rsync.git
17 years agoDocument the new --log-file option.
Wayne Davison [Tue, 9 May 2006 18:31:24 +0000 (18:31 +0000)]
Document the new --log-file option.

17 years agoOutput the progress message as FCLIENT instead of FINFO.
Wayne Davison [Tue, 9 May 2006 18:31:19 +0000 (18:31 +0000)]
Output the progress message as FCLIENT instead of FINFO.

17 years agoGot rid of the call to log_exit() now that the cleanup code handles this.
Wayne Davison [Tue, 9 May 2006 18:31:16 +0000 (18:31 +0000)]
Got rid of the call to log_exit() now that the cleanup code handles this.

17 years ago- Turned some FINFO rprintf() calls into FCLIENT calls that don't go
Wayne Davison [Tue, 9 May 2006 18:31:13 +0000 (18:31 +0000)]
- Turned some FINFO rprintf() calls into FCLIENT calls that don't go
  into a client-side log-file.
- Added a couple FLOG rprintf() calls.

17 years ago- Use the new logfile_* variables that are set by the --log-file
Wayne Davison [Tue, 9 May 2006 18:31:10 +0000 (18:31 +0000)]
- Use the new logfile_* variables that are set by the --log-file
  option instead of the old daemon_log_* variables.
- Renamed the "logfile" variable "logfile_fp".
- Enhanced the logic in rwrite() to handle client-side FLOG output.

17 years agoSome demon_log_* variables changed into logfile_* variables that are
Wayne Davison [Tue, 9 May 2006 18:31:06 +0000 (18:31 +0000)]
Some demon_log_* variables changed into logfile_* variables that are
also used by the new --log-file option.

17 years agoAdded --log-file and several logfile_* support variables.
Wayne Davison [Tue, 9 May 2006 18:31:03 +0000 (18:31 +0000)]
Added --log-file and several logfile_* support variables.

17 years agoRevised the logic for when to call log_exit() to handle the
Wayne Davison [Tue, 9 May 2006 18:31:00 +0000 (18:31 +0000)]
Revised the logic for when to call log_exit() to handle the
new --log-file option.

17 years agoTweaked the comment about log-message categories.
Wayne Davison [Tue, 9 May 2006 18:30:57 +0000 (18:30 +0000)]
Tweaked the comment about log-message categories.

17 years agoGot rid of an unused variable.
Wayne Davison [Tue, 9 May 2006 18:00:15 +0000 (18:00 +0000)]
Got rid of an unused variable.

17 years agoMention the latest bugfix and improved an old description.
Wayne Davison [Tue, 9 May 2006 17:46:04 +0000 (17:46 +0000)]
Mention the latest bugfix and improved an old description.

17 years agoIf the --partial-dir value is an absolute path and we use a file in
Wayne Davison [Tue, 9 May 2006 17:38:47 +0000 (17:38 +0000)]
If the --partial-dir value is an absolute path and we use a file in
it as an alternate basis file, make sure that we delete it when the
associated destination file is successfully updated.

17 years agoImproved the doc for --existing and --ignore-existing.
Wayne Davison [Fri, 5 May 2006 20:41:17 +0000 (20:41 +0000)]
Improved the doc for --existing and --ignore-existing.

17 years ago- Moved the sanitizing of the partial_dir value here from options.c.
Wayne Davison [Fri, 5 May 2006 16:00:33 +0000 (16:00 +0000)]
- Moved the sanitizing of the partial_dir value here from options.c.
- Call die_on_unsafe_path() in a few needed places.
- Handle the server-excluded checks for an absolute partial_dir and
  the --*-dest options.

17 years ago- Added some calls to die_on_unsafe_path() to the path-sanitizing code.
Wayne Davison [Fri, 5 May 2006 15:58:44 +0000 (15:58 +0000)]
- Added some calls to die_on_unsafe_path() to the path-sanitizing code.
- Moved the sanitizing of the partial_dir value into main.c.

17 years ago- Added a call to die_on_unsafe_path() to handle_partial_dir().
Wayne Davison [Fri, 5 May 2006 15:54:20 +0000 (15:54 +0000)]
- Added a call to die_on_unsafe_path() to handle_partial_dir().
- Fixed some strange slash handling in partial_dir_fname().

17 years agoA few minor changes to match the recent --help changes in
Wayne Davison [Fri, 5 May 2006 06:24:25 +0000 (06:24 +0000)]
A few minor changes to match the recent --help changes in
options.c.

17 years ago- Don't check HAVE_LINK directly, use SUPPORT_HARD_LINKS instead.
Wayne Davison [Fri, 5 May 2006 06:16:22 +0000 (06:16 +0000)]
- Don't check HAVE_LINK directly, use SUPPORT_HARD_LINKS instead.
- Updated a few items in the --help text.

17 years agoCall the new die_on_unsafe_path() function in a few places.
Wayne Davison [Fri, 5 May 2006 05:57:48 +0000 (05:57 +0000)]
Call the new die_on_unsafe_path() function in a few places.

17 years agoCall the new die_on_unsafe_path() function in a couple spots.
Wayne Davison [Fri, 5 May 2006 05:56:37 +0000 (05:56 +0000)]
Call the new die_on_unsafe_path() function in a couple spots.

17 years agoAdded a new function named die_on_unsafe_path(). This is used
Wayne Davison [Fri, 5 May 2006 05:55:40 +0000 (05:55 +0000)]
Added a new function named die_on_unsafe_path().  This is used
by a non-chroot daemon to ensure that a user-supplied path does
not contain an unsafe symlink element.

17 years ago- Don't check HAVE_LINK directly, use SUPPORT_HARD_LINKS instead.
Wayne Davison [Fri, 5 May 2006 05:53:36 +0000 (05:53 +0000)]
- Don't check HAVE_LINK directly, use SUPPORT_HARD_LINKS instead.
- Tweaked Paul's recent change just a tad.
- Fixed the output of --compare-dest when SUPPORT_HARD_LINKS is
  not defined.

17 years agoCast file->mode to an int when printing it via %o.
Wayne Davison [Fri, 5 May 2006 05:40:17 +0000 (05:40 +0000)]
Cast file->mode to an int when printing it via %o.

17 years agoFix generator.c to avoid calling do_link() on systems that do not support
Paul Green [Thu, 4 May 2006 19:00:37 +0000 (19:00 +0000)]
Fix generator.c to avoid calling do_link() on systems that do not support
the link() function.  The changes test OK on SuSE Linux and on my own
Stratus VOS system, but I would be grateful if other folks could
double-check this change.  Paul G.

17 years agoA few more improvements.
Wayne Davison [Thu, 4 May 2006 02:42:55 +0000 (02:42 +0000)]
A few more improvements.

17 years agoA few minor twiddles.
Wayne Davison [Thu, 4 May 2006 02:38:58 +0000 (02:38 +0000)]
A few minor twiddles.

17 years agoAdded/clarified the latest bug-fix info.
Wayne Davison [Thu, 4 May 2006 02:29:58 +0000 (02:29 +0000)]
Added/clarified the latest bug-fix info.

17 years agoCast the enum in rwrite() when calling send_msg().
Wayne Davison [Wed, 3 May 2006 16:58:36 +0000 (16:58 +0000)]
Cast the enum in rwrite() when calling send_msg().

17 years agoUse send_msg() when forwarding a log message, not io_multiplex_write().
Wayne Davison [Wed, 3 May 2006 16:17:49 +0000 (16:17 +0000)]
Use send_msg() when forwarding a log message, not io_multiplex_write().
This allows send_msg() to decide if the message should be sent right
away, or should be deferred.

17 years agoChanged read_msg_fd() to always use send_msg()/rwrite() instead of
Wayne Davison [Wed, 3 May 2006 16:17:46 +0000 (16:17 +0000)]
Changed read_msg_fd() to always use send_msg()/rwrite() instead of
sometimes calling io_multiplex_write()/rwrite() or msg_list_add().
This simplifies the code that does the defer_forwarding_messages
checking because it can now be done in just one place, in send_msg()
(because rwrite() also calls send_msg() now when forwarding a msg).

17 years agoA small optimization to the new code in msg2sndr_flush().
Wayne Davison [Wed, 3 May 2006 06:32:59 +0000 (06:32 +0000)]
A small optimization to the new code in msg2sndr_flush().

17 years agoMade log_init() call timestring() instead of just localtime() prior to a
Wayne Davison [Wed, 3 May 2006 06:11:11 +0000 (06:11 +0000)]
Made log_init() call timestring() instead of just localtime() prior to a
chroot (note that timestring() makes use of localtime() plus strftime()).

17 years agoImproved the last bugfix description.
Wayne Davison [Wed, 3 May 2006 05:37:00 +0000 (05:37 +0000)]
Improved the last bugfix description.

17 years agoMention the fix for the "unexpected tag 3" bug.
Wayne Davison [Wed, 3 May 2006 05:22:01 +0000 (05:22 +0000)]
Mention the fix for the "unexpected tag 3" bug.

17 years agoFixed a bug where deferred MSG_INFO/MSG_ERROR/MSG_LOG messages were
Wayne Davison [Wed, 3 May 2006 05:19:01 +0000 (05:19 +0000)]
Fixed a bug where deferred MSG_INFO/MSG_ERROR/MSG_LOG messages were
not being forwarded from the generator to the sender using rwrite()
(and MSG_LOG messages didn't need to be deferred at all).

17 years agoAdded module_dirlen now that utils.o needs it.
Wayne Davison [Wed, 3 May 2006 00:49:31 +0000 (00:49 +0000)]
Added module_dirlen now that utils.o needs it.

17 years ago- Call safe_stat() instead of do_stat() when a sanitizing daemon might
Wayne Davison [Wed, 3 May 2006 00:48:32 +0000 (00:48 +0000)]
- Call safe_stat() instead of do_stat() when a sanitizing daemon might
  need to ensure that it doesn't follow any unsafe symlinks.
- Got rid of startdir_depth code (in favor of util.c's curr_dir_depth).
- Use the new calling syntax for sanitize_path().

17 years ago- Call safe_stat() instead of do_stat() when a sanitizing daemon might
Wayne Davison [Wed, 3 May 2006 00:48:29 +0000 (00:48 +0000)]
- Call safe_stat() instead of do_stat() when a sanitizing daemon might
  need to ensure that it doesn't follow any unsafe symlinks.
- Don't sanitize the string of a symlink -- just use it verbatim.

17 years ago- Call safe_stat() instead of do_stat() when a sanitizing daemon might
Wayne Davison [Wed, 3 May 2006 00:48:26 +0000 (00:48 +0000)]
- Call safe_stat() instead of do_stat() when a sanitizing daemon might
  need to ensure that it doesn't follow any unsafe symlinks.
- Use the new calling syntax for sanitize_path().

17 years ago- Added curr_dir_depth, a variable that keeps track of the depth past
Wayne Davison [Wed, 3 May 2006 00:48:23 +0000 (00:48 +0000)]
- Added curr_dir_depth, a variable that keeps track of the depth past
  the module root when a sanitizing daemon is running.
- Use the new calling syntax for sanitize_path().
- Updated count_dir_elements() to not count a . component.
- Improved sanitize_path() so that it can take a symlink target string
  and combine it with the symlink's name when the target is relative.
  When working with a symlink, the routine also returns NULL if any ..
  elements try to escape past the module root.
- The new routine safe_stat(), is used when sanitizing paths (i.e. when
  the daemon does not have chroot enabled).  This manually follows a
  chain of symlinks, ensuring they don't try to escape the module.

17 years agoUse the new calling syntax for sanitize_path().
Wayne Davison [Wed, 3 May 2006 00:48:20 +0000 (00:48 +0000)]
Use the new calling syntax for sanitize_path().

17 years agoTweaked some variable names in readlink_stat() and link_stat().
Wayne Davison [Tue, 2 May 2006 19:52:28 +0000 (19:52 +0000)]
Tweaked some variable names in readlink_stat() and link_stat().

17 years agoMention the latest bug-fixes.
Wayne Davison [Tue, 2 May 2006 17:38:34 +0000 (17:38 +0000)]
Mention the latest bug-fixes.

17 years agoTest the new symlink sanitizing fix by using ::test-scratchdir/to
Wayne Davison [Tue, 2 May 2006 17:13:11 +0000 (17:13 +0000)]
Test the new symlink sanitizing fix by using ::test-scratchdir/to
as the destination and including a symlink with a leading "../" .

17 years agoTweaked our test config file to include transfer logging that
Wayne Davison [Tue, 2 May 2006 17:11:42 +0000 (17:11 +0000)]
Tweaked our test config file to include transfer logging that
includes %i, and to have a test-scratchdir module.

17 years agoMoved the sanitizing of a symlink's link string here from flist.c.
Wayne Davison [Tue, 2 May 2006 16:52:51 +0000 (16:52 +0000)]
Moved the sanitizing of a symlink's link string here from flist.c.

17 years agoFixed a problem in the --link-dest processing where -I (--ignore-times)
Wayne Davison [Tue, 2 May 2006 16:46:48 +0000 (16:46 +0000)]
Fixed a problem in the --link-dest processing where -I (--ignore-times)
or --size-only could adversely affect our ability to find identical
files to hard-link.

17 years agoSet a new variable, startdir_depth, to indicate how deep the
Wayne Davison [Tue, 2 May 2006 16:42:20 +0000 (16:42 +0000)]
Set a new variable, startdir_depth, to indicate how deep the
destination dir is in a daemon module's path.

17 years agoGot rid of the sanitizing of a symlink's link string because it
Wayne Davison [Tue, 2 May 2006 16:36:07 +0000 (16:36 +0000)]
Got rid of the sanitizing of a symlink's link string because it
is happening too early to work correctly (before we know what
our destination-directory's depth in the module will be).

17 years agoWe can't check the --*-dest args against the server filter rules
Wayne Davison [Fri, 28 Apr 2006 17:34:53 +0000 (17:34 +0000)]
We can't check the --*-dest args against the server filter rules
here because we don't know what the args are going to be relative
to yet.

17 years agoWe sanitize the --*-dest args here so they can be relative to the
Wayne Davison [Fri, 28 Apr 2006 16:58:20 +0000 (16:58 +0000)]
We sanitize the --*-dest args here so they can be relative to the
destination directory.

17 years agoMoved the sanitizing of the --*-dest args into main.c.
Wayne Davison [Fri, 28 Apr 2006 16:57:39 +0000 (16:57 +0000)]
Moved the sanitizing of the --*-dest args into main.c.

18 years agoUpdated the FSF's address to an even newer one.
Wayne Davison [Tue, 25 Apr 2006 23:51:12 +0000 (23:51 +0000)]
Updated the FSF's address to an even newer one.

18 years ago- Updated the address for the FSF in the opening comment.
Wayne Davison [Tue, 25 Apr 2006 20:23:34 +0000 (20:23 +0000)]
- Updated the address for the FSF in the opening comment.
- Standardized the format of the opening comment, including adding a
  brief description of what's in the file for those that lacked it.
- Added some missing copyright lines.
- Some minor whitespace tweaks (in a few of the files).

18 years agoTweaked some whitespace to match the latest version from autoconf.
Wayne Davison [Mon, 24 Apr 2006 23:43:45 +0000 (23:43 +0000)]
Tweaked some whitespace to match the latest version from autoconf.

18 years agoIncremented the version to 2.6.9cvs.
Wayne Davison [Mon, 24 Apr 2006 22:24:40 +0000 (22:24 +0000)]
Incremented the version to 2.6.9cvs.

18 years agoRolled over the news file for the next version.
Wayne Davison [Mon, 24 Apr 2006 22:23:56 +0000 (22:23 +0000)]
Rolled over the news file for the next version.

18 years agoUpdated to the version dated 2006-02-23.
Wayne Davison [Mon, 24 Apr 2006 22:20:10 +0000 (22:20 +0000)]
Updated to the version dated 2006-02-23.

18 years agoPreparing for release of 2.6.8 v2.6.8
Wayne Davison [Sat, 22 Apr 2006 15:39:57 +0000 (15:39 +0000)]
Preparing for release of 2.6.8

18 years agoMention the phase-unknown changes.
Wayne Davison [Fri, 21 Apr 2006 16:57:13 +0000 (16:57 +0000)]
Mention the phase-unknown changes.

18 years ago- Got rid of the (mostly useless) io_read_phase and io_write_phase
Wayne Davison [Fri, 21 Apr 2006 16:40:19 +0000 (16:40 +0000)]
- Got rid of the (mostly useless) io_read_phase and io_write_phase
  variables.  In the future it might be useful to get rsync to
  mention what kind of a message it was trying to write, but the
  'phase "unknown"' output was ubiquitous and confusing.
- Got rid of the completely unused write_int_named() function.

18 years agoGot rid of the setting of io_write_phase.
Wayne Davison [Fri, 21 Apr 2006 16:36:30 +0000 (16:36 +0000)]
Got rid of the setting of io_write_phase.

18 years agoMention the security fix for patches/xattrs.diff.
Wayne Davison [Fri, 21 Apr 2006 00:05:30 +0000 (00:05 +0000)]
Mention the security fix for patches/xattrs.diff.

18 years agoImproved the --owner and --group sections.
Wayne Davison [Tue, 18 Apr 2006 18:58:07 +0000 (18:58 +0000)]
Improved the --owner and --group sections.

18 years agoTwiddle the trailing output in non-live mode.
Wayne Davison [Tue, 18 Apr 2006 06:43:17 +0000 (06:43 +0000)]
Twiddle the trailing output in non-live mode.

18 years agoFixed a typo in the new code.
Wayne Davison [Tue, 18 Apr 2006 06:38:58 +0000 (06:38 +0000)]
Fixed a typo in the new code.

18 years ago- Fixed --copy-dest behavior on a system where CAN_HARDLINK_SYMLINK
Wayne Davison [Tue, 18 Apr 2006 05:48:46 +0000 (05:48 +0000)]
- Fixed --copy-dest behavior on a system where CAN_HARDLINK_SYMLINK
  and/or CAN_HARDLINK_SPECIAL is not defined.
- Changed --link-dest behavior on such systems to work just like
  --copy-dest for the non-hard-linkable files.

18 years agoA slight improvement to the logic in try_dests_non().
Wayne Davison [Tue, 18 Apr 2006 05:24:02 +0000 (05:24 +0000)]
A slight improvement to the logic in try_dests_non().

18 years agoMention the latest bugfix.
Wayne Davison [Mon, 17 Apr 2006 19:55:16 +0000 (19:55 +0000)]
Mention the latest bugfix.

18 years agoFixed a bug where special files were not getting hard-linked via
Wayne Davison [Mon, 17 Apr 2006 19:53:41 +0000 (19:53 +0000)]
Fixed a bug where special files were not getting hard-linked via
the --link-dest option.

18 years agoDon't use the -C option to make (since it's not portable).
Wayne Davison [Mon, 17 Apr 2006 18:15:02 +0000 (18:15 +0000)]
Don't use the -C option to make (since it's not portable).

18 years agoMention the latest fixes.
Wayne Davison [Mon, 17 Apr 2006 18:00:09 +0000 (18:00 +0000)]
Mention the latest fixes.

18 years agoUse the new CAN_HARDLINK_SYMLINK and CAN_HARDLINK_SPECIAL defines
Wayne Davison [Mon, 17 Apr 2006 17:56:10 +0000 (17:56 +0000)]
Use the new CAN_HARDLINK_SYMLINK and CAN_HARDLINK_SPECIAL defines
from configure instead of the old (and not quite correct) run-time
checking.

18 years agoAdded two new checks: 1. Does link() hard-link symlinks (not the
Wayne Davison [Mon, 17 Apr 2006 17:53:47 +0000 (17:53 +0000)]
Added two new checks:  1. Does link() hard-link symlinks (not the
referent)?  2. Does link() hard-link a fifo?

18 years agoDon't use -C with make because it's not portable.
Wayne Davison [Mon, 17 Apr 2006 17:37:13 +0000 (17:37 +0000)]
Don't use -C with make because it's not portable.

18 years agoGot rid of the unused REPLACE_INET_NTOA and REPLACE_INET_ATON checks.
Wayne Davison [Thu, 13 Apr 2006 16:54:24 +0000 (16:54 +0000)]
Got rid of the unused REPLACE_INET_NTOA and REPLACE_INET_ATON checks.

18 years agoGot rid of inet_ntoa -> rep_inet_ntoa tom-foolery because rsync
Wayne Davison [Thu, 13 Apr 2006 16:53:15 +0000 (16:53 +0000)]
Got rid of inet_ntoa -> rep_inet_ntoa tom-foolery because rsync
doesn't use inet_ntoa() anymore.

18 years agoGot rid of the replacement inet_aton() and inet_ntoa() functions
Wayne Davison [Thu, 13 Apr 2006 16:52:15 +0000 (16:52 +0000)]
Got rid of the replacement inet_aton() and inet_ntoa() functions
because no code in rsync has used these functions since 2.4.8.

18 years agoImproved a sentence.
Wayne Davison [Wed, 12 Apr 2006 16:59:32 +0000 (16:59 +0000)]
Improved a sentence.

18 years agoA few minor changes to some messages and a couple variable improvements.
Wayne Davison [Wed, 12 Apr 2006 16:35:53 +0000 (16:35 +0000)]
A few minor changes to some messages and a couple variable improvements.

18 years agoPreparing for release of 2.6.8pre1 v2.6.8pre1
Wayne Davison [Wed, 12 Apr 2006 16:23:46 +0000 (16:23 +0000)]
Preparing for release of 2.6.8pre1

18 years agoMoved the include of netinet/ip.h into socket.c.
Wayne Davison [Tue, 11 Apr 2006 00:48:55 +0000 (00:48 +0000)]
Moved the include of netinet/ip.h into socket.c.

18 years agoMoved the include of netinet/ip.h here along with its apparent
Wayne Davison [Tue, 11 Apr 2006 00:48:28 +0000 (00:48 +0000)]
Moved the include of netinet/ip.h here along with its apparent
prerequisite, netinet/in_systm.h .

18 years agoTweaked a comment.
Wayne Davison [Tue, 11 Apr 2006 00:28:02 +0000 (00:28 +0000)]
Tweaked a comment.

18 years agoInclude netinet/ip.h .
Wayne Davison [Sun, 9 Apr 2006 13:51:56 +0000 (13:51 +0000)]
Include netinet/ip.h .

18 years agoMention the read-position fix.
Wayne Davison [Sat, 8 Apr 2006 16:43:21 +0000 (16:43 +0000)]
Mention the read-position fix.

18 years agoFixed the value of map->p_fd_offset when a read() fails or is
Wayne Davison [Sat, 8 Apr 2006 16:37:50 +0000 (16:37 +0000)]
Fixed the value of map->p_fd_offset when a read() fails or is
abbreviated.

18 years agoUse --delete-after instead of --delete.
Wayne Davison [Sat, 8 Apr 2006 16:36:57 +0000 (16:36 +0000)]
Use --delete-after instead of --delete.

18 years agoWe don't need log_initialised exported after all.
Wayne Davison [Sat, 8 Apr 2006 16:04:34 +0000 (16:04 +0000)]
We don't need log_initialised exported after all.

18 years agoMoved the calls to client_addr() and client_name() below the lp_load()
Wayne Davison [Sat, 8 Apr 2006 16:03:16 +0000 (16:03 +0000)]
Moved the calls to client_addr() and client_name() below the lp_load()
call so that their potential to output log messages can't cause the
log-file to initialize before we know what "log name" to use.

18 years agoClarified a few items.
Wayne Davison [Thu, 6 Apr 2006 07:54:03 +0000 (07:54 +0000)]
Clarified a few items.

18 years agoA slightly better fix than the last version.
Wayne Davison [Thu, 6 Apr 2006 07:38:54 +0000 (07:38 +0000)]
A slightly better fix than the last version.

18 years agoMade log_initialised public.
Wayne Davison [Thu, 6 Apr 2006 07:33:12 +0000 (07:33 +0000)]
Made log_initialised public.

18 years agoThere's no need for lp_load() to check am_daemon.
Wayne Davison [Thu, 6 Apr 2006 07:28:22 +0000 (07:28 +0000)]
There's no need for lp_load() to check am_daemon.

18 years agoImproved the bind()-failure message to mention what family failed.
Wayne Davison [Sun, 2 Apr 2006 21:15:37 +0000 (21:15 +0000)]
Improved the bind()-failure message to mention what family failed.

18 years agoGot rid of duplicate rsync: prefix from an rsyserr() call.
Wayne Davison [Sun, 2 Apr 2006 15:02:02 +0000 (15:02 +0000)]
Got rid of duplicate rsync: prefix from an rsyserr() call.

18 years agoImproved a couple sentences.
Wayne Davison [Sun, 2 Apr 2006 14:23:22 +0000 (14:23 +0000)]
Improved a couple sentences.

18 years ago- Mention the logging fix for a single-use daemon.
Wayne Davison [Fri, 31 Mar 2006 17:56:19 +0000 (17:56 +0000)]
- Mention the logging fix for a single-use daemon.
- Mention the fatal error for when --daemon can't be honored.

18 years agoLog the connect message after calling log_init().
Wayne Davison [Fri, 31 Mar 2006 17:37:21 +0000 (17:37 +0000)]
Log the connect message after calling log_init().

18 years ago- Complain if the user specified --delete without -r or -d.
Wayne Davison [Tue, 28 Mar 2006 23:09:36 +0000 (23:09 +0000)]
- Complain if the user specified --delete without -r or -d.
- Got rid of the saw_delete_* vars.