libranet 3.0 install a few months ago was pulling my hair out when I couldn't connect to anything until I finally stumbled over the fact that Libranet automagically installs starts a firewall during a standard install. Cheers! cmr On 8/3/05, Mike Reinehr [EMAIL PROTECTED] wrote: On Wednesday 03 August 2005 11:42 am, Stack Stack wrote: Thanks, That at least fixed the problem on the Debian box. However, still the same error on the Red Hat box. I reinstalled the fd client on the Red Hat box. I also went through every bacula file that I could find on both boxes, and verified that the password was set properly. Still the same error on the Red Hat box. Any suggestions? I am beginning to think that the password isn't the problem, and that I have some other setting messed up. Thanks for the help, ~Stack~ On 8/3/05, Danie Theron [EMAIL PROTECTED] wrote: Stack Stack wrote: Hey Guys, Well I have Bacula running on my Debian box now. I have been toying with it for a few hours, and have figured out a lot. I got different folders, files, and drives being backed up. So I decided to test out adding another box. I am attempting to add a Red Hat 9 box. I installed the Client rpm, and setup the config files on both machines as I understand they should be according to the tutorial file. When I run a backup from the Debian box, I get this error: ERROR in authenticate.c:217 Unable to authenticate console *UserAgent* at client:/*IP addy*/:36131. Not certain what to do on that. When I run ./bconsole from the Red Hat box I get this error: Connecting to Director /*My IP*/:9102 Director authorization problem. Most likely the passwords do not agree. Please see http://www.bacula.org/html-manual/faq.html#AuthorizationErrors for help. ERR=1999 Authorization failed. I checked the passwords, they are correct. I checked the port numbers, they are correct. So I went to the website, and its a dead link. I couldn't find any errors listed as 1999, so I am quite baffled as to what is going on. If I left something out, let me know. I appreciate any help. Thanks Guys, ~Stack~ You might want to add the RH box to your hosts file /etc/hosts. Hope this helps If the problem was solved by adding the RH system name to your /etc/hosts file on the Debian system, then the problem was a DNS resolver problem. Have you tried adding the Debian system name to the /etc/hosts file on your RH system? You're likely having the same DNS resolver problem there, as well. cmr -- Debian 'Sarge': Registered Linux User #241964 More laws, less justice. -- Marcus Tullius Ciceroca, 42 BC --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ 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 --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf
Re: [Bacula-users] Adding a Red Hat box to be backed up
I do not have a firewall running on either system at this time, and am able to connect to each system by ssh. I mentioned before that these were software test boxes before we made any drastic changes to the production machines. Well one of the tests we were running, just borked up the RH box. On the bright side this will give me a chance to test a clean install. Maybe my problems are from something completely unrelated. I appreciate the suggestions, I will post again if I have a problem on the box. Thanks, ~Stack~ On 8/3/05, Mike Reinehr [EMAIL PROTECTED] wrote: On Wednesday 03 August 2005 01:47 pm, Stack Stack wrote: Once I saw that it worked on one system(Debian), I made sure that it was implemented on the other (Red Hat). I am going to attempt a client install on another system. Hopefully this will point out which box is having the problem. Thanks for the suggestion, ~Stack~ Ok, but, while we're on the subject of networking have you tried the other obvious things like pinging the respective systems, one from the other, just to verify basic connectivity. Also, do you have any firewall software running on either system, which might be blocking ports? I did a new Libranet 3.0 install a few months ago was pulling my hair out when I couldn't connect to anything until I finally stumbled over the fact that Libranet automagically installs starts a firewall during a standard install. Cheers! cmr On 8/3/05, Mike Reinehr [EMAIL PROTECTED] wrote: On Wednesday 03 August 2005 11:42 am, Stack Stack wrote: Thanks, That at least fixed the problem on the Debian box. However, still the same error on the Red Hat box. I reinstalled the fd client on the Red Hat box. I also went through every bacula file that I could find on both boxes, and verified that the password was set properly. Still the same error on the Red Hat box. Any suggestions? I am beginning to think that the password isn't the problem, and that I have some other setting messed up. Thanks for the help, ~Stack~ On 8/3/05, Danie Theron [EMAIL PROTECTED] wrote: Stack Stack wrote: Hey Guys, Well I have Bacula running on my Debian box now. I have been toying with it for a few hours, and have figured out a lot. I got different folders, files, and drives being backed up. So I decided to test out adding another box. I am attempting to add a Red Hat 9 box. I installed the Client rpm, and setup the config files on both machines as I understand they should be according to the tutorial file. When I run a backup from the Debian box, I get this error: ERROR in authenticate.c:217 Unable to authenticate console *UserAgent* at client:/*IP addy*/:36131. Not certain what to do on that. When I run ./bconsole from the Red Hat box I get this error: Connecting to Director /*My IP*/:9102 Director authorization problem. Most likely the passwords do not agree. Please see http://www.bacula.org/html-manual/faq.html#AuthorizationErrors for help. ERR=1999 Authorization failed. I checked the passwords, they are correct. I checked the port numbers, they are correct. So I went to the website, and its a dead link. I couldn't find any errors listed as 1999, so I am quite baffled as to what is going on. If I left something out, let me know. I appreciate any help. Thanks Guys, ~Stack~ You might want to add the RH box to your hosts file /etc/hosts. Hope this helps If the problem was solved by adding the RH system name to your /etc/hosts file on the Debian system, then the problem was a DNS resolver problem. Have you tried adding the Debian system name to the /etc/hosts file on your RH system? You're likely having the same DNS resolver problem there, as well. cmr -- Debian 'Sarge': Registered Linux User #241964 More laws, less justice. -- Marcus Tullius Ciceroca, 42 BC --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement *
[Bacula-users] Problems with Exclude?
i've tried to build
a fileset that excludes directories based on wild card, but I don't believe it's
working. Tried looking in the manual again, but as far as I see it's
correct.
Bacula Version:
1.36.3
FileSet
setting:
FileSet
{Name = "Voicemail Greetings"Include
{Options { signature=MD5
verify=5
compression=GZIPexclude=yes
wilddir=INBOX* }File =
/var/spool/VOIP/voicemail/default}}
The above fileset
calls a directory that contains all the phone numbers on a system as
directories. In each directory there's an INBOX directory as well as several
files and other directories. We do not wish to save voicemails (contained in
INBOX), only the other files.
Any
ideas?
Sherwood
McGowan
Re: [Bacula-users] umount blocks further usage
On Wed, 2005-08-03 at 12:55 -0700, Jesse Keating wrote: So release does not unload the drive, so I would have to pair that with an mtx call to unload. This I'd rather not do, but I suppose I should. Hrm, release doesn't close the /dev/nst0 device either. update slots leaves the device as closed or non-existent. Will this pose a problem if I unload the tape out from under bacula-sd while the device is open? Testing here.. -- Jesse Keating GameHouse -- Systems Engineer --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Bacula Crashing suddenly
no recent changes other than putting doublequotes around the Wilddir statement as previously mentioned, still using same bacula version (1.36.3) Getting the following from my CentOS 4 system (uses latest MySQL release 4.12) Using host libthread_db library "/lib/libthread_db.so.1". 0x0017a38b in waitpid () from /lib/i686/libpthread.so.0 $1 = "testing-dir", '\0' repeats 18 times $2 = 0x81b2230 "bacula-dir" $3 = 0x81b2dd0 "/sbin/bacula-dir" $4 = "MySQL" $5 = 0x80c18dc "1.36.3 (22 April 2005)" $6 = 0x80ba66b "i686-redhat-linux-gnu" $7 = 0x80ba664 "redhat" $8 = 0x80af9e5 "" #0 0x0017a38b in waitpid () from /lib/i686/libpthread.so.0 #1 0xb4b0 in ?? () #2 0x in ?? () #0 0x0017a38b in waitpid () from /lib/i686/libpthread.so.0 No symbol table info available. #1 0xb4b0 in ?? () No symbol table info available. #2 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. Any ideas why this is suddenly happening? It did it occasionally BEFORE the wilddir change as well. Sherwood McGowan
Re: [Bacula-users] umount blocks further usage
On Wed, 2005-08-03 at 13:15 -0700, Jesse Keating wrote: Hrm, release doesn't close the /dev/nst0 device either. update slots leaves the device as closed or non-existent. Will this pose a problem if I unload the tape out from under bacula-sd while the device is open? Testing here.. When I tried to do an update slots command after doing an mtx unload it paused for probably 3 to 5 minutes before actually listing the slots. Perhaps it was timing out on a drive that it thought was there. So this is not going to work as well as just doing update slots for now. -- Jesse Keating GameHouse -- Systems Engineer --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] umount blocks further usage
On Wednesday 03 August 2005 21:55, Jesse Keating wrote: On Sat, 2005-07-30 at 09:32 +0200, Kern Sibbald wrote: Try the release command. It was designed to release the drive but not to block it as the unmount command does. You may need to explicitly do an unload of the drive using mtx directly. If that is the case, it might be an interesting RFC for Bacula to do that when it is dealing with an autochanger. As far as I know, no one is using release so it may or may not work. So release does not unload the drive, so I would have to pair that with an mtx call to unload. This I'd rather not do, but I suppose I should. I'll work on drafting an RFC to handle non-blocking unloading of the drive for the purpose of being able to do magazine maint at convenience but not block backups if the tape system isn't gotten to within the day. Unless I am missing something, there is no such thing as a nonblocking unloading of a drive. I know you're working on support for multi-drive tape loaders. How would these be referenced differently than just a single storage device? Now one can say 'unload storage=Tape' with drive stuff, would it be something like 'unload storage=Tape drive=1' ? This is in the doc, though not explicitly. -- Best regards, Kern ( /\ V_V --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula Crashing suddenly
On Wednesday 03 August 2005 23:02, Sherwood McGowan wrote: no recent changes other than putting doublequotes around the Wilddir statement as previously mentioned, still using same bacula version (1.36.3) Getting the following from my CentOS 4 system (uses latest MySQL release 4.12) Using host libthread_db library /lib/libthread_db.so.1. 0x0017a38b in waitpid () from /lib/i686/libpthread.so.0 $1 = testing-dir, '\0' repeats 18 times $2 = 0x81b2230 bacula-dir $3 = 0x81b2dd0 /sbin/bacula-dir $4 = MySQL $5 = 0x80c18dc 1.36.3 (22 April 2005) $6 = 0x80ba66b i686-redhat-linux-gnu $7 = 0x80ba664 redhat $8 = 0x80af9e5 #0 0x0017a38b in waitpid () from /lib/i686/libpthread.so.0 #1 0xb4b0 in ?? () #2 0x in ?? () #0 0x0017a38b in waitpid () from /lib/i686/libpthread.so.0 No symbol table info available. #1 0xb4b0 in ?? () No symbol table info available. #2 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. #0 0x in ?? () No symbol table info available. Any ideas why this is suddenly happening? It did it occasionally BEFORE the wilddir change as well. No. You will need to ensure that Bacula is built with debug info, then probably run it manually under the debugger as described in the Kaboom chapter of the manual. Unfortunately, the above traceback is not detailed enough to even start ... -- Best regards, Kern ( /\ V_V --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] RE: [Bacula-devel] Issue unloading drive in changer
Hi Chris,
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf
Of Chris Wright
Sent: Wednesday, 03 August, 2005 13:33
To: bacula-users@lists.sourceforge.net;
[EMAIL PROTECTED]
Subject: [Bacula-devel] Issue unloading drive in changer
Hopefully this info will be helpful to someone.
I'm installing bacula for the first time. I am using version
1.36.3. I have a HP A4853 DLT Library and a HP DLT7000
drive. I was able to use the changer to load tapes, get the
status, etc, but it would not unload the drive. In this
situation, I need to issue the command mt eject before I
could unload the drive. I modified the file mtx-changer to include
MT=mt
(the block for unload is now)
unload)
# echo Doing mtx -f $ctl unload $slot $drive
#
# enable the following line if you need to eject the cartridge
# mt -f $device offline
This line sould do basically the same thing.
if test x$slot = x; then
${MT} eject
${MTX} -f $ctl unload
else
$MT eject
${MTX} -f $ctl unload $slot $drive
fi
;;
Everything with the changer and drive seems to be working
now. Hope this info can help somebody and maybe be
incorporated by the developers
Chris
---
SF.Net email is Sponsored by the Better Software Conference EXPO
September 19-22, 2005 * San Francisco, CA * Development
Lifecycle Practices
Agile Plan-Driven Development * Managing Projects Teams *
Testing QA
Security * Process Improvement Measurement *
http://www.sqe.com/bsce5sf
___
Bacula-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/bacula-devel
Thanks,
Chris
---
SF.Net email is Sponsored by the Better Software Conference EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile Plan-Driven Development * Managing Projects Teams * Testing QA
Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
RE: [Bacula-users] status director and pruning
Aren't the AutoPrune and VolumeRentention fields written to the volume when
it is labelled? I seem to remember something in the manual about having to
update your volumes if you change certain attributes regarding
Pruning/Recycling/Retention.
Thanks,
Chris
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf
Of Russell Howe
Sent: Wednesday, 03 August, 2005 08:13
To: bacula-users@lists.sourceforge.net
Subject: Re: [Bacula-users] status director and pruning
Kern Sibbald wrote:
On Wednesday 03 August 2005 14:18, Russell Howe wrote:
Should it be possible (via 'AutoPrune = no' options in all the right
places, together with 'Prune {Job,Volume,File}s = no') to
stop status
director from doing a prune operation when it lists which tapes it
thinks it will need for upcoming jobs?
In principle, yes. If you can find a case where it goes
off and prune (quite
possible since it is not a job as such), then I'll be happy
to fix it.
It seems to prune every time for me. I'll take a look at the database
and check what the code's doing... IIRC it was a fairly
simple sequence
of steps that ran when status director was issued. I started looking
before, but got sidetracked...
--
Russell Howe
[EMAIL PROTECTED]
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
---
SF.Net email is Sponsored by the Better Software Conference EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile Plan-Driven Development * Managing Projects Teams * Testing QA
Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Restore job that spans more than one tape.
--- [EMAIL PROTECTED] wrote: I recently ran a backup job that spanned 4 tapes (DLT-IV) and since this is one of the first backups on this system (x86_64) I renamed the source directory and tried to do a complete restore to verify that my data is safe. When I ran the restore the job stopped at the end of the first tape stating that there were no more volumes to restore and the last file was the wrong size. What am I doing wrong? I was running bacula 1.34.6 to back up the data but this version would not restore it at all (all output files were zero bytes) so switched to version 1.36.3. Which I ran this restore job. Here is the last few lines from the gnome console: 02-Aug 20:35 dicom-sd: End of Volume at file 52 on device /dev/st0, Volume TMImages- 02-Aug 20:35 dicom-sd: TMImagesRestore.2005-08-02_16.37.46 Warning: Wrong Volume mounted on device /dev/st0: Wanted TMImages-0001 have TMImages- 02-Aug 20:35 dicom-dir: TMImagesRestore.2005-08-02_16.37.46 Fatal error: sql_get.c:823 sql_get.c:823 query SELECT MediaId,VolumeName,VolJobs,VolFiles,VolBlocks,VolBytes,VolMounts,VolErrors,VolWrites,MaxVolBytes,VolCapacityBytes,MediaType,VolStatus,PoolId,VolRetention,VolUseDuration,MaxVolJobs,MaxVolFiles,Recycle,Slot,FirstWritten,LastWritten,InChanger,EndFile,EndBlock FROM Media WHERE VolumeName='TMImages-0001' failed: ERROR: column endfile does not exist 02-Aug 20:35 dicom-sd: TMImagesRestore.2005-08-02_16.37.46 Warning: acquire.c:192 Error getting Volume info: 1997 Volume TMImages-0001 not in catalog. 02-Aug 20:35 dicom-sd: End of all volumes. 02-Aug 20:35 dicom-fd: TMImagesRestore.2005-08-02_16.37.46 Error: attribs.c:339 File size of restored file /mnt/md1/clean/images/TM/Original/Site1/Site001/4606/7004582.img not correct. Original 32559616, restored 23986176. 02-Aug 20:35 dicom-dir: TMImagesRestore.2005-08-02_16.37.46 Error: Bacula 1.36.3 (22Apr05): 02-Aug-2005 20:35:44 JobId: 719 Job: TMImagesRestore.2005-08-02_16.37.46 Client: dicom-fd Start time: 02-Aug-2005 16:37:49 End time: 02-Aug-2005 20:35:44 Files Expected: 9,315 Files Restored: 4,524 Bytes Restored: 51,707,737,523 Rate: 3622.3 KB/s FD Errors: 1 FD termination status: Error SD termination status: Error Termination:*** Restore Error *** 02-Aug 20:35 dicom-dir: message.c:462 Mail prog: bsmtp: bsmtp.c:276 Fatal connect error to localhost: ERR=Connection refused 02-Aug 20:35 dicom-dir: TMImagesRestore.2005-08-02_16.37.46 Error: message.c:473 Mail program terminated in error. CMD=/usr/sbin/bsmtp -h localhost -f (Bacula) [EMAIL PROTECTED] -s Bacula: Restore Fatal Error of dicom-fd Full [EMAIL PROTECTED] ERR=Child exited with code 1 Could this be a result of a bad upgrade of the database? When I ran the upgrade script it said that it ran seccusfully, however there were warnings that user bacula was not the data base owner. I am using a postgresql database on a second linux box. Thanks in Advance, John M. Drescher --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users It looks like the database was indeed the problem. After examining the update script and the database I determined that the database was not fully updated because of the owner problem. The user bacula was not the owner so permissions were preventing the modification of the tables. After getting the proper username / password from the database admin I was able to fix the problem. I restarted the job and it appears to be working as I just remotely logged in and it was waiting for the second tape. John --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
