+Item 23: Improve Bacula's tape and drive usage and cleaning management
+ Date: 8 November 2005, November 11, 2005
+ Origin: Adam Thornton <athornton at sinenomine dot net>,
+ Arno Lehmann <al at its-lehmann dot de>
+ Status:
+
+ What: Make Bacula manage tape life cycle information, tape reuse
+ times and drive cleaning cycles.
+
+ Why: All three parts of this project are important when operating
+ backups.
+ We need to know which tapes need replacement, and we need to
+ make sure the drives are cleaned when necessary. While many
+ tape libraries and even autoloaders can handle all this
+ automatically, support by Bacula can be helpful for smaller
+ (older) libraries and single drives. Limiting the number of
+ times a tape is used might prevent tape errors when using
+ tapes until the drives can't read it any more. Also, checking
+ drive status during operation can prevent some failures (as I
+ [Arno] had to learn the hard way...)
+
+ Notes: First, Bacula could (and even does, to some limited extent)
+ record tape and drive usage. For tapes, the number of mounts,
+ the amount of data, and the time the tape has actually been
+ running could be recorded. Data fields for Read and Write
+ time and Number of mounts already exist in the catalog (I'm
+ not sure if VolBytes is the sum of all bytes ever written to
+ that volume by Bacula). This information can be important
+ when determining which media to replace. The ability to mark
+ Volumes as "used up" after a given number of write cycles
+ should also be implemented so that a tape is never actually
+ worn out. For the tape drives known to Bacula, similar
+ information is interesting to determine the device status and
+ expected life time: Time it's been Reading and Writing, number
+ of tape Loads / Unloads / Errors. This information is not yet
+ recorded as far as I [Arno] know. A new volume status would
+ be necessary for the new state, like "Used up" or "Worn out".
+ Volumes with this state could be used for restores, but not
+ for writing. These volumes should be migrated first (assuming
+ migration is implemented) and, once they are no longer needed,
+ could be moved to a Trash pool.
+
+ The next step would be to implement a drive cleaning setup.
+ Bacula already has knowledge about cleaning tapes. Once it
+ has some information about cleaning cycles (measured in drive
+ run time, number of tapes used, or calender days, for example)
+ it can automatically execute tape cleaning (with an
+ autochanger, obviously) or ask for operator assistance loading
+ a cleaning tape.
+
+ The final step would be to implement TAPEALERT checks not only
+ when changing tapes and only sending the information to the
+ administrator, but rather checking after each tape error,
+ checking on a regular basis (for example after each tape
+ file), and also before unloading and after loading a new tape.
+ Then, depending on the drives TAPEALERT state and the known
+ drive cleaning state Bacula could automatically schedule later
+ cleaning, clean immediately, or inform the operator.
+
+ Implementing this would perhaps require another catalog change
+ and perhaps major changes in SD code and the DIR-SD protocol,
+ so I'd only consider this worth implementing if it would
+ actually be used or even needed by many people.
+
+ Implementation of these projects could happen in three distinct
+ sub-projects: Measuring Tape and Drive usage, retiring
+ volumes, and handling drive cleaning and TAPEALERTs.
+
+Item 24: Multiple threads in file daemon for the same job
+ Date: 27 November 2005
+ Origin: Ove Risberg (Ove.Risberg at octocode dot com)
+ Status:
+
+ What: I want the file daemon to start multiple threads for a backup
+ job so the fastest possible backup can be made.
+
+ The file daemon could parse the FileSet information and start
+ one thread for each File entry located on a separate
+ filesystem.
+
+ A confiuration option in the job section should be used to
+ enable or disable this feature. The confgutration option could
+ specify the maximum number of threads in the file daemon.
+
+ If the theads could spool the data to separate spool files
+ the restore process will not be much slower.
+
+ Why: Multiple concurrent backups of a large fileserver with many
+ disks and controllers will be much faster.
+
+Item 25: Archival (removal) of User Files to Tape
+ Date: Nov. 24/2005
+ Origin: Ray Pengelly [ray at biomed dot queensu dot ca
+ Status:
+
+ What: The ability to archive data to storage based on certain parameters
+ such as age, size, or location. Once the data has been written to
+ storage and logged it is then pruned from the originating
+ filesystem. Note! We are talking about user's files and not
+ Bacula Volumes.
+
+ Why: This would allow fully automatic storage management which becomes
+ useful for large datastores. It would also allow for auto-staging
+ from one media type to another.
+
+ Example 1) Medical imaging needs to store large amounts of data.
+ They decide to keep data on their servers for 6 months and then put
+ it away for long term storage. The server then finds all files
+ older than 6 months writes them to tape. The files are then removed
+ from the server.
+
+ Example 2) All data that hasn't been accessed in 2 months could be
+ moved from high-cost, fibre-channel disk storage to a low-cost
+ large-capacity SATA disk storage pool which doesn't have as quick of
+ access time. Then after another 6 months (or possibly as one
+ storage pool gets full) data is migrated to Tape.
+
+
+
+
+========== Items on put hold by Kern ============================
+
+Item h1: Split documentation
+ Origin: Maxx <maxxatworkat gmail dot com>
+ Date: 27th July 2006
+ Status: Approved, awaiting implementation
+
+ What: Split documentation in several books
+
+ Why: Bacula manual has now more than 600 pages, and looking for
+ implementation details is getting complicated. I think
+ it would be good to split the single volume in two or
+ maybe three parts:
+
+ 1) Introduction, requirements and tutorial, typically
+ are useful only until first installation time
+
+ 2) Basic installation and configuration, with all the
+ gory details about the directives supported 3)
+ Advanced Bacula: testing, troubleshooting, GUI and
+ ancillary programs, security managements, scripting,
+ etc.
+
+ Notes: This is a project that needs to be done, and will be implemented,
+ but it is really a developer issue of timing, and does not
+ needed to be included in the voting.
+
+
+Item h2: Implement support for stacking arbitrary stream filters, sinks.
+Date: 23 November 2006
+Origin: Landon Fuller <landonf@threerings.net>
+Status: Planning. Assigned to landonf.
+
+ What: Implement support for the following:
+ - Stacking arbitrary stream filters (eg, encryption, compression,
+ sparse data handling))
+ - Attaching file sinks to terminate stream filters (ie, write out
+ the resultant data to a file)
+ - Refactor the restoration state machine accordingly
+
+ Why: The existing stream implementation suffers from the following:
+ - All state (compression, encryption, stream restoration), is
+ global across the entire restore process, for all streams. There are
+ multiple entry and exit points in the restoration state machine, and
+ thus multiple places where state must be allocated, deallocated,
+ initialized, or reinitialized. This results in exceptional complexity
+ for the author of a stream filter.
+ - The developer must enumerate all possible combinations of filters
+ and stream types (ie, win32 data with encryption, without encryption,
+ with encryption AND compression, etc).
+
+ Notes: This feature request only covers implementing the stream filters/
+ sinks, and refactoring the file daemon's restoration implementation
+ accordingly. If I have extra time, I will also rewrite the backup
+ implementation. My intent in implementing the restoration first is to
+ solve pressing bugs in the restoration handling, and to ensure that
+ the new restore implementation handles existing backups correctly.
+
+ I do not plan on changing the network or tape data structures to
+ support defining arbitrary stream filters, but supporting that
+ functionality is the ultimate goal.
+
+ Assistance with either code or testing would be fantastic.
+
+ Notes: Kern: this project has a lot of merit, and we need to do it, but
+ it is really an issue for developers rather than a new feature
+ for users, so I have removed it from the voting list, but kept it
+ here, but at some point, it will be implemented.
+
+Item h3: Filesystem watch triggered backup.
+ Date: 31 August 2006
+ Origin: Jesper Krogh <jesper@krogh.cc>
+ Status:
+
+ What: With inotify and similar filesystem triggeret notification
+ systems is it possible to have the file-daemon to monitor
+ filesystem changes and initiate backup.
+
+ Why: There are 2 situations where this is nice to have.
+ 1) It is possible to get a much finer-grained backup than
+ the fixed schedules used now.. A file created and deleted
+ a few hours later, can automatically be caught.
+
+ 2) The introduced load on the system will probably be
+ distributed more even on the system.
+
+ Notes: This can be combined with configration that specifies
+ something like: "at most every 15 minutes or when changes
+ consumed XX MB".
+
+Kern Notes: I would rather see this implemented by an external program
+ that monitors the Filesystem changes, then uses the console
+
+
+Item h4: Directive/mode to backup only file changes, not entire file
+ Date: 11 November 2005
+ Origin: Joshua Kugler <joshua dot kugler at uaf dot edu>
+ Marek Bajon <mbajon at bimsplus dot com dot pl>
+ Status:
+
+ What: Currently when a file changes, the entire file will be backed up in
+ the next incremental or full backup. To save space on the tapes
+ it would be nice to have a mode whereby only the changes to the
+ file would be backed up when it is changed.
+
+ Why: This would save lots of space when backing up large files such as
+ logs, mbox files, Outlook PST files and the like.
+
+ Notes: This would require the usage of disk-based volumes as comparing
+ files would not be feasible using a tape drive.
+
+ Notes: Kern: I don't know how to implement this. Put on hold until someone
+ provides a detailed implementation plan.
+
+
+Item h5: Implement multiple numeric backup levels as supported by dump
+Date: 3 April 2006
+Origin: Daniel Rich <drich@employees.org>
+Status:
+What: Dump allows specification of backup levels numerically instead of just
+ "full", "incr", and "diff". In this system, at any given level, all
+ files are backed up that were were modified since the last backup of a
+ higher level (with 0 being the highest and 9 being the lowest). A
+ level 0 is therefore equivalent to a full, level 9 an incremental, and
+ the levels 1 through 8 are varying levels of differentials. For
+ bacula's sake, these could be represented as "full", "incr", and
+ "diff1", "diff2", etc.
+
+Why: Support of multiple backup levels would provide for more advanced backup
+ rotation schemes such as "Towers of Hanoi". This would allow better
+ flexibility in performing backups, and can lead to shorter recover
+ times.
+
+Notes: Legato Networker supports a similar system with full, incr, and 1-9 as
+ levels.
+
+Notes: Kern: I don't see the utility of this, and it would be a *huge*
+ modification to existing code.
+
+Item h6: Implement NDMP protocol support
+ Origin: Alan Davis
+ Date: 06 March 2007
+ Status:
+
+ What: Network Data Management Protocol is implemented by a number of
+ NAS filer vendors to enable backups using third-party
+ software.
+
+ Why: This would allow NAS filer backups in Bacula without incurring
+ the overhead of NFS or SBM/CIFS.
+
+ Notes: Further information is available:
+ http://www.ndmp.org
+ http://www.ndmp.org/wp/wp.shtml
+ http://www.traakan.com/ndmjob/index.html
+
+ There are currently no viable open-source NDMP
+ implementations. There is a reference SDK and example
+ app available from ndmp.org but it has problems
+ compiling on recent Linux and Solaris OS'. The ndmjob
+ reference implementation from Traakan is known to
+ compile on Solaris 10.
+
+ Notes: Kern: I am not at all in favor of this until NDMP becomes
+ an Open Standard or until there are Open Source libraries
+ that interface to it.
+
+Item h7: Commercial database support
+ Origin: Russell Howe <russell_howe dot wreckage dot org>
+ Date: 26 July 2006
+ Status:
+
+ What: It would be nice for the database backend to support more
+ databases. I'm thinking of SQL Server at the moment, but I guess Oracle,
+ DB2, MaxDB, etc are all candidates. SQL Server would presumably be
+ implemented using FreeTDS or maybe an ODBC library?
+
+ Why: We only really have one database server, which is MS SQL Server
+ 2000. Maintaining a second one for the backup software (we grew out of
+ SQLite, which I liked, but which didn't work so well with our database
+ size). We don't really have a machine with the resources to run
+ postgres, and would rather only maintain a single DBMS. We're stuck with
+ SQL Server because pretty much all the company's custom applications
+ (written by consultants) are locked into SQL Server 2000. I can imagine
+ this scenario is fairly common, and it would be nice to use the existing
+ properly specced database server for storing Bacula's catalog, rather
+ than having to run a second DBMS.
+
+ Notes: This might be nice, but someone other than me will probably need to
+ implement it, and at the moment, proprietary code cannot legally be
+ mixed with Bacula GPLed code. This would be possible only providing
+ the vendors provide GPLed (or OpenSource) interface code.
+
+Item h8: Incorporation of XACML2/SAML2 parsing
+ Date: 19 January 2006
+ Origin: Adam Thornton <athornton@sinenomine.net>
+ Status: Blue sky
+
+ What: XACML is "eXtensible Access Control Markup Language" and
+ "SAML is the "Security Assertion Markup Language"--an XML standard
+ for making statements about identity and authorization. Having these
+ would give us a framework to approach ACLs in a generic manner, and
+ in a way flexible enough to support the four major sorts of ACLs I
+ see as a concern to Bacula at this point, as well as (probably) to
+ deal with new sorts of ACLs that may appear in the future.
+
+ Why: Bacula is beginning to need to back up systems with ACLs
+ that do not map cleanly onto traditional Unix permissions. I see
+ four sets of ACLs--in general, mutually incompatible with one
+ another--that we're going to need to deal with. These are: NTFS
+ ACLs, POSIX ACLs, NFSv4 ACLS, and AFS ACLS. (Some may question the
+ relevance of AFS; AFS is one of Sine Nomine's core consulting
+ businesses, and having a reputable file-level backup and restore
+ technology for it (as Tivoli is probably going to drop AFS support
+ soon since IBM no longer supports AFS) would be of huge benefit to
+ our customers; we'd most likely create the AFS support at Sine Nomine
+ for inclusion into the Bacula (and perhaps some changes to the
+ OpenAFS volserver) core code.)
+
+ Now, obviously, Bacula already handles NTFS just fine. However, I
+ think there's a lot of value in implementing a generic ACL model, so
+ that it's easy to support whatever particular instances of ACLs come
+ down the pike: POSIX ACLS (think SELinux) and NFSv4 are the obvious
+ things arriving in the Linux world in a big way in the near future.
+ XACML, although overcomplicated for our needs, provides this
+ framework, and we should be able to leverage other people's
+ implementations to minimize the amount of work *we* have to do to get
+ a generic ACL framework. Basically, the costs of implementation are
+ high, but they're largely both external to Bacula and already sunk.
+
+ Notes: As you indicate this is a bit of "blue sky" or in other words,
+ at the moment, it is a bit esoteric to consider for Bacula.
+
+Item h9: Archive data
+ Date: 15/5/2006
+ Origin: calvin streeting calvin at absentdream dot com
+ Status:
+
+ What: The abilty to archive to media (dvd/cd) in a uncompressed format
+ for dead filing (archiving not backing up)
+
+ Why: At work when jobs are finished and moved off of the main file
+ servers (raid based systems) onto a simple Linux file server (ide based
+ system) so users can find old information without contacting the IT
+ dept.
+
+ So this data dosn't realy change it only gets added to,
+ But it also needs backing up. At the moment it takes
+ about 8 hours to back up our servers (working data) so
+ rather than add more time to existing backups i am trying
+ to implement a system where we backup the acrhive data to
+ cd/dvd these disks would only need to be appended to
+ (burn only new/changed files to new disks for off site
+ storage). basialy understand the differnce between
+ achive data and live data.
+
+ Notes: Scan the data and email me when it needs burning divide
+ into predefined chunks keep a recored of what is on what
+ disk make me a label (simple php->mysql=>pdf stuff) i
+ could do this bit ability to save data uncompresed so
+ it can be read in any other system (future proof data)
+ save the catalog with the disk as some kind of menu
+ system
+
+ Notes: Kern: I don't understand this item, and in any case, if it
+ is specific to DVD/CDs, which we do not recommend using,
+ it is unlikely to be implemented except as a user
+ submitted patch.
+
+
+Item h10: Clustered file-daemons
+ Origin: Alan Brown ajb2 at mssl dot ucl dot ac dot uk
+ Date: 24 July 2006
+ Status:
+ What: A "virtual" filedaemon, which is actually a cluster of real ones.
+
+ Why: In the case of clustered filesystems (SAN setups, GFS, or OCFS2, etc)
+ multiple machines may have access to the same set of filesystems
+
+ For performance reasons, one may wish to initate backups from
+ several of these machines simultaneously, instead of just using
+ one backup source for the common clustered filesystem.
+
+ For obvious reasons, normally backups of $A-FD/$PATH and
+ B-FD/$PATH are treated as different backup sets. In this case
+ they are the same communal set.
+
+ Likewise when restoring, it would be easier to just specify
+ one of the cluster machines and let bacula decide which to use.
+
+ This can be faked to some extent using DNS round robin entries
+ and a virtual IP address, however it means "status client" will
+ always give bogus answers. Additionally there is no way of
+ spreading the load evenly among the servers.
+
+ What is required is something similar to the storage daemon
+ autochanger directives, so that Bacula can keep track of
+ operating backups/restores and direct new jobs to a "free"
+ client.
+
+ Notes: Kern: I don't understand the request enough to be able to
+ implement it. A lot more design detail should be presented
+ before voting on this project.
+
+
+========== Already implemented ================================
+
+Item n: make changing "spooldata=yes|no" possible for
+ manual/interactive jobs
+ Origin: Marc Schiffbauer <marc@schiffbauer.net>
+ Date: 12 April 2007)
+ Status:
+
+ What: Make it possible to modify the spooldata option
+ for a job when being run from within the console.
+ Currently it is possible to modify the backup level
+ and the spooldata setting in a Schedule resource.
+ It is also possible to modify the backup level when using
+ the "run" command in the console.
+ But it is currently not possible to to the same
+ with "spooldata=yes|no" like:
+
+ run job=MyJob level=incremental spooldata=yes
+
+ Why: In some situations it would be handy to be able to switch
+ spooldata on or off for interactive/manual jobs based on
+ which data the admin expects or how fast the LAN/WAN
+ connection currently is.
+
+ Notes: ./.
+
+============= Empty Feature Request form ===========
+Item n: One line summary ...
+ Date: Date submitted
+ Origin: Name and email of originator.
+ Status:
+
+ What: More detailed explanation ...
+
+ Why: Why it is important ...
+
+ Notes: Additional notes or features (omit if not used)
+============== End Feature Request form ==============