Martin Pool [Thu, 24 Jan 2002 08:08:56 +0000 (08:08 +0000)]
write_batch_csums_file: Opaque IO buffers should be void*.
Martin Pool [Thu, 24 Jan 2002 08:07:35 +0000 (08:07 +0000)]
Fix cast that was breaking HP/UX.
Martin Pool [Thu, 24 Jan 2002 08:05:22 +0000 (08:05 +0000)]
Move both calls to getpeername into a common wrapper function that
handles IPV4_MAPPED addresses.
Martin Pool [Thu, 24 Jan 2002 07:22:38 +0000 (07:22 +0000)]
Must get declarations from addrinfo.h before prototypes.
Martin Pool [Thu, 24 Jan 2002 05:57:49 +0000 (05:57 +0000)]
Another size_t warning
Martin Pool [Thu, 24 Jan 2002 05:57:22 +0000 (05:57 +0000)]
check_name: Print out name *before* clobbering it.
Martin Pool [Thu, 24 Jan 2002 05:54:47 +0000 (05:54 +0000)]
Try to fix error on Solaris:
+ [ -f /export/home/build/build_farm/rsync/testtmp.symlink-ignore/to/referent ]
+ [ -d /export/home/build/build_farm/rsync/testtmp.symlink-ignore/to/from ]
+ [ -L /export/home/build/build_farm/rsync/testtmp.symlink-ignore/to/dangling ]
./testsuite/symlink-ignore.test: test: argument expected
Solaris says it supports -L, so I'm not sure what's happening...
Martin Pool [Thu, 24 Jan 2002 05:41:46 +0000 (05:41 +0000)]
If name lookup fails, then show the relevant IP address in the error message.
Martin Pool [Thu, 24 Jan 2002 04:49:07 +0000 (04:49 +0000)]
At the connections that just get a list of modules are not logged,
but they should be.
Martin Pool [Thu, 24 Jan 2002 04:41:09 +0000 (04:41 +0000)]
rsync_module: If host-based access fails, show the exact name/address
used for the check in the error message. (Just in case...)
Martin Pool [Thu, 24 Jan 2002 04:36:00 +0000 (04:36 +0000)]
Show command used to start connection child.
Martin Pool [Thu, 24 Jan 2002 04:26:55 +0000 (04:26 +0000)]
indent -kr -i8
Martin Pool [Thu, 24 Jan 2002 04:24:12 +0000 (04:24 +0000)]
Doc.
Oops, connection program message was in the wrong place.
Martin Pool [Thu, 24 Jan 2002 04:21:42 +0000 (04:21 +0000)]
Show helper program with -v when opening connection.
Doc.
Martin Pool [Thu, 24 Jan 2002 04:19:41 +0000 (04:19 +0000)]
Doc.
Martin Pool [Thu, 24 Jan 2002 04:07:07 +0000 (04:07 +0000)]
Fix const
Martin Pool [Thu, 24 Jan 2002 04:03:06 +0000 (04:03 +0000)]
Unbreak the old behavior of using UNKNOWN as a hostname if any of the
addr->name->addr translations fail, because people might count on this
in "hosts deny" lines.
Martin Pool [Thu, 24 Jan 2002 03:31:28 +0000 (03:31 +0000)]
Fix inverted sense of error check.
Martin Pool [Thu, 24 Jan 2002 03:28:20 +0000 (03:28 +0000)]
Refactor client_name() into smaller functions.
Better messages for DNS failure.
If we can get a reverse name for an IP address, but not confirm that
it is correct using a forward lookup then we still proceed to use the
name, but also emit a warning.
Martin Pool [Thu, 24 Jan 2002 03:03:20 +0000 (03:03 +0000)]
Doc.
Martin Pool [Thu, 24 Jan 2002 03:01:42 +0000 (03:01 +0000)]
Bump version to 2.5.2pre2.
Martin Pool [Thu, 24 Jan 2002 02:41:38 +0000 (02:41 +0000)]
log_formatted: Fill the log buffer with nuls to make sure we cannot
accidentally leave the string unterminated.
Martin Pool [Thu, 24 Jan 2002 02:33:45 +0000 (02:33 +0000)]
Another harmless size_t warning.
Martin Pool [Wed, 23 Jan 2002 08:08:48 +0000 (08:08 +0000)]
Doc.
Martin Pool [Wed, 23 Jan 2002 08:04:54 +0000 (08:04 +0000)]
More size_t fixes.
Martin Pool [Wed, 23 Jan 2002 07:57:43 +0000 (07:57 +0000)]
Just for variety we have some socklen_t fixes too.
Martin Pool [Wed, 23 Jan 2002 07:54:13 +0000 (07:54 +0000)]
Refactor code in send_sums to remove repeated ternaries.
Martin Pool [Wed, 23 Jan 2002 07:52:52 +0000 (07:52 +0000)]
indent -kr -i8
Martin Pool [Wed, 23 Jan 2002 07:48:35 +0000 (07:48 +0000)]
Fix more ints that ought to be size_t's.
Martin Pool [Wed, 23 Jan 2002 07:42:30 +0000 (07:42 +0000)]
Doc.
do_hard_links() actually only looks at the global hardlink table, so
it can be a (void) fn. (Another gcc warning...)
Martin Pool [Wed, 23 Jan 2002 07:36:23 +0000 (07:36 +0000)]
Fix another int that ought to be a size_t.
Martin Pool [Wed, 23 Jan 2002 07:34:26 +0000 (07:34 +0000)]
indent -kr -i8
Martin Pool [Wed, 23 Jan 2002 07:32:29 +0000 (07:32 +0000)]
Another signedness fix to quieten Sun cc warning.
Martin Pool [Wed, 23 Jan 2002 07:28:38 +0000 (07:28 +0000)]
Skip this for now; it's a known bug
Martin Pool [Wed, 23 Jan 2002 07:22:03 +0000 (07:22 +0000)]
Bump version to 2.5.2pre1.
Martin Pool [Wed, 23 Jan 2002 07:18:58 +0000 (07:18 +0000)]
Note batch-mode changes.
Martin Pool [Wed, 23 Jan 2002 07:12:57 +0000 (07:12 +0000)]
Note about proxy authentication and SOCKS.
Martin Pool [Wed, 23 Jan 2002 06:48:13 +0000 (06:48 +0000)]
Indent.
Martin Pool [Wed, 23 Jan 2002 05:59:10 +0000 (05:59 +0000)]
Update thankyou list.
Clarify email addresses.
Martin Pool [Wed, 23 Jan 2002 05:53:58 +0000 (05:53 +0000)]
Note that batch mode is currently experimental.
Martin Pool [Wed, 23 Jan 2002 05:51:06 +0000 (05:51 +0000)]
Patch from Jos Backus -- Fix breakage from dev_t to DEV64_T in batch
mode.
Also, drop -f and -F for batch mode: these should be reserved for
options that are more commonly used.
It also appends a newline to the argvs file and skips adding the
source directory to the command line.
Martin Pool [Wed, 23 Jan 2002 04:57:18 +0000 (04:57 +0000)]
Signedness security patch from Sebastian Krahmer <krahmer@suse.de> --
in some cases we were not sufficiently careful about reading integers
from the network.
Also, make sure log messages are always nul-terminated.
Martin Pool [Wed, 23 Jan 2002 03:52:06 +0000 (03:52 +0000)]
If using gcc, then also turn on -W to get even more warnings.
Remove obsolete message.
Martin Pool [Mon, 21 Jan 2002 01:57:13 +0000 (01:57 +0000)]
Don't need to reread configuration on SIGHUP because we always do that
on new connections.
David Dykstra [Tue, 15 Jan 2002 21:25:55 +0000 (21:25 +0000)]
Aack! Since released version 2.5.0, the --whole-file option was accidentally
changed to --whole. Change it back. Anybody who wants to use this option
with any version of rsync will have to switch to using -W.
Martin Pool [Tue, 15 Jan 2002 11:50:32 +0000 (11:50 +0000)]
Change gratuituous strlcat's into strlcpy, since we already know the
length of the existing string.
Martin Pool [Tue, 15 Jan 2002 11:32:30 +0000 (11:32 +0000)]
Clearer doc.
Martin Pool [Tue, 15 Jan 2002 11:20:26 +0000 (11:20 +0000)]
Add --enable-profile to turn on gprof. This is not perfect, because
it seems to only write to ./gmon.out, and that causes trouble when
there are several rsync processes in the same directory. But you can
make it work.
Martin Pool [Tue, 15 Jan 2002 10:47:59 +0000 (10:47 +0000)]
Update
Martin Pool [Tue, 15 Jan 2002 10:43:51 +0000 (10:43 +0000)]
The Ted T'so school of program optimization: make progress visible and
people will think it's faster.
So now with --progress rsync will show you how many files it has seen
as it builds the file_list.
Martin Pool [Tue, 15 Jan 2002 10:04:48 +0000 (10:04 +0000)]
Remove unused variable.
Martin Pool [Tue, 15 Jan 2002 10:04:11 +0000 (10:04 +0000)]
Oops, getaddrinfo returns an error code (not -1) for error. -Wall is
good.
Martin Pool [Tue, 15 Jan 2002 09:53:27 +0000 (09:53 +0000)]
Define _GNU_SOURCE so that we get all necessary prototypes.
Martin Pool [Tue, 15 Jan 2002 09:43:21 +0000 (09:43 +0000)]
If using GCC, try to turn on -Wall. I want to be clean with respect
to -Wall.
Martin Pool [Mon, 14 Jan 2002 00:16:51 +0000 (00:16 +0000)]
Fix for <samba.org/cgi-bin/rsync/incoming?id=3750>
temp files must be opened through do_open so that binary modes is used
on cygwin. (Chris Boucher)
Martin Pool [Sun, 13 Jan 2002 23:57:13 +0000 (23:57 +0000)]
Ignore testtmp directories.
Martin Pool [Fri, 11 Jan 2002 08:37:42 +0000 (08:37 +0000)]
DOc.
Martin Pool [Fri, 11 Jan 2002 08:25:32 +0000 (08:25 +0000)]
Always use 64-bit ino_t and dev_t internally, so that we can detect
hardlinks if coming from a larger platform. Add heaps of comments
explaining why this is so.
Martin Pool [Fri, 11 Jan 2002 08:24:34 +0000 (08:24 +0000)]
More comments about IPv6 stuff.
If a reverse name lookup fails, show the name that we were trying to
look up.
Martin Pool [Fri, 11 Jan 2002 08:08:34 +0000 (08:08 +0000)]
When checking what was copied, use specific -d -f -L flags to test
rather than -e. (Perhaps Solaris doesn't have test -e?)
Martin Pool [Fri, 11 Jan 2002 08:02:43 +0000 (08:02 +0000)]
The current version of rsync is expected to fail to eliminate all
duplicates from list.
Martin Pool [Fri, 11 Jan 2002 08:01:05 +0000 (08:01 +0000)]
Add concept of expected-failure.
Martin Pool [Fri, 11 Jan 2002 07:41:50 +0000 (07:41 +0000)]
Give cleaner output from "make check"
Martin Pool [Fri, 11 Jan 2002 07:30:50 +0000 (07:30 +0000)]
Bump version number.
Now finished merging across work from experimental BK repository.
Martin Pool [Fri, 11 Jan 2002 07:29:53 +0000 (07:29 +0000)]
Improved duplicates test: check that each file is
copied once and exactly once.
Martin Pool [Fri, 11 Jan 2002 07:29:02 +0000 (07:29 +0000)]
Check whether code to eliminate duplicate filenames works
(hint: it does not.)
Martin Pool [Fri, 11 Jan 2002 07:26:39 +0000 (07:26 +0000)]
Better mallinfo() output in --stats
Martin Pool [Fri, 11 Jan 2002 07:25:54 +0000 (07:25 +0000)]
Update copyright
Martin Pool [Fri, 11 Jan 2002 07:24:31 +0000 (07:24 +0000)]
Doc
Martin Pool [Fri, 11 Jan 2002 07:16:11 +0000 (07:16 +0000)]
Look for mallinfo() and use it to display
heap usage information in --stats
Martin Pool [Fri, 11 Jan 2002 07:15:16 +0000 (07:15 +0000)]
In protocol version 26, always
send 64-bit ino_t and dev_t. We also need to try to use 64-bit
ino_t internally *even if* this platform does not have 64-bit
inums itself, because we need to find duplicate inums when
coming from a larger platform with --hardlinks.
Martin Pool [Fri, 11 Jan 2002 07:11:43 +0000 (07:11 +0000)]
Autoindent
Add copyright
Martin Pool [Fri, 11 Jan 2002 07:11:23 +0000 (07:11 +0000)]
Improved test framework and test for hardlink handling
Martin Pool [Fri, 11 Jan 2002 07:10:25 +0000 (07:10 +0000)]
Bump version
Martin Pool [Fri, 11 Jan 2002 07:09:53 +0000 (07:09 +0000)]
Merge ChangeSet@1.12: Add test case for -H
Martin Pool [Fri, 11 Jan 2002 07:09:22 +0000 (07:09 +0000)]
Merge ChangeSet@1.12: tls now shows number of links to a file to aid in testing -H
Martin Pool [Fri, 11 Jan 2002 07:07:49 +0000 (07:07 +0000)]
Merge ChangeSet@1.10: Documentation about flist scalabilityTODO
Martin Pool [Fri, 11 Jan 2002 07:07:30 +0000 (07:07 +0000)]
Merge ChangeSet@1.9: Documentation about flist scalability
Martin Pool [Fri, 11 Jan 2002 07:05:30 +0000 (07:05 +0000)]
Merge ChangeSet@1.4: Documentation about flist scalability
Martin Pool [Fri, 11 Jan 2002 07:04:37 +0000 (07:04 +0000)]
Merge ChangeSet@1.4: Documentation about future development.
Martin Pool [Thu, 3 Jan 2002 07:18:17 +0000 (07:18 +0000)]
Oops, fix date.
rsync-bugs [Thu, 3 Jan 2002 07:11:10 +0000 (07:11 +0000)]
preparing for release of 2.5.1
Martin Pool [Thu, 3 Jan 2002 07:08:35 +0000 (07:08 +0000)]
Cleanup.
David Dykstra [Thu, 20 Dec 2001 15:33:13 +0000 (15:33 +0000)]
Removed debugging statement that was added to sig_int() by the rsync+ patch
integration. It was causing the format of the daemon log to be messed up
because of the leading \n.
Martin Pool [Thu, 20 Dec 2001 01:23:28 +0000 (01:23 +0000)]
Suggestion from David Stein
verbose output
Indicate whether files are new, updated, or deleted
Martin Pool [Tue, 18 Dec 2001 06:48:56 +0000 (06:48 +0000)]
Typo
Martin Pool [Tue, 18 Dec 2001 06:47:40 +0000 (06:47 +0000)]
Get rid of global_opts struct as suggested by Dave -- too many
problems with initialization.
Martin Pool [Tue, 18 Dec 2001 06:45:28 +0000 (06:45 +0000)]
Documentation fixes based on mail from Edward Welbourne, and an
attempted explanation of rsync's symbolic-link handling.
Martin Pool [Tue, 18 Dec 2001 06:26:26 +0000 (06:26 +0000)]
Add an "unsafe" symlink to the symlink test case so we can see what happens.
Martin Pool [Tue, 18 Dec 2001 06:25:31 +0000 (06:25 +0000)]
Note about hardlink performance.
Martin Pool [Tue, 18 Dec 2001 06:21:33 +0000 (06:21 +0000)]
Refactor code for setting local address on outgoing connections. If a
local address is specified, then try all addrinfo records for it.
Martin Pool [Tue, 18 Dec 2001 05:54:57 +0000 (05:54 +0000)]
Get rid of global_opts struct as suggested by Dave -- too many
problems with initialization.
Change the algorithm from trying to open an inbound socket with
getaddrinfo: keep trying suggested addresses until we find one on
which we can both get a socket and bind. Not convinced this is the
best, but it's probably better.
Martin Pool [Tue, 18 Dec 2001 01:33:56 +0000 (01:33 +0000)]
Note rsyncd-over-ssh and documentation TODOs.
Martin Pool [Tue, 18 Dec 2001 01:32:27 +0000 (01:32 +0000)]
Fix from Jeff Garzik for inet_ntop prototype errors on some Linux
distributions: we were failing to define HAVE_INET_NTOP, so our
prototype in rsync.h came through.
Also rerun autoheader, and have comment for HAVE_SOCKADDR_STORAGE.
Martin Pool [Tue, 18 Dec 2001 01:07:20 +0000 (01:07 +0000)]
Update README so that Paulus doesn't get bug reports anymore :-)
I think this document needs to be reworked to better explain how to
use rsync, but not right now.
David Dykstra [Fri, 14 Dec 2001 18:25:51 +0000 (18:25 +0000)]
When INET6 is not defined, meaning that IPv6 is not supported, need to
initalize the global_opts.af_hint to AF_INET or systems such as Linux that
have a native getaddrinfo() because they support IPv6 will attempt to
create IPv6 sockets. This brings up a problem with the new global_opts
structure; in order to initialize them to a value other than 0, we need to
explicitly initialize them all in an order that matches the order in
rsync.h. I think that's more inconvenient & error prone than keeping
global variables.
David Dykstra [Fri, 14 Dec 2001 18:00:54 +0000 (18:00 +0000)]
open_socket_in was attempting to try all the protocols returned from
getaddrinfo(), but only if a corresponding call to socket() returned one of
three *NOSUPPORT errno codes. A Redhat 6.2 system was observed returning
EINVAL instead so it never went on to try IPv4. This update adds EINVAL to
the list. Question: why not always continue through the list regardless of
what errno is?
Martin Pool [Fri, 14 Dec 2001 05:55:21 +0000 (05:55 +0000)]
Verbose.
Martin Pool [Fri, 14 Dec 2001 05:54:24 +0000 (05:54 +0000)]
Only show test output if it failed.
Martin Pool [Fri, 14 Dec 2001 05:52:25 +0000 (05:52 +0000)]
Be less verbose.