9 - Document running multiple Jobs
10 - Document that two Verifys at same time on same client do not work.
11 - Document how to recycle a tape in 7 days even if the backup takes a long time.
12 - Document default config file locations.
13 - Document better includes (does it cross file systems ?).
14 - Document specifically how to add new File daemon to config files.
15 - Document forcing a new tape to be used.
16 - Document "Error in message.c:500 Mail program terminated in error.
19 --bindir is wrong and does not reflect prefix= in the *_sqlite_* scripts
21 --top level configure options are not passed to the depkgs, particularly
23 --Also, it might be better to split the depkgs location from the --with-sqlite
25 --should be able to specify e.g. --with-sqlite=/opt/local and have it find
26 lib, bin, sbin for itself
27 I tried this and it didn't find sqlite.h
28 --sd.conf password does not match dir.conf storage password
31 - Make Restore report an error if FD or SD term codes are not OK.
32 - Convert all %x substitution variables, which are hard to remember
33 and read to %(variable-name)s. Idea from TMDA.
34 - Report volume write rate.
35 - Fix db_get_job_volume_names() to return array of strings.
36 - Report compression % and other compression statistics if turned on.
37 - Add JobLevel in FD status (but make sure it is defined).
38 - Eliminate MySQL shared libraries from smtp and daemons not
40 - Pass "Catalog Files = no" to storage daemon to eliminate
42 - Make Pool resource handle Counter resources.
43 - Remove NextId for SQLite. Optimize.
44 - Fix gethostbyname() to use gethostbyname_r()
45 - Cleanup path/filename separation in sql_get.c and sql_create.c
46 - Implement ./configure --with-client-only
47 - Strip trailing / from Include
48 - Move all SQL statements into a single location.
49 - Cleanup db_update_media and db_update_pool
50 - Add UA rc and history files.
51 - put termcap (used by console) in ./configure and
52 allow -with-termcap-dir.
53 - Remove JobMediaId it is not used.
54 - Enhance time and size scanning routines.
55 - Fix Autoprune for Volumes to respect need for full save.
56 - DateWritten may be wrong.
57 - Fix Win32 config file definition name on /install
58 - When we are at EOM, we must ask each job to write JobMedia
59 record (update_volume_info).
60 - No READLINE_SRC if found in alternate directory.
61 - Add Client FS/OS id (Linux, Win95/98, ...).
62 - Put Windows files in Windows stream?
64 ====== 31 May 2002 ========
65 Now that Bacula 1.20 is released, virtually all the
66 basic features are implemented (some are still quite
67 primitive though). Over the next month or two, I'm
68 planning to focus on the following items:
71 - Fix any bugs I find or you report.
72 - Finish the implementation of automatic pruning
73 (add pruning of Restore and Verify jobs).
74 - Make sure pruning of Volumes won't prune the
75 only backup of a FileSet
78 - Improve the Restore capabilities of Bacula
79 - Restore to most recent system state (i.e.
80 figure out what tapes need to be mounted and
82 - Restore to a particular time (perhaps several
83 variations -- e.g. before date, after date).
84 - Interactive Restore where you get to select
85 what files are to be restored (much like the Unix
86 "restore" program permits). Now that we have a
87 catalog of all files saved, it would be nice to
89 - Restore options (overwrite, overwrite if older,
90 overwrite if newer, never overwrite, ...)
91 - Improve the standalone programs (bls and bextract)
92 to have pattern matching capabilities (e.g. restore
93 by FileSet, Job, JobType, JobLevel, ...).
94 - Ideally after each Job, Bacula could write out a
95 set of commands to a file that if later feed to
96 bextract would restore your system to the current
97 state (at least for the saved FileSet). This would
98 provide a simple disaster recovery that could be
99 initiated from a "floppy" and one simple ASCII control
100 file. I'm not exactly sure how to do this, but it
101 shouldn't be too hard and I'll
102 be trying to go in this direction.
105 - Implement tape verification to ensure that the data
106 written for a particular Job can really be read.
107 - Compare tape File attributes to Catalog.
108 (File attributes are size, dates, MD5, but not
110 - Compare tape to Client files (attributes, or
114 - With the current Bacula 1.21 (not yet in the CVS) I
115 expect there is about 95% chance that running multiple
116 simultaneous Jobs will actually work without stepping
117 on each other. I'm planning to try this sometime soon.
122 - Rework Storage daemon with new rwl_lock routines.
123 - Implement Label templates
124 - Pass JCR to database routines permitting better error printing.
127 - Verify against Full.
130 mysqldump -f --opt bacula >bacula
134 - Probably add End of Data tape records (this would make
135 the tape format incompatible with the previous version).
136 - I'll most likely enhance the current tape format
137 in the way that I previously described, which will make
138 some of the labels incompatible, but the change will
139 not affect the current restore code since it does not
140 look at the details of the labels.
141 - I may add a few more waiting conditions in the Storage
142 daemon where it will current immediately aborts a
143 Job if the necessary resources are not available (e.g.
144 tape is being written and a read request arrives).
145 - Write an applet for Linux.
147 - Remove PoolId from Job table, it exists in Media.
148 - Allow commands to detach or run in background.
149 - Write better dump of Messages resource.
150 - Fix status delay on storage daemon during rewind.
151 - Add VerNo to each Session label record.
152 - Add Job to Session records.
153 - Add VOLUME_CAT_INFO to the EOS tape record (as
154 well as to the EOD record).
155 - Add SD message variables to control operator wait time
156 - Maximum Operator Wait
157 - Minimum Message Interval
158 - Maximum Message Interval
159 - Add EOM handling variables
161 - Require EOD records
162 - Send Operator message when cannot read tape label.
163 - Think about how to handle I/O error on MTEOM.
164 - If Storage daemon aborts a job, ensure that this
165 is printed in the error message.
166 - Verify level=Volume (scan only), level=Data (compare of data to file).
167 Verify level=Catalog, level=InitCatalog
168 - Scan tape contents into database.
170 - Cold start full restore (restore catalog then
171 user selects what to restore). Write summary file containing only
172 Job, Media, and Catalog information. Store on another machine.
173 - Dump/Restore database
176 - Implement first cut of Catalog Retention period (remove old
177 entries from database).
178 - Add SessionTime/Id filters to bextract.
180 - Ensure that Start/End File/Block are correct.
181 - Add keyword search to show command in Console.
182 - If MySQL database is not running, job terminates with
183 wierd type and wierd error code.
184 - Write a regression script
185 - Report bad status from smtp or mail program.
186 - Fix Win2000 error with no messages during startup.
187 - Add estimate to Console
188 - Events : tape has more than xxx bytes.
189 - In Storage daemon, status should include job cancelled.
190 - Write general list maintenance subroutines.
191 - Implement immortal format with EDOs.
192 - Restrict characters permitted in a Resource name.
193 - Restore file xx or files xx, yy to their most recent values.
194 - Provide definitive identification of type in backup.
195 - Complete code in Bacula Resources -- this will permit
196 reading a new config file at any time.
197 - Document new Console
198 - Handle ctl-c in Console
199 - Test restore of Windows backup
200 - Implement LabelTemplate (at least first cut).
201 - Implement script driven addition of File daemon to
204 - Bug: anonymous Volumes requires mount in some cases.
205 - see setgroup and user for Bacula p4-5 of stunnel.c
206 - Implement new serialize subroutines
207 send(socket, "string", &Vol, "uint32", &i, NULL)
208 - Add save type to Session label.
209 - Correct date on Session label.
210 - On I/O error, write EOF, then try to write again.
211 - Audit all UA commands to ensure that we always prompt where
213 - If ./btape is called without /dev, assume argument is
214 a Storage resource name.
215 - Put memory utilization in Status output of each daemon
216 if full status requested or if some level of debug on.
217 - Make database type selectable by .conf files i.e. at runtime
218 - gethostbyname failure in bnet_connect() continues
219 generating errors -- should stop.
220 - Don't create a volume that is already written. I.e. create only once.
221 - If error at end of tape, implement some way to kill waiting processes.
222 - Get correct block/file information in Catalog, pay attention
224 - Add HOST to Volume label.
225 - Set flag for uname -a. Add to Volume label.
226 - Implement throttled work queue.
227 - Write bscan program that will syncronize the DB Media record with
228 the contents of the Volume -- for use after a crash.
229 - Check for EOT at ENOSPC or EIO or ENXIO (unix Pc)
230 - Allow multiple Storage specifications (or multiple names on
231 a single Storage specification) in the Job record. Thus a job
232 can be backed up to a number of storage devices.
233 - Implement full MediaLabel code.
234 - Implement dump label to UA
235 - Copy volume using single drive.
236 - Copy volume with multiple driven (same or different block size).
237 - Add block size (min, max) to Vol label.
238 - Concept of VolumeSet during restore which is a list
239 of Volume names needed.
240 - Restore files modified after date
241 - Restore file modified before date
242 - Emergency restore info:
244 - Backup working directory
246 - Restore options (do not overwrite)
247 - Restore -- do nothing but show what would happend
248 - Authentication between SD and FD
249 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
250 - Send Volumes needed during restore to Console
251 - Put Job statistics in End Session Label (files saved,
252 total bytes, start time, ...).
253 - Put FileSet name in the SOS label.
254 - Implement Restore FileSet=
255 - Write a scanner for the UA (keyword, scan-routine, result, prompt).
256 - Create a protocol.h and protocol.c where all protocol messages
258 - If SD cannot open a drive, make it periodically retry.
259 - Put Bacula version somewhere in Job stream, probably Start Session
261 - Remove duplicate fields from jcr (e.g. jcr.level and
263 - Timout a job or terminate if link goes down, or reopen link and query.
264 - Define how we handle times to avoid problem with Unix dates (2049 ?).
265 - The daemons should know when one is already
266 running and refuse to run a second copy.
267 - Fill all fields in Vol/Job Header -- ensure that everything
268 needed is written to tape. Think about restore to Catalog
269 from tape. Client record needs improving.
270 - Find general solution for sscanf size problems (as well
271 as sprintf. Do at run time?
273 - Concept of precious tapes (cannot be reused).
274 - Allow FD to run from inetd ???
275 - Preprocessing command per file.
276 - Postprocessing command per file (when restoring).
278 - Restore should get Device and Pool information from
279 job record rather than from config.
280 - Make SD send attribute stream to DR but first
281 buffering to file, then sending only when the
282 files are written to tape.
283 - Autolabel should be specified by DR instead of SD.
284 - Ability to recreate the catalog from a tape.
285 - Find out how to get the system tape block limits, e.g.:
286 Apr 22 21:22:10 polymatou kernel: st1: Block limits 1 - 245760 bytes.
287 Apr 22 21:22:10 polymatou kernel: st0: Block limits 2 - 16777214 bytes.
290 - AutoScan (check checksum of tape)
291 - Format command = "format /dev/nst0"
295 - Seek resolution (usually corresponds to buffer size)
296 - EODErrorCode=ENOSPC or code
297 - Partial Read error code
298 - Partial write error code
299 - Nonformatted read error
300 - Nonformatted write error
301 - WriteProtected error
305 - IgnoreCloseErrors=yes
315 - Eliminate duplicate File records to shrink database.
316 - FD sends unsaved file list to Director at end of job.
317 - Implement InsertUniqueDB.
318 - Write a Storage daemon that uses pipes and
319 standard Unix programs to write to the tape.
321 - Need something that monitors the JCR queue and
322 times out jobs by asking the deamons where they are.
323 - Add daemon JCR JobId=0 to have a daemon context
327 - Client (list of clients to force client)
328 - Devices (list of devices to force device)
333 - Recycle from other pools: Yes, No
334 - Recycle to other pools: Yes, no
337 - Optional MediaType to force media?
348 if(connect(sockfd, (struct sockaddr * ) (& addr), sizeof(addr)) .lt. 0){
354 linger.l_linger = 60;
355 i = setsockopt(sockfd, SOL_SOCKET, SO_LINGER, (char *) &linger,
358 fl = fcntl(sockfd, F_GETFL);
359 fcntl(sockfd, F_SETFL, fl & (~ O_NONBLOCK) & (~ O_NDELAY));
361 - Add "0nnn" in front of all sscanf %s fields
362 to prevent field overflow.
364 What: jobid or file list
365 From: tape, file, ...
366 Where: original location, another path
367 How: Always replace, Replace if newer, Never replace
368 Report: files restored; files not restored; errors; warnings
370 - Enhance Jmsg code to permit buffering and saving to disk.
371 - Probably create a jcr with JobId=0 as a master
372 catchall if jcr not found or if operation involves
374 - device driver = "xxxx" for drives.
375 - restart: paranoid: read label fsf to
376 eom read append block, and go
377 super-paranoid: read label, read all files
378 in between, read append block, and go
379 verify: backspace, read append block, and go
380 permissive: same as above but frees drive
381 if tape is not valid.
383 - Ensure that /dev/null works
384 - File daemon should build list of files skipped, and then
385 at end of save retry and report any errors.
386 - Need report class for messages. Perhaps
387 report resource where report=group of messages
388 - Extract what=(session_id|file_list); where
390 - enhance scan_attrib and rename scan_jobtype, and
391 fill in code for "since" option
392 - dir_config: get rid of all printfs
393 - To buffer messages, we need associated jobid and Director name.
394 - Need to save contents of FileSet to tape?
395 - Director needs a time after which the report status is sent
396 anyway -- or better yet, a retry time for the job.
397 Don't reschedule a job if previous incarnation is still running.
398 - Figure out how to do a "full" restore from catalog
399 - Figure out how to save the catalog (possibly a special FileSet).
400 - Figure out how to restore the catalog.
401 - Figure out how to put a Volume into the catalog (from the tape)
402 - Figure out how to do a restore from a Volume
403 - Some way to automatically backup everything is needed????
404 - Need a structure for pending actions:
406 - termination status (part of buffered msgs?)
407 - Concept of grouping Storage devices and job can use
408 any of a number of devices
410 Read, Write, Clean, Delete
411 - Login to Bacula; Bacula users with different permissions:
413 - Tape recycle destination
414 - Job Schedule Status
418 - File daemon should pass Director the operating system info
419 to be stored in the Client Record (or verified that it has
421 - Store info on each file system type (probably in the job header on tape.
422 This could be the output of df; or perhaps some sort of /etc/mtab record.
425 - Use media 1 time (so that we can do 6 days of incremental
426 backups before switching to another tape) (already)
427 specify # times (jobs)
428 specify bytes (already)
429 specify time (seconds, hours, days)
430 - Implement FSM (File System Modules).
431 - Identify unchanged or "system" files and save them to a
432 special tape thus removing them from the standard
433 backup FileSet -- BASE backup.
434 - Turn virutally all sprintfs into snprintfs.
435 - Heartbeat between daemons.
436 - Audit M_ error codes to ensure they are correct and
438 - Add variable break characters to lex analyzer.
439 Either a bit mask or a string of chars so that
440 the caller can change the break characters.
441 - Make a single T_BREAK to replace T_COMMA, etc.
442 - Ensure that File daemon and Storage daemon can
443 continue a save if the Director goes down (this
444 is NOT currently the case). Must detect socket error,
445 buffer messages for later.
448 Done: (see kernsdone for more)
449 --the console script is broken as installed and has to be hand-massaged with
450 paths, config files etc.
451 - Termination status in FD for Verify = C -- incorrect.
452 - Implement alter_sqlite_tables
453 - Fix scheduler -- see "Hourly cycle". It doesn't do both each
454 hour, rather it alternates between 0:05 and 0:35.
455 - Create Counter DB records.