-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Josh Fisher wrote: > Support wrote: >> Dear Shon >> >> I have a similar situation and solved it byusing this stargegy. >> >> Firstly the disk volumes should be treated as "Tape" drives in that only >> onle volume can be opened at a time but you can have many concurrent jobs >> writing to that volume - my setting is for a maximum of 5 concurrent jobs/ >> > The problem with concurrent jobs writing to the same volume is that job > data is interleaved on the volume. For tape, the interleaving will make > restores painfully slow. Even for disk storage it will slow things down. > I prefer to have one job at a time write to a volume so that the jobs > are placed sequentially in the volume. This would require multiple > drives for tape, but for disk it just requires multiple device > resources, each writing to a separate volume. The volumes "loaded" in > those devices are regular files and can be in the same partition or even > the same directory.
Spooling also solves this problem. Concurrent jobs are then written to tape in a sane manner. This helps with shoeshining on a tape drive too, as data is fed as fast as it can be written. >> I write data to a LaCie 1 TB disc using Firewire and get throughput of >> about 60 GB / hr. >> >> If you have several customers and can group them then one option is to >> partition your disc so that each group uses a different partition and this >> can be seen as a different "tape" drive and the jobs should write to each >> partition concurrently. >> > > They can be grouped by directories in the same partition. Partitioning > will not likely make use of the full storage capacity of the drive > because one partition will fill up first, leaving wasted space on the > others. Grouping by directory and using a single partition on the drive > allows the full capacity of the drive to be used. > >> As for full backups I use job migration - ie the jobs are written to disc >> and then migrated to tape. >> >> If you need my config files I will email them to you. >> >> Regards >> Stephen Carr >> >> Shon Stephens wrote: >> >>> I'm trying to understand job concurrency in Bacula and what strategy I >>> should use for backing up clients. >>> >>> Its likely that I will have to backup around 50 clients. My strategy is to >>> write incrementals to disk volumes and fulls to tapes. I assume its >>> possible >>> for Bacula to write to multiple disk volumes simultaneously if configured >>> to >>> do so, but what about tapes? I have my tapes divided by customer so that >>> >>> Customer1 Pool - 8 volumes, 2 in slots each week >>> Customer2 Pool - 8 volumes, 2 in slots each week >>> >>> Can Bacula write a job from 2 different clients(file daemons) in Customer1 >>> Pool at the same time? How does this work? I understand that Bacula can't >>> write a job from Customer1 Pool and Customer2 Pool at the same time >>> because >>> these are different volumes. >>> >>> Also can bacula simultaneously spool to disk and write to tape for the >>> same >>> job? >>> >>> I'm looking for good advice on how to streamline Bacula so that I'm not >>> always running backups or have to many jobs waiting on resources in use. >>> >>> Thanks, >>> Shon >>> ------------------------------------------------------------------------- >>> This SF.net email is sponsored by: Splunk Inc. >>> Still grepping through log files to find problems? Stop. >>> Now Search log events and configuration files using AJAX and a browser. >>> Download your FREE copy of Splunk now >> >>> http://get.splunk.com/_______________________________________________ >>> Bacula-users mailing list >>> Bacula-users@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/bacula-users >>> >>> >> >> >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by: Splunk Inc. >> Still grepping through log files to find problems? Stop. >> Now Search log events and configuration files using AJAX and a browser. >> Download your FREE copy of Splunk now >> http://get.splunk.com/ >> _______________________________________________ >> Bacula-users mailing list >> Bacula-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/bacula-users >> > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. > Still grepping through log files to find problems? Stop. > Now Search log events and configuration files using AJAX and a browser. > Download your FREE copy of Splunk now >> http://get.splunk.com/ > _______________________________________________ > Bacula-users mailing list > Bacula-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bacula-users - -- ---- _ _ _ _ ___ _ _ _ |Y#| | | |\/| | \ |\ | | |Ryan Novosielski - Systems Programmer II |$&| |__| | | |__/ | \| _| |[EMAIL PROTECTED] - 973/972.0922 (2-0922) \__/ Univ. of Med. and Dent.|IST/AST - NJMS Medical Science Bldg - C630 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGrjfvmb+gadEcsb4RAmmqAJ9DFjCnz2CEh7mad8hunqTmODdqRQCdExxl hShxk5ju9L3FXL6qqumnOP0= =USbX -----END PGP SIGNATURE----- ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
