1 Technical notes on version 1.39
7 - Fix the block listing bug pointed out by Arno.
8 - Update the po files at Eric's request.
11 - Fix Maximum Changer Wait, Maximum Open Wait, Maximum Rewind Wait to
12 accept time qualifiers.
13 - Make catalog respect ACL.
14 - Add recycle count to Media record.
16 Changes to 1.39.6-beta1 24Feb04
18 - Add Rudolf Cejka's new rc-chio-changer.
19 - Implement pulling Volume from Scratch Pool if the
20 Volume is in the autochanger.
21 - Implement additional command arguments for update Volume.
25 - Back port changes to 1.38.5
26 - Fix recycle SQL for StorageId.
27 - Fix misplaced quote in mtx-changer script.
29 - Note. Your database must be updated, or you must create
30 a new database. I have not yet incremented the database level.
32 - Add LocationId, DeviceId, and MediaTypeId to Media record.
34 - Implement create/get mac record in database for adding extended
35 Migration data to the job record.
36 - Add new MAC table to update/make database scripts.
37 - Return Storage name used when getting VolumeNames for a job.
38 - Change bsr file keyword Storage to Device, which is more accurate.
39 - Ensure that Mac records are pruned/purged.
40 - Tweak SD tools to deal with changing media type.
41 - Integrate more dev.c subroutines as methods (e.g. strerror, bsr, ...)
42 - Pass pointer to dcr pointer to acquire_device_for_read() so
43 that the subroutine can switch devices, and hence dcrs.
44 - Modify the multiple MediaType read code to re-use the same
45 dcr when switching devices. This makes the code much more
47 - Integrate patch from Karl Hakimian that reads JobIds, FileIndexes
48 from a table for restore.
49 - Add Storage name to VolParams, but it really should be Device.
51 - Add disk-changer to scripts directory + configure/Makefile
52 - Eliminate PoolId from jcr -- it is in jcr->jr.PoolId
53 - Implement store_bit scanner to replace store_yesno. Mostly done.
54 - Implement new store_bool that stores in a bool.
55 - Add true/false to yes/no conf directives.
56 - Make first cut at changing appropriate store_yesno to store_bool.
57 - Complete implementation of Pool storage devices.
58 - Move starting clones to job.c
59 - Move create_restore_bootstrap_file() to job.c
60 - Make copy_storage() more general to be able to handle
62 - Cleanup a lot of migration code for manual running, including
64 - Move getting a scratch Volume into a subroutine.
65 - Make all places a Volume that is added to a pool to
67 - Fix bug in autochanger recycle code (improper edit 64 bit).
68 - Fix segfault in restore command when no value specified.
69 - Start adding code to handle multiple MediaTypes in restore.
70 - Eliminate race condition in getting Volume name for
71 two drive autochanger.
72 - More debug code in autochanger.
73 - Add storage keyword to bootstrap file, add parsing.
74 - Move slot in bsr file into Volume record as there will be
75 a different slot for each Volume.
78 - Implement Pool storage overrides.
80 - Implement first cut of Migration.
81 - Implement mysql_use_result() from patch by Karl Hakimian.
82 This reduces significantly the memory consumption during
83 the restore tree building, and hence runs faster too.
84 - Implement StorageId patch supplied by user (reggie) in bug
85 #536. This should permit Bacula to work correctly with two
87 - Implement Job listing variations suggested by a user.
88 - Move updating bootstrap code in backup.c to subroutine
89 update_bootstrap_file().
90 - Add new job status elapsed time and bytes written user
91 friendly job report output patch sent by a user.
92 - Implement a storage list in Pools.
93 - Separate out setup_job() code from run_job().
94 - Get migration working -- lots of changes in mac.c in both
96 - Apply patch from user (Eric Bollinger I think) that fixes a
97 DIR crash when no arguments are supplied to a dot command.
98 - Fix typo (strcpy->strcmp) in ua_prune.c as reported by Martin.
99 - Fix command arg name->volume in label command of gnome-console.
100 - Fix SD acquire.c to release correct DCR by explicitly testing
101 on the dcr address rather than trying to devine if it is a read
102 or write dcr. This failed in error conditions when the device was
106 - Apply user supplied patch for more readable rate output
108 - Continue implementing migration.
109 - Implement support for removable filesystems in SD.
110 - Ensure that btraceback scripts can be read by anyone.
111 - Replace dvd-freespace and dvd-writepart by dvd-handler.
112 - Correct bug where canceling restore before the FD contacts
113 the SD causes the drive to be left in read mode.
114 - Move ofline_or_rewind into DEVICE::close().
115 - Eliminate close_device.
116 - Convert several dev subroutines to methods (e.g. bsf,
118 - Eliminate force_close_device().
119 - Implement Device Type directive in Device resource that
120 can have values File, Tape, Fifo, DVD, or Prog.
121 - Add has_cap() method to Device.
124 - Add patch from bug #527 to allow RedHat user to specify
125 options/user/group for starting each daemon in
128 - Add two new queries to query.sql provided by Arno. One
129 list volumes known to the Storage device, and the other
130 lists volumes possibly needing replacement (error, ...).
132 - Add periodic (every 24 hours) garbage collection of memory
133 pool by releasing free buffers.
135 - Correct bug counting sized (for display only) in smartall.c
136 - Print FD mempool stats if debug > 0 rather than 5.
138 - Make db_lock() mutex error fail the job rather than abort
139 Bacula. Canceling the job caused the mutex to fail.
140 - Correct bug in alist.c that re-allocated the list if the
141 number of items goes to zero.
142 - Move the reservation system thread locking to the top level
143 so that one job at a time tries all possible drives before
145 - Implement a reservation 'fail' message queue that is built
146 and destroyed on each pass through the reservation system.
147 These messages are displayed in a 'Jobs waiting to reserve
148 a drive' list during a 'status storage='. Note, multiple
149 messages will generally print for each JobId because they
150 represent the different problems with either the same drive
151 or different drives. If this output proves too confusing
152 of voluminous, I will display it only when debug level 1
153 or greater is enabled in the SD.
155 - Add enable/disable job=<job-name>. This command prevents
156 the specified job from being scheduled. Even when disabled,
157 the job can be manually started from the console.
158 - During 'update slots' clear all InChanger flags where the
159 StorageId is zero (old Media records).
163 - Fix autochanger code to strip leading spaces from returned
164 slots number. Remove bc from chio-changer.
165 - Back port a bit of 1.39 crypto code to reduce diffs.
166 - Fix first call to autochanger that missed close()ing the
167 drive. Put close() just before each run_program(). Fixes
170 - Add PoolId to Job record when updating it at job start time.
172 - Pull in more code from 1.39 so that there are fewer file
173 differences (the new ua_dotcmds.c, base64.h, crypto.h
174 hmac.c jcr.c (dird and lib) lib.h md5.h parse_conf.c
175 util.c. Aside from ua_dotcmds.c these are mostly crypto
177 - Implement new method of walking the jcr chain. The
178 incr/dec of the use_count is done within the walking
179 routines. This should prevent a jcr from being freed
180 from under the walk routines.
185 - Start implementing Verify list output.
186 - Move the suitable_drive flag to a better place to prevent
187 premature termination of the reservation if all drives
188 are busy -- should fix Arno's diff/inc pool failures.
190 - Add mutex to single thread VSS code in Win32.
192 - Simplify code in askdir.c that waits for creating an appendable
193 volume so that it can handle multiple returns from the wait
195 - Modify the wait code to permit multiple returns.
196 - Return a zero when "autochanger drives" is called and
197 it is not an autochanger.
198 - Make rewind_dev() a method taking a DCR as an argument.
199 This permits closing and reopening the drive if the
200 rewind fails as happens if the drive was loaded while the
201 file descriptor was open. This refreshes the file descriptor.
202 - Remove the ST_OPENED flag and always rely on fd < 0 for knowing
203 if the device is open or not. This should eliminate
205 - Return error if reserve cannot find at least one suitable device.
206 - Make wait_for_sysop() return correct state information.
207 - Fix Win32 state file problem. write was not using compat
208 code. This should fix bug #500.
210 - Modify gui on command to set only GUI mode and not batch.
211 - Modify .messages command to always print messages regardless
213 - If GUI mode is on, suppress automatic printing of
215 - Delete old bnet packet code.
216 - Ignore new BNET_START_SELECT and BNET_END_SELECT signals in
218 - Modify restore command in wx-console to set gui on and to use
219 only .messages instead of messages. Hopefully this fixes bug
221 - Fix seg fault in exit of acquire when canceling a job --
222 reported by Wolfgang Denk
223 - Pull in latest reservation system changes from 1.38
224 - Make .messages command always print messages regardless
225 of the automessages flag.
227 - Fix seg fault if user labels a drive directory bug #513
228 - Remove quotes around Version as it breaks things.
230 - Merge in Aleksandar Milivojevic's mods to the spec file.
231 - Apply sparse code fix for raw drives and fifos. Bug 506
232 - Thorsten fixed Unicode cd problem with wx-console bug 505.
234 - Correct reservation system to do a last ditch try
235 for any mounted volume, then anyone anywhere.
236 - Add quotes around table Version because of
237 error in MySQL 4.1.15 -- bug report submitted.
238 - Correct some minor problems with btape in the fill
240 - Updates to ssh-tunnel from Joshua Kugler.
241 - Added a report.pl program from Jonas Bjorklund.
242 - Simplify the O_NONBLOCK open() code for tape drives,
243 and always open nonblocking.
244 - Do not wait for open() if EIO returned (shouldn't happen).
245 - Eliminate 3 argument to tape open().
246 - Correct the slot # edited in the 3995 Bad autochanger unload
248 - With -S on bscan (show progress) do not divide by zero.
250 - Make cancel pthread_cond_signal() pthread_cond_broadcast().
251 - When dcr is freed, also broadcast dev->wait_next_vol signal.
252 - Remove unused code in wait_for_device.
253 - Make wait_for_device() always return after 60 seconds of wait.
257 - Add stubs for non-crypto build.
259 - Use localhost if no network configured
261 - Eliminated duplicate MaxVolBytes in cat update -- bug 509.
262 - Remove debug print.
263 - Add bail_out in error during state file reading.
265 - Merge changes made to 1.38.3 into HEAD
266 - Add stubs for pygtk-console code
267 - Create Makefile.in for pygtk-console code
269 - Merge updates into 1.38 branch
270 - Update specs to include mysql4 define.
271 - Fix when attributes are sent, must be after binit().
272 - Stop read_record() if status not ok in second loop.
273 - Return rec->FileIndex in dcr->VolLastIndex for normal
274 and partial records in read_record(). This allows bscan
275 to get FileIndex at EOT correct.
276 - Fix butil.c to correctly set dcr -- fixes seg fault in bls.
278 - Fix Win32 built to work with new crypto code.
279 - Apply patch supplied by user (slightly modified) to fix
280 correct detection of holes in block devices and FIFOs.
282 - Apply patch supplied by user (slightly modified)
283 to fix SD hang with multiple pools and bad client
286 - Add nagios plugin to the examples directory. Submitted by
288 - Remove warning message about multiple saves of hardlinked files
289 from find_one.c as it can generate too many warning messages.
290 - Modify most restore error messages to be queued so that they
291 appear at the end of the job rather than mixted with the restore
292 listing where they could be "lost".
294 - Reset timeout values before select() per patch from
295 Frank Sweetser for problems with non-blocking sockets.
296 - Unlink the state file if either reading or writing it gets
297 errors. Hopefully this will fix Win32 exit problems.
298 - Add sanity check in append.c to ensure that dcr is not NULL.
299 This can happen if multiple drive autochanger SCSI control
300 channel and drive indicies do not correspond.
302 - Get next volume from Scratch pool before creating a volume.
303 - Set new Pool defaults in Vol when moved from Scratch Pool.
304 - Remove argument from create_bacula_database for SQLite as it
306 - Add back index code so that two drive autochangers can get
308 - Change a bunch of debug levels to aid debugging autochangers.
309 - Fix reservation so that mutexes are properly applied.
310 - Rework reservation algorithm so that two drives can be used
313 - Landon merged his data encription changes into the HEAD
314 - Apply days keyword patch from Alexander.Bergolth at wu-wien.ac.at
315 If this patch is applied, the number of days can be specified with
316 "list nextvol days=xx"
319 My use case is to be able to preview the next scheduled job (and the
320 next tape to be used) on fridays if there are no scheduled jobs during
324 - Fix font code in gnome2 console user patch. Fixes bug #501.
325 - Fix malformatted bnet error message that caused seg fault
327 - Applied user patch to improve README.vc8 in src/win32.
329 - Add Migrate, Copy, Archive Job types (some where there)
330 - Correct some more editing of JobId's (for 64 bit compatibility).
331 - Ensure that StorageId is stored in Media record when ever possible.
332 - Add Migration Job to Job.
333 - Add Migration Time, Migration High Bytes, Migration Low Bytes
334 Next Pool to Pool resource.
335 - Add more code to mac.c (migration archive copy).
336 - Change Start Storage daemon job to require read and write storage
338 - Pass read storage data to SD as well as write storage data.
339 - Remove old code from winservice.cpp
340 - Break on error in scan.
341 - Fix typo in signal.c
342 - Separate read/write DCR in SD. Add jcr->read_dcr.
343 - Cleanup how find_device() works.
344 - Add read output to Status in SD.
347 - Add red-black btree routines
349 - Remove abs() in bfile.c so that it compiles on Solaris.
352 - Fix crash in tray-monitor when daemon disconnects. Bug #479.
353 - Fix bnet-server bug found on OpenBSD. Bug #486
354 - Fix cancel failure bug. Bug #481
355 - Fix failure when Pool name has spaces. Bug #487
356 - Fix SD crash in autochanger code. Mutex failure. Bug #488
357 - Fix a couple of free()s in src/filed/acl.c
358 - Fix memory overrun in bfile.c in building OS X resource
359 fork filename. Bug #489
360 - Add Pool name to SD status output.
362 - Apply SunOS patch for ACLs submitted by David Duchscher.
363 - Make sure to set storage before trying to set drive.
364 - Add bacula_mail_summary.sh to examples directory. It makes
365 a single email summary of any number of jobs. Submitted
367 - Make sure when we do a mount to unblock the device even
368 if the drive could not be opened.
370 - Remove the USE_WIN32STREAMEXTRACTION #defines (always on)
371 and correct a few minor problems to make it build on Linux.
373 - Remove delete of CVS from all Makefiles
374 - Fix seg fault when clicking on Add button in wx-console
375 restore panel. Bug #470.
376 - Fix copyright date and URL typo -- bug #468.
377 - Change autostart install for FreeBSD to look for rc.conf
378 rather than rc.local as suggested fix for bug #466.
379 - Apply patch supplied by Eric Bollinger to fix PostgreSQL
380 grant on status. Bug #465
381 - Apply patch supplied by Eric Bollinger to fix PostgreSQL
382 update script. Bug #464
383 - Tweak #ifdefing a bit in new Win32 stream code.
384 - Fix #ifdeffing for FD_NO_SEND_TEST.
385 - Add documentation of performance #defines