Commit | Line | Data |
---|---|---|
46ef7d1d | 1 | -*- indented-text -*- |
a0365806 | 2 | |
259c3e72 MP |
3 | BUGS --------------------------------------------------------------- |
4 | ||
f9b9e2f0 MP |
5 | rsync-url barfs on upload |
6 | ||
7 | rsync foo rsync://localhost/transfer/ | |
8 | ||
9 | Fix the parser. | |
10 | ||
11 | ||
259c3e72 MP |
12 | There seems to be a bug with hardlinks |
13 | ||
14 | mbp/2 build$ ls -l /tmp/a /tmp/b -i | |
15 | /tmp/a: | |
16 | total 32 | |
17 | 2568307 -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a1 | |
18 | 2568307 -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a2 | |
19 | 2568307 -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a3 | |
20 | 2568310 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 a4 | |
21 | 2568310 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 a5 | |
22 | 2568310 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b1 | |
23 | 2568310 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b2 | |
24 | 2568310 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b3 | |
25 | ||
26 | /tmp/b: | |
27 | total 32 | |
28 | 2568309 -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a1 | |
29 | 2568309 -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a2 | |
30 | 2568309 -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a3 | |
31 | 2568311 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 a4 | |
32 | 2568311 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 a5 | |
33 | 2568311 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b1 | |
34 | 2568311 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b2 | |
35 | 2568311 -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b3 | |
36 | mbp/2 build$ rm -r /tmp/b && ./rsync -avH /tmp/a/ /tmp/b | |
37 | building file list ... done | |
38 | created directory /tmp/b | |
39 | ./ | |
40 | a1 | |
41 | a4 | |
42 | a2 => a1 | |
43 | a3 => a2 | |
44 | wrote 350 bytes read 52 bytes 804.00 bytes/sec | |
45 | total size is 232 speedup is 0.58 | |
46 | mbp/2 build$ rm -r /tmp/b | |
47 | mbp/2 build$ ls -l /tmp/b | |
48 | ls: /tmp/b: No such file or directory | |
49 | mbp/2 build$ rm -r /tmp/b && ./rsync -avH /tmp/a/ /tmp/b | |
50 | rm: cannot remove `/tmp/b': No such file or directory | |
51 | mbp/2 build$ rm -f -r /tmp/b && ./rsync -avH /tmp/a/ /tmp/b | |
52 | building file list ... done | |
53 | created directory /tmp/b | |
54 | ./ | |
55 | a1 | |
56 | a4 | |
57 | a2 => a1 | |
58 | a3 => a2 | |
59 | wrote 350 bytes read 52 bytes 804.00 bytes/sec | |
60 | total size is 232 speedup is 0.58 | |
61 | mbp/2 build$ ls -l /tmp/b | |
62 | total 32 | |
63 | -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a1 | |
64 | -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a2 | |
65 | -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a3 | |
66 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 a4 | |
67 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 a5 | |
68 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b1 | |
69 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b2 | |
70 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b3 | |
71 | mbp/2 build$ ls -l /tmp/a | |
72 | total 32 | |
73 | -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a1 | |
74 | -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a2 | |
75 | -rw-rw-r-- 3 mbp mbp 29 Mar 25 17:30 a3 | |
76 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 a4 | |
77 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 a5 | |
78 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b1 | |
79 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b2 | |
80 | -rw-rw-r-- 5 mbp mbp 29 Mar 25 17:30 b3 | |
46ef7d1d | 81 | |
33d213bb | 82 | |
e4724e5c MP |
83 | Progress indicator can produce corrupt output when transferring directories: |
84 | ||
85 | main/binary-arm/ | |
86 | main/binary-arm/admin/ | |
87 | main/binary-arm/base/ | |
88 | main/binary-arm/comm/8.56kB/s 0:00:52 | |
89 | main/binary-arm/devel/ | |
90 | main/binary-arm/doc/ | |
91 | main/binary-arm/editors/ | |
92 | main/binary-arm/electronics/s 0:00:53 | |
93 | main/binary-arm/games/ | |
94 | main/binary-arm/graphics/ | |
95 | main/binary-arm/hamradio/ | |
96 | main/binary-arm/interpreters/ | |
97 | main/binary-arm/libs/6.61kB/s 0:00:54 | |
98 | main/binary-arm/mail/ | |
99 | main/binary-arm/math/ | |
100 | main/binary-arm/misc/ | |
101 | ||
7e28fca1 | 102 | |
e4724e5c | 103 | lchmod |
e4724e5c | 104 | I don't think we handle this properly on systems that don't have the |
7e28fca1 MP |
105 | call. Are there any such? |
106 | ||
e4724e5c | 107 | |
5ba268ef MP |
108 | Cross-test versions |
109 | Part of the regression suite should be making sure that we don't | |
110 | break backwards compatibility: old clients vs new servers and so | |
111 | on. Ideally we would test the cross product of versions. | |
112 | ||
113 | It might be sufficient to test downloads from well-known public | |
114 | rsync servers running different versions of rsync. This will give | |
115 | some testing and also be the most common case for having different | |
116 | versions and not being able to upgrade. | |
117 | ||
f5e4eadb MP |
118 | --no-blocking-io might be broken |
119 | ||
120 | in the same way as --no-whole-file; somebody needs to check. | |
121 | ||
8bd1a73e MP |
122 | Do not rely on having a group called "nobody" |
123 | ||
124 | http://www.linuxbase.org/spec/refspecs/LSB_1.1.0/gLSB/usernames.html | |
125 | ||
126 | On Debian it's "nogroup" | |
e4724e5c | 127 | |
3d90ec14 MP |
128 | DAEMON -------------------------------------------------------------- |
129 | ||
130 | server-imposed bandwidth limits | |
131 | ||
132 | rsyncd over ssh | |
133 | ||
134 | There are already some patches to do this. | |
135 | ||
136 | BitKeeper uses a server whose login shell is set to bkd. That's | |
137 | probably a reasonable approach. | |
138 | ||
139 | ||
140 | FEATURES ------------------------------------------------------------ | |
33d213bb | 141 | |
33d213bb | 142 | |
32e83406 | 143 | --dry-run is too dry |
642a979a MP |
144 | |
145 | Mark Santcroos points out that -n fails to list files which have | |
a577af90 | 146 | only metadata changes, though it probably should. |
642a979a MP |
147 | |
148 | There may be a Debian bug about this as well. | |
149 | ||
150 | ||
a2d2e5c0 MP |
151 | use chroot |
152 | ||
153 | If the platform doesn't support it, then don't even try. | |
154 | ||
155 | If running as non-root, then don't fail, just give a warning. | |
156 | (There was a thread about this a while ago?) | |
157 | ||
158 | http://lists.samba.org/pipermail/rsync/2001-August/thread.html | |
159 | http://lists.samba.org/pipermail/rsync/2001-September/thread.html | |
160 | ||
642a979a | 161 | |
595f2d4d MP |
162 | supplementary groups |
163 | ||
164 | Perhaps allow supplementary groups to be specified in rsyncd.conf; | |
165 | then make the first one the primary gid and all the rest be | |
166 | supplementary gids. | |
167 | ||
168 | ||
8f4455f2 MP |
169 | File list structure in memory |
170 | ||
171 | Rather than one big array, perhaps have a tree in memory mirroring | |
a577af90 | 172 | the directory tree. |
8f4455f2 MP |
173 | |
174 | This might make sorting much faster! (I'm not sure it's a big CPU | |
a577af90 | 175 | problem, mind you.) |
8f4455f2 MP |
176 | |
177 | It might also reduce memory use in storing repeated directory names | |
a577af90 | 178 | -- again I'm not sure this is a problem. |
0e5a1f83 | 179 | |
a2d2e5c0 MP |
180 | Performance |
181 | ||
182 | Traverse just one directory at a time. Tridge says it's possible. | |
a6a3c3df MP |
183 | |
184 | At the moment rsync reads the whole file list into memory at the | |
185 | start, which makes us use a lot of memory and also not pipeline | |
186 | network access as much as we could. | |
187 | ||
0e5a1f83 MP |
188 | |
189 | Handling duplicate names | |
190 | ||
b3e6c815 | 191 | We need to be careful of duplicate names getting into the file list. |
d2e9d069 MP |
192 | See clean_flist(). This could happen if multiple arguments include |
193 | the same file. Bad. | |
b3e6c815 MP |
194 | |
195 | I think duplicates are only a problem if they're both flowing | |
196 | through the pipeline at the same time. For example we might have | |
197 | updated the first occurrence after reading the checksums for the | |
198 | second. So possibly we just need to make sure that we don't have | |
a577af90 | 199 | both in the pipeline at the same time. |
b3e6c815 MP |
200 | |
201 | Possibly if we did one directory at a time that would be sufficient. | |
202 | ||
203 | Alternatively we could pre-process the arguments to make sure no | |
58379559 MP |
204 | duplicates will ever be inserted. There could be some bad cases |
205 | when we're collapsing symlinks. | |
b3e6c815 MP |
206 | |
207 | We could have a hash table. | |
208 | ||
d2e9d069 MP |
209 | The root of the problem is that we do not want more than one file |
210 | list entry referring to the same file. At first glance there are | |
211 | several ways this could happen: symlinks, hardlinks, and repeated | |
212 | names on the command line. | |
213 | ||
214 | If names are repeated on the command line, they may be present in | |
215 | different forms, perhaps by traversing directory paths in different | |
216 | ways, traversing paths including symlinks. Also we need to allow | |
217 | for expansion of globs by rsync. | |
218 | ||
219 | At the moment, clean_flist() requires having the entire file list in | |
220 | memory. Duplicate names are detected just by a string comparison. | |
221 | ||
222 | We don't need to worry about hard links causing duplicates because | |
223 | files are never updated in place. Similarly for symlinks. | |
224 | ||
225 | I think even if we're using a different symlink mode we don't need | |
226 | to worry. | |
227 | ||
0e5a1f83 MP |
228 | Unless we're really clever this will introduce a protocol |
229 | incompatibility, so we need to be able to accept the old format as | |
230 | well. | |
231 | ||
232 | ||
a6a3c3df MP |
233 | Memory accounting |
234 | ||
235 | At exit, show how much memory was used for the file list, etc. | |
236 | ||
b3e6c815 MP |
237 | Also we do a wierd exponential-growth allocation in flist.c. I'm |
238 | not sure this makes sense with modern mallocs. At any rate it will | |
239 | make us allocate a huge amount of memory for large file lists. | |
240 | ||
0e5a1f83 | 241 | |
a6a3c3df MP |
242 | Hard-link handling |
243 | ||
244 | At the moment hardlink handling is very expensive, so it's off by | |
a577af90 | 245 | default. It does not need to be so. |
a6a3c3df | 246 | |
0e5a1f83 MP |
247 | Since most of the solutions are rather intertwined with the file |
248 | list it is probably better to fix that first, although fixing | |
249 | hardlinks is possibly simpler. | |
250 | ||
a6a3c3df MP |
251 | We can rule out hardlinked directories since they will probably |
252 | screw us up in all kinds of ways. They simply should not be used. | |
253 | ||
254 | At the moment rsync only cares about hardlinks to regular files. I | |
255 | guess you could also use them for sockets, devices and other beasts, | |
256 | but I have not seen them. | |
257 | ||
258 | When trying to reproduce hard links, we only need to worry about | |
a577af90 | 259 | files that have more than one name (nlinks>1 && !S_ISDIR). |
a6a3c3df MP |
260 | |
261 | The basic point of this is to discover alternate names that refer to | |
262 | the same file. All operations, including creating the file and | |
263 | writing modifications to it need only to be done for the first name. | |
264 | For all later names, we just create the link and then leave it | |
265 | alone. | |
266 | ||
267 | If hard links are to be preserved: | |
268 | ||
269 | Before the generator/receiver fork, the list of files is received | |
270 | from the sender (recv_file_list), and a table for detecting hard | |
271 | links is built. | |
272 | ||
273 | The generator looks for hard links within the file list and does | |
274 | not send checksums for them, though it does send other metadata. | |
275 | ||
276 | The sender sends the device number and inode with file entries, so | |
277 | that files are uniquely identified. | |
278 | ||
279 | The receiver goes through and creates hard links (do_hard_links) | |
280 | after all data has been written, but before directory permissions | |
281 | are set. | |
282 | ||
283 | At the moment device and inum are sent as 4-byte integers, which | |
284 | will probably cause problems on large filesystems. On Linux the | |
285 | kernel uses 64-bit ino_t's internally, and people will soon have | |
286 | filesystems big enough to use them. We ought to follow NFS4 in | |
287 | using 64-bit device and inode identification, perhaps with a | |
288 | protocol version bump. | |
289 | ||
290 | Once we've seen all the names for a particular file, we no longer | |
291 | need to think about it and we can deallocate the memory. | |
292 | ||
293 | We can also have the case where there are links to a file that are | |
294 | not in the tree being transferred. There's nothing we can do about | |
295 | that. Because we rename the destination into place after writing, | |
296 | any hardlinks to the old file are always going to be orphaned. In | |
297 | fact that is almost necessary because otherwise we'd get really | |
298 | confused if we were generating checksums for one name of a file and | |
299 | modifying another. | |
300 | ||
301 | At the moment the code seems to make a whole second copy of the file | |
302 | list, which seems unnecessary. | |
303 | ||
304 | We should have a test case that exercises hard links. Since it | |
305 | might be hard to compare ./tls output where the inodes change we | |
306 | might need a little program to check whether several names refer to | |
307 | the same file. | |
a2d2e5c0 | 308 | |
a2d2e5c0 | 309 | |
bde47ca7 | 310 | |
411acbbc | 311 | Handling IPv6 on old machines |
bde47ca7 | 312 | |
411acbbc MP |
313 | The KAME IPv6 patch is nice in theory but has proved a bit of a |
314 | nightmare in practice. The basic idea of their patch is that rsync | |
315 | is rewritten to use the new getaddrinfo()/getnameinfo() interface, | |
316 | rather than gethostbyname()/gethostbyaddr() as in rsync 2.4.6. | |
317 | Systems that don't have the new interface are handled by providing | |
318 | our own implementation in lib/, which is selectively linked in. | |
c7d692c3 | 319 | |
411acbbc MP |
320 | The problem with this is that it is really hard to get right on |
321 | platforms that have a half-working implementation, so redefining | |
322 | these functions clashes with system headers, and leaving them out | |
323 | breaks. This affects at least OSF/1, RedHat 5, and Cobalt, which | |
324 | are moderately improtant. | |
325 | ||
326 | Perhaps the simplest solution would be to have two different files | |
327 | implementing the same interface, and choose either the new or the | |
328 | old API. This is probably necessary for systems that e.g. have | |
329 | IPv6, but gethostbyaddr() can't handle it. The Linux manpage claims | |
330 | this is currently the case. | |
331 | ||
332 | In fact, our internal sockets interface (things like | |
333 | open_socket_out(), etc) is much narrower than the getaddrinfo() | |
334 | interface, and so probably simpler to get right. In addition, the | |
335 | old code is known to work well on old machines. | |
336 | ||
337 | We could drop the rather large lib/getaddrinfo files. | |
338 | ||
339 | ||
340 | Other IPv6 stuff: | |
341 | ||
c33e3e39 MP |
342 | Implement suggestions from http://www.kame.net/newsletter/19980604/ |
343 | and ftp://ftp.iij.ad.jp/pub/RFC/rfc2553.txt | |
344 | ||
345 | If a host has multiple addresses, then listen try to connect to all | |
346 | in order until we get through. (getaddrinfo may return multiple | |
c10b0bdd | 347 | addresses.) This is kind of implemented already. |
c33e3e39 MP |
348 | |
349 | Possibly also when starting as a server we may need to listen on | |
350 | multiple passive addresses. This might be a bit harder, because we | |
351 | may need to select on all of them. Hm. | |
352 | ||
a2d2e5c0 MP |
353 | Define a syntax for IPv6 literal addresses. Since they include |
354 | colons, they tend to break most naming systems, including ours. | |
355 | Based on the HTTP IPv6 syntax, I think we should use | |
356 | ||
a577af90 | 357 | rsync://[::1]/foo/bar [::1]::bar |
a2d2e5c0 MP |
358 | |
359 | which should just take a small change to the parser code. | |
360 | ||
b17dd0c4 | 361 | |
5aafd07b MP |
362 | Errors |
363 | ||
364 | If we hang or get SIGINT, then explain where we were up to. Perhaps | |
365 | have a static buffer that contains the current function name, or | |
366 | some kind of description of what we were trying to do. This is a | |
367 | little easier on people than needing to run strace/truss. | |
368 | ||
369 | "The dungeon collapses! You are killed." Rather than "unexpected | |
370 | eof" give a message that is more detailed if possible and also more | |
a577af90 | 371 | helpful. |
5aafd07b | 372 | |
89b0a3d9 MP |
373 | If we get an error writing to a socket, then we should perhaps |
374 | continue trying to read to see if an error message comes across | |
375 | explaining why the socket is closed. I'm not sure if this would | |
376 | work, but it would certainly make our messages more helpful. | |
377 | ||
b17dd0c4 | 378 | What happens if a directory is missing -x attributes. Do we lose |
a577af90 PG |
379 | our load? (Debian #28416) Probably fixed now, but a test case would |
380 | be good. | |
b17dd0c4 MP |
381 | |
382 | ||
5575de14 MP |
383 | File attributes |
384 | ||
385 | Device major/minor numbers should be at least 32 bits each. See | |
386 | http://lists.samba.org/pipermail/rsync/2001-November/005357.html | |
387 | ||
388 | Transfer ACLs. Need to think of a standard representation. | |
389 | Probably better not to even try to convert between NT and POSIX. | |
390 | Possibly can share some code with Samba. | |
5aafd07b | 391 | |
28a69e25 MP |
392 | Empty directories |
393 | ||
394 | With the current common --include '*/' --exclude '*' pattern, people | |
395 | can end up with many empty directories. We might avoid this by | |
396 | lazily creating such directories. | |
397 | ||
c6e27b60 | 398 | |
28a69e25 MP |
399 | zlib |
400 | ||
a577af90 | 401 | Perhaps don't use our own zlib. |
c6e27b60 MP |
402 | |
403 | Advantages: | |
404 | ||
405 | - will automatically be up to date with bugfixes in zlib | |
406 | ||
407 | - can leave it out for small rsync on e.g. recovery disks | |
408 | ||
409 | - can use a shared library | |
410 | ||
411 | - avoids people breaking rsync by trying to do this themselves and | |
412 | messing up | |
413 | ||
414 | Should we ship zlib for systems that don't have it, or require | |
415 | people to install it separately? | |
416 | ||
417 | Apparently this will make us incompatible with versions of rsync | |
418 | that use the patched version of rsync. Probably the simplest way to | |
419 | do this is to just disable gzip (with a warning) when talking to old | |
420 | versions. | |
421 | ||
5ba268ef MP |
422 | |
423 | logging | |
424 | ||
425 | Perhaps flush stdout after each filename, so that people trying to | |
426 | monitor progress in a log file can do so more easily. See | |
427 | http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=48108 | |
428 | ||
429 | At the connections that just get a list of modules are not logged, | |
430 | but they should be. | |
431 | ||
432 | If a child of the rsync daemon dies with a signal, we should notice | |
433 | that when we reap it and log a message. | |
434 | ||
435 | Keep stderr and stdout properly separated (Debian #23626) | |
436 | ||
595f2d4d MP |
437 | After we get the @RSYNCD greeting from the server, we know it's |
438 | version but we have not yet sent the command line, so we could just | |
a577af90 | 439 | remove the -z option if the server is too old. |
595f2d4d MP |
440 | |
441 | For ssh invocation it's not so simple, because we actually use the | |
442 | command line to start the remote process. However, we only actually | |
443 | do compression in token.c, and we could therefore once we discover | |
444 | the remote version emit an error if it's too old. I'm not sure if | |
445 | that's a good tradeoff or not. | |
446 | ||
28a69e25 | 447 | |
92325ada MP |
448 | proxy authentication |
449 | ||
450 | Allow RSYNC_PROXY to be http://user:pass@proxy.foo:3128/, and do | |
a577af90 | 451 | HTTP Basic Proxy-Authentication. |
92325ada MP |
452 | |
453 | Multiple schemes are possible, up to and including the insanity that | |
454 | is NTLM, but Basic probably covers most cases. | |
455 | ||
456 | SOCKS | |
457 | ||
458 | Add --with-socks, and then perhaps a command-line option to put them | |
459 | on or off. This might be more reliable than LD_PRELOAD hacks. | |
460 | ||
5ba268ef MP |
461 | FAT support |
462 | ||
a577af90 PG |
463 | rsync to a FAT partition on a Unix machine doesn't work very well at |
464 | the moment. I think we get errors about invalid filenames and | |
5ba268ef MP |
465 | perhaps also trying to do atomic renames. |
466 | ||
a577af90 PG |
467 | I guess the code to do this is currently #ifdef'd on Windows; |
468 | perhaps we ought to intelligently fall back to it on Unix too. | |
5ba268ef MP |
469 | |
470 | ||
27741d9f MP |
471 | Better statistics: |
472 | ||
473 | <Rasmus> mbp: hey, how about an rsync option that just gives you the | |
474 | summary without the list of files? And perhaps gives more | |
475 | information like the number of new files, number of changed, | |
a577af90 PG |
476 | deleted, etc. ? <mbp> Rasmus: nice idea <mbp> there is --stats |
477 | <mbp> but at the moment it's very tridge-oriented <mbp> rather than | |
478 | user-friendly <mbp> it would be nice to improve it <mbp> that would | |
479 | also work well with --dryrun | |
27741d9f | 480 | |
e53fe9a2 MP |
481 | TDB: |
482 | ||
483 | Rather than storing the file list in memory, store it in a TDB. | |
484 | ||
485 | This *might* make memory usage lower while building the file list. | |
486 | ||
487 | Hashtable lookup will mean files are not transmitted in order, | |
488 | though... hm. | |
489 | ||
490 | This would neatly eliminate one of the major post-fork shared data | |
491 | structures. | |
492 | ||
493 | ||
97e1254a MP |
494 | chmod: |
495 | ||
a577af90 PG |
496 | On 12 Mar 2002, Dave Dykstra <dwd@bell-labs.com> wrote: > If we |
497 | would add an option to do that functionality, I would vote for one > | |
498 | that was more general which could mask off any set of permission bits | |
499 | and > possibly add any set of bits. Perhaps a chmod-like syntax if it | |
500 | could be > implemented simply. | |
97e1254a | 501 | |
a577af90 | 502 | I think that would be good too. For example, people uploading files |
97e1254a MP |
503 | to a web server might like to say |
504 | ||
505 | rsync -avzP --chmod a+rX ./ sourcefrog.net:/home/www/sourcefrog/ | |
506 | ||
507 | Ideally the patch would implement as many of the gnu chmod semantics | |
508 | as possible. I think the mode parser should be a separate function | |
a577af90 PG |
509 | that passes back something like (mask,set) description to the rest |
510 | of the program. For bonus points there would be a test case for the | |
97e1254a MP |
511 | parser. |
512 | ||
8bd1a73e MP |
513 | Possibly also --chown |
514 | ||
36692011 MP |
515 | (Debian #23628) |
516 | ||
97e1254a | 517 | |
3c1edccb MP |
518 | --diff |
519 | ||
520 | Allow people to specify the diff command. (Might want to use wdiff, | |
521 | gnudiff, etc.) | |
522 | ||
523 | Just diff the temporary file with the destination file, and delete | |
524 | the tmp file rather than moving it into place. | |
525 | ||
526 | Interaction with --partial. | |
527 | ||
528 | Security interactions with daemon mode? | |
529 | ||
530 | (Suggestion from david.e.sewell) | |
531 | ||
532 | ||
a628b069 MP |
533 | Incorrect timestamps (Debian #100295) |
534 | ||
535 | A bit hard to believe, but apparently it happens. | |
6d19c674 MP |
536 | |
537 | ||
538 | Check "refuse options works" | |
539 | ||
540 | We need a test case for this... | |
541 | ||
542 | Was this broken when we changed to popt? | |
6479c2ed MP |
543 | |
544 | ||
bd0ad74f MP |
545 | PERFORMANCE ---------------------------------------------------------- |
546 | ||
547 | MD4 file_sum | |
548 | ||
549 | If we're doing a local transfer, or using -W, then perhaps don't | |
550 | send the file checksum. If we're doing a local transfer, then | |
551 | calculating MD4 checksums uses 90% of CPU and is unlikely to be | |
552 | useful. | |
553 | ||
554 | Indeed for transfers over zlib or ssh we can also rely on the | |
555 | transport to have quite strong protection against corruption. | |
556 | ||
557 | Perhaps we should have an option to disable this, analogous to | |
a577af90 PG |
558 | --whole-file, although it would default to disabled. The file |
559 | checksum takes up a definite space in the protocol -- we can either | |
560 | set it to 0, or perhaps just leave it out. | |
bd0ad74f MP |
561 | |
562 | MD4 | |
563 | ||
564 | Perhaps borrow an assembler MD4 from someone? | |
565 | ||
566 | Make sure we call MD4 with properly-sized blocks whenever possible | |
567 | to avoid copying into the residue region? | |
568 | ||
6479c2ed MP |
569 | String area code |
570 | ||
571 | Test whether this is actually faster than just using malloc(). If | |
572 | it's not (anymore), throw it out. | |
a628b069 MP |
573 | |
574 | ||
a2d2e5c0 MP |
575 | PLATFORMS ------------------------------------------------------------ |
576 | ||
577 | Win32 | |
578 | ||
579 | Don't detach, because this messes up --srvany. | |
580 | ||
581 | http://sources.redhat.com/ml/cygwin/2001-08/msg00234.html | |
582 | ||
5ba268ef | 583 | |
0e23e41d MP |
584 | DEVELOPMENT ---------------------------------------------------------- |
585 | ||
586 | Splint | |
587 | ||
588 | Build rsync with SPLINT to try to find security holes. Add | |
589 | annotations as necessary. Keep track of the number of warnings | |
590 | found initially, and see how many of them are real bugs, or real | |
591 | security bugs. Knowing the percentage of likely hits would be | |
592 | really interesting for other projects. | |
593 | ||
f5a95bb5 MP |
594 | Torture test |
595 | ||
596 | Something that just keeps running rsync continuously over a data set | |
597 | likely to generate problems. | |
598 | ||
599 | Cross-testing | |
600 | ||
601 | Run current rsync versions against significant past releases. | |
602 | ||
43a4dc10 MP |
603 | Memory debugger |
604 | ||
3a79260d | 605 | jra recommends Valgrind: |
43a4dc10 MP |
606 | |
607 | http://devel-home.kde.org/~sewardj/ | |
608 | ||
25ff30e8 MP |
609 | Release script |
610 | ||
611 | Update spec files | |
612 | ||
613 | Build tar file; upload | |
614 | ||
615 | Send announcement to mailing list and c.o.l.a. | |
616 | ||
617 | Make freshmeat announcement | |
618 | ||
619 | Update web site | |
620 | ||
621 | ||
622 | ||
e9c4c301 MP |
623 | TESTING -------------------------------------------------------------- |
624 | ||
625 | Cross-test versions | |
626 | ||
627 | Part of the regression suite should be making sure that we don't | |
a577af90 PG |
628 | break backwards compatibility: old clients vs new servers and so on. |
629 | Ideally we would test both up and down from the current release to | |
630 | all old versions. | |
25ff30e8 MP |
631 | |
632 | We might need to omit broken old versions, or versions in which | |
633 | particular functionality is broken | |
e9c4c301 MP |
634 | |
635 | It might be sufficient to test downloads from well-known public | |
636 | rsync servers running different versions of rsync. This will give | |
637 | some testing and also be the most common case for having different | |
638 | versions and not being able to upgrade. | |
639 | ||
25ff30e8 MP |
640 | |
641 | Test on kernel source | |
642 | ||
643 | Download all versions of kernel; unpack, sync between them. Also | |
644 | sync between uncompressed tarballs. Compare directories after | |
645 | transfer. | |
646 | ||
647 | Use local mode; ssh; daemon; --whole-file and --no-whole-file. | |
648 | ||
649 | Use awk to pull out the 'speedup' number for each transfer. Make | |
650 | sure it is >= x. | |
651 | ||
652 | ||
e9c4c301 MP |
653 | Test large files |
654 | ||
655 | Sparse and non-sparse | |
656 | ||
657 | Mutator program | |
658 | ||
659 | Insert bytes, delete bytes, swap blocks, ... | |
660 | ||
661 | configure option to enable dangerous tests | |
662 | ||
663 | If tests are skipped, say why. | |
664 | ||
b73b51a9 MP |
665 | Test daemon feature to disallow particular options. |
666 | ||
599dc93c MP |
667 | Pipe program that makes slow/jerky connections. |
668 | ||
a577af90 PG |
669 | Versions of read() and write() that corrupt the stream, or abruptly |
670 | fail | |
76533c52 | 671 | |
6479c2ed MP |
672 | Separate makefile target to run rough tests -- or perhaps just run |
673 | them every time? | |
674 | ||
717eb9b8 MP |
675 | Test "refuse options" works |
676 | ||
677 | What about for --recursive? | |
678 | ||
679 | If you specify an unrecognized option here, you should get an error. | |
680 | ||
e9c4c301 | 681 | |
7c583c73 MP |
682 | DOCUMENTATION -------------------------------------------------------- |
683 | ||
684 | Update README | |
685 | ||
b73b51a9 MP |
686 | Keep list of open issues and todos on the web site |
687 | ||
688 | Update web site from CVS | |
689 | ||
5af50297 MP |
690 | |
691 | Perhaps redo manual as SGML | |
692 | ||
693 | The man page is getting rather large, and there is more information | |
694 | that ought to be added. | |
695 | ||
696 | TexInfo source is probably a dying format. | |
697 | ||
698 | Linuxdoc looks like the most likely contender. I know DocBook is | |
699 | favoured by some people, but it's so bloody verbose, even with emacs | |
700 | support. | |
701 | ||
702 | ||
a2d2e5c0 MP |
703 | BUILD FARM ----------------------------------------------------------- |
704 | ||
705 | Add machines | |
706 | ||
a2d2e5c0 MP |
707 | Cygwin (on different versions of Win32?) |
708 | ||
709 | HP-UX variants (via HP?) | |
33d213bb | 710 | |
5aafd07b MP |
711 | SCO |
712 | ||
46ef7d1d | 713 | |
62b68c80 | 714 | LOGGING -------------------------------------------------------------- |
a2d2e5c0 | 715 | |
62b68c80 MP |
716 | Perhaps flush stdout after each filename, so that people trying to |
717 | monitor progress in a log file can do so more easily. See | |
718 | http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=48108 | |
a2d2e5c0 | 719 | |
62b68c80 MP |
720 | At the connections that just get a list of modules are not logged, |
721 | but they should be. | |
722 | ||
723 | If a child of the rsync daemon dies with a signal, we should notice | |
724 | that when we reap it and log a message. | |
725 | ||
726 | Keep stderr and stdout properly separated (Debian #23626) | |
727 | ||
728 | Use a separate function for reporting errors; prefix it with | |
729 | "rsync:" or "rsync(remote)", or perhaps even "rsync(local | |
730 | generator): ". | |
a2d2e5c0 | 731 | |
50f2f002 MP |
732 | verbose output |
733 | ||
734 | Indicate whether files are new, updated, or deleted | |
735 | ||
d834adc1 MP |
736 | At end of transfer, show how many files were or were not transferred |
737 | correctly. | |
738 | ||
62b68c80 MP |
739 | -vv |
740 | ||
741 | Explain *why* every file is transferred or not (e.g. "local mtime | |
742 | 123123 newer than 1283198") | |
743 | ||
744 | ||
8ff9d697 MP |
745 | debugging of daemon |
746 | ||
747 | Add an rsyncd.conf parameter to turn on debugging on the server. | |
748 | ||
749 | ||
62b68c80 MP |
750 | |
751 | NICE ----------------------------------------------------------------- | |
752 | ||
753 | --no-detach and --no-fork options | |
754 | ||
755 | Very useful for debugging. Also good when running under a | |
756 | daemon-monitoring process that tries to restart the service when the | |
757 | parent exits. | |
758 | ||
759 | hang/timeout friendliness | |
3d90ec14 | 760 | |
a2d2e5c0 MP |
761 | internationalization |
762 | ||
763 | Change to using gettext(). Probably need to ship this for platforms | |
a577af90 | 764 | that don't have it. |
a2d2e5c0 MP |
765 | |
766 | Solicit translations. | |
767 | ||
3d90ec14 MP |
768 | Does anyone care? Before we bother modifying the code, we ought to |
769 | get the manual translated first, because that's possibly more useful | |
770 | and at any rate demonstrates desire. | |
771 | ||
a577af90 | 772 | rsyncsh |
46ef7d1d MP |
773 | |
774 | Write a small emulation of interactive ftp as a Pythonn program | |
775 | that calls rsync. Commands such as "cd", "ls", "ls *.c" etc map | |
776 | fairly directly into rsync commands: it just needs to remember the | |
777 | current host, directory and so on. We can probably even do | |
778 | completion of remote filenames. | |
25ff30e8 MP |
779 | |
780 | ||
781 | RELATED PROJECTS ----------------------------------------------------- | |
782 | ||
783 | http://rsync.samba.org/rsync-and-debian/ | |
784 | ||
785 | rsyncable gzip patch | |
786 | ||
787 | Exhaustive, tortuous testing | |
788 | ||
789 | Cleanups? | |
790 | ||
791 | rsyncsplit as alternative to real integration with gzip? | |
792 | ||
793 | reverse rsync over HTTP Range | |
794 | ||
795 | Goswin Brederlow suggested this on Debian; I think tridge and I | |
796 | talked about it previous in relation to rproxy. | |
a577af90 PG |
797 | |
798 |