/[path-sandbox]/trunk/ChangeLog
Gentoo

Contents of /trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 40 - (show annotations) (download)
Mon Mar 21 23:30:13 2005 UTC (9 years, 6 months ago) by azarah
File size: 16566 byte(s)
Allow lchown a symlink in write-allowed path pointing to write-denied target.

1 # ChangeLog for Path Sandbox
2 # Copyright 1999-2004 Gentoo Foundation; Distributed under the GPL v2
3 # $Header$
4
5 22 Mar 2005; Martin Schlemmer <azarah@gentoo.org> libsandbox.c:
6 Allow lchown a symlink in write-allowed path pointing to write-denied
7 target.
8
9 21 Mar 2005; Marius Mauch <genone@gentoo.org> libsandbox.c:
10 Also show resolved symlink names in the log.
11
12 14 Mar 2005; Martin Schlemmer <azarah@gentoo.org> Makefile.am, libsandbox.c:
13 Seems -nostdlib was the problem with the constructor/destructor - remove it
14 from Makefile.am, and change the constructor/destructor names again.
15
16 14 Mar 2005; Martin Schlemmer <azarah@gentoo.org> libsandbox.c:
17 Also rename the _init() and _fini() declarations.
18
19 14 Mar 2005; Martin Schlemmer <azarah@gentoo.org> libsandbox.c, sandbox.c:
20 Fixup the constructor/destructor function names again (they should be _init()
21 and _fini() it seems, and not being called caused sandbox_lib_path to be
22 unset, and thus breaking the execve() wrapper's LD_PRELOAD protection).
23 Add both the path in given SANDBOX_x variable, as well as its symlink
24 resolved path in init_env_entries(). Modify filter_path() to be able to
25 resolve paths without resolving symlinks, as well as to be able to resolve
26 symlinks. Fix a possible segfault in check_access(). Add symlink resolving
27 to check_access() resolving bug #31019. Add 'hack' for unlink, as the fix
28 for bug #31019 cause access violations if we try to remove a symlink that is
29 not in protected path, but points to a protected path. Fix a memory leak in
30 sandbox.c (sandbox_pids_file in main()). Fix the realpath() calls in main()
31 (sandbox.c) being unchecked. Fix the debug logname not having the pid in it
32 (pid_string was uninitialized). General syntax cleanups.
33
34 09 Mar 2005; Brian Harring <ferringb@gentoo.org> sandbox.c: Fixed the
35 infamous "pids file is not a regular file" w/out newline bug.
36
37 09 Mar 2005; Brian Harring <ferringb@gentoo.org> Makefile.am, configure.in:
38 Correct libc_version path detection, since it was screwing up if libdir !=
39 "/lib/".
40
41 02 Mar 2005; Martin Schlemmer <azarah@gentoo.org> libsandbox.c:
42 Hack to make sure sandboxed process cannot remove a device node, bug #79836.
43
44 02 Mar 2005; Martin Schlemmer <azarah@gentoo.org> Makefile.am:
45 Fix symbols.in not added to dist.
46
47 02 Mar 2005; Martin Schlemmer <azarah@gentoo.org> Makefile.am, canonicalize.c,
48 getcwd.c, libsandbox.c, sandbox.c, sandbox.h, sandbox_futils.c:
49 White space fixes.
50
51 02 Mar 2005; Martin Schlemmer <azarah@gentoo.org> Makefile.am, canonicalize.c,
52 configure.in, getcwd.c, libsandbox.c, symbols.in:
53 Fix inverse test logic in canonicalize.c, use a strncpy. Fix gcc warning in
54 getcwd.c. Add symbols.in and logic to Makefile.am to generate symbol versions
55 for glibc and other libc's that use this. Update libsandbox.c to use these
56 symbol versions if available. Fix exec wrapper to re-export LD_PRELOAD if the
57 process unset it.
58
59 01 Mar 2005; Brian Harring <ferringb@gentoo.org> libsandbox.c:
60 killed off _init and _fini in favor of
61 void __attribute__ ((constructor)) init_func and
62 void __attribute__ ((destructor)) closing_func. _(init|func) were deprecated.
63
64 06 Dec 2004; Brian Harring <ferringb@gentoo.org> Makefile.am, libsandbox.c,
65 canonicalize.c, getcwd.c: Fixed compilation *again*. Hopefully cvs is done
66 having the hick-ups.
67
68 04 Dec 2004; Brian Harring <ferringb@gentoo.org> libsandbox.c, getcwd.c,
69 Makefile.am: Fixed compilation.
70
71 01 Dec 2004; Brian Harring <ferringb@gentoo.org> aclocal.m4:
72 Gutted the bugger so it stops checking for a c++ and fortran compiler.
73 Do *not* regenerate aclocal.m4 for making a release until a better
74 solution is created.
75
76 20 Nov 2004; Brian Harring <ferringb@gentoo.org> Makefile.am, sandbox_futils.c:
77 Removal of more hardcoded paths.
78
79 20 Nov 2004; Brian Harring <ferringb@gentoo.org> Makefile.am, configure.in,
80 sandbox_futils.c: tweaks to install sandbox.bashrc, and use it.
81
82 19 Nov 2004; Brian Harring <ferringb@gentoo.org>:
83 Sandbox is now autotooled, create-localdecls needs to be killed and the code
84 shifted into configure.in. Currently builds *one* libsandbox.so- if multiple
85 are desired (-m64 and -m32 for amd64), the ebuild should do it (imo).
86 To get to a point of testing, automake && autoconf; created requisite files w/
87 a(utomake|clocal)-1.8, and autoconf 2.59. Installs to /usr/, instead of
88 /lib and /usr/lib/portage/bin.
89
90 14 Nov 2004; Brian Harring <ferringb@gentoo.org> libsandbox.c, sandbox.c:
91 closing out bug #70225, potential overflow of the sandbox_pids_file var.
92
93 07 Nov 2004; Brian Harring <ferringb@gentoo.org> libsandbox.c: c99 standard,
94 (think it was at least) allows intermixing of code and data segments. bug #70351
95 should be fixed by this.
96
97 03 Nov 2004; Brian Harring <ferringb@gentoo.org> libsandbox.c, sandbox_futils.c:
98 futils fix from bug #65201 via solar, and libsandbox log path checks via #69137
99
100 02 Aug 2004; Nicholas Jones <carpaski@gentoo.org> libsandbox.c: Code from
101 Seth Robertson that tracked down all adjuct flags for read operations that
102 do not invoke a write operation.
103
104 04 Apr 2004; Nicholas Jones <carpaski@gentoo.org> libsandbox.c, sandbox.c:
105 Another fix from jstubbs regarding a free() on a stack variable for the
106 environment -- tracking now prevents extraneous free()'s segfault.
107
108 04 Apr 2004; Nicholas Jones <carpaski@gentoo.org> libsandbox.c, sandbox.c:
109 J. Stubbs tracked down a new bug where mkdir was failing to the patch on
110 the lstat in mkdir... it now only returns 0 or -1 as documented for mkdir.
111 Also remove the errno = ESUCCESS settings as documentation points out that
112 a library isn't allowed to do that.
113
114 04 Apr 2004; Nicholas Jones <carpaski@gentoo.org> libsandbox.c: Added a
115 file_security_check() function to check random potential exploits on files
116 that sandbox is to load and read -- Normally sandboxpids.tmp. This fixes
117 the 'system-crippling' exploits (bug 21923) and catches a few other
118 potential problems.
119
120 20 Mar 2004; Nicholas Jones <carpaski@gentoo.org> Makefile: Updates for
121 32/64 bit sandbox. Made CC and LD '?=' values to allow passed in CC to work.
122
123 20 Mar 2004; Nicholas Jones <carpaski@gentoo.org> libsandbox.c:
124 bug 42048 -- Fixed the lstat/errno conditions for mkdir <caleb@g.o>.
125 Added the 64/32 bit sandbox patch for AMD64 bug 32963 <brad/azarah>.
126
127 29 Feb 2004; Martin Schlemmer <azarah@gentoo.org> sandbox.c, sandbox_futils.c :
128 Fix permissions and group of pids file and logs. Permissions should be 0664
129 and group should be 'portage'. Bug #34260.
130
131 28 Feb 2004; Martin Schlemmer <azarah@gentoo.org> libsandbox.c :
132 Besides a small cleanup, redo how we replace LD_PRELOAD in the environ passed
133 to the real execve (in our execve wrapper). Seems that on some arches (sparc
134 among others) do not allow us to tamper with the readonly copy passed to
135 execve, so pass our own copy of the environment. Bug #42290.
136
137 11 Jan 2004; Nicholas Jones <carpaski@gentoo.org> create-decls:
138 Changed tail to head and added a notice about duration of glibc check.
139
140 21 Dec 2003; Nicholas Jones <carpaski@gentoo.org> create-decls:
141 Changed the glibc subversion check to use /usr/bin/* instead of /bin/sh
142 as there isn't a guarentee that it is dynamic.
143
144 02 Nov 2003; Martin Schlemmer <azarah@gentoo.org> libsandbox.c :
145 If 'file' passed to before_syscall(const char *func, const char *file) is
146 invalid, we should set errno to ENOENT, and not EINVAL. This should
147 close bug #32238.
148
149 14 Oct 2003; Martin Schlemmer <azarah@gentoo.org> libsandbox.c :
150 Fix a bug that occurs mainly on 64bit arch, where the file passed to
151 the functions we wrap, is invalid, and then cause canonicalize to pass
152 garbage to before_syscall(), thanks to great detective work from
153 Andrea Luzzardi <al@sig11.org> (bug #29846).
154
155 13 Oct 2003; Martin Schlemmer <azarah@gentoo.org> create-localdecls :
156 Add a uClibc detection patch from Peter S. Mazinger <ps.m@gmx.net>.
157
158 13 Oct 2003; Martin Schlemmer <azarah@gentoo.org> libsandbox.c :
159 Fix a bug in libsandbox.c 's checking in the rename wrapper - it basically
160 only checked the destination patch, and not the source, so we could move
161 a protected file to a unprotected directory, and then delete/modify it.
162 Thanks to Andrea Luzzardi (scox) <al@sig11.org>, bug #30992, for this fix.
163
164 12 Oct 2003; Nicholas Jones <carpaski@gentoo.org> sandbox.c :
165 Added python2.3 to the predict section/variable.
166
167 28 Sep 2003; Martin Schlemmer <azarah@gentoo.org> libsandbox.c, sandbox.c,
168 sandbox.h, sandbox_futils.c :
169 Add support to set the pids file via SANDBOX_PIDS_FILE at startup. If
170 it is not set, it will revert to its old value.
171
172 27 Sep 2003; Martin Schlemmer <azarah@gentoo.org> libsandbox.c :
173 Fix our mkdir wrapper to check if the dir exist, and return EEXIST if so,
174 rather than failing with a violation, bug #29748.
175
176 27 Jul 2003; Martin Schlemmer <azarah@gentoo.org> libsandbox.c :
177 Fix canonicalize() to ignore calls with path = "".
178
179 27 Jul 2003; Martin Schlemmer <azarah@gentoo.org> getcwd.c, libsandbox.c,
180 sandbox_futils.c, canonicalize.c :
181 Once again coreutils fails, as my systems had 2.5 kernel, the getcwd system
182 call handled strings larger than PATH_MAX (bug #21766). It however does not
183 work the same on 2.4 kernels.
184
185 To fix, I added the posix implementation of getcwd() (from glibc cvs) that
186 do not need the system call. We use the default getcwd() function via a
187 wrapper (egetcwd), and then lstat the returned path. If lstat fails, it
188 means the current directory was removed, OR that the the system call for
189 getcwd failed (curious is that it do not fail and return NULL or set
190 errno, but rather just truncate the retured directory - usually from the
191 start), and if so, we use the generic getcwd() function (__egetcwd). Note
192 that we do not use the generic version all the time, as it calls lstat()
193 a great number of times, and performance degrade much.
194
195 29 Jun 2003; Martin Schlemmer <azarah@gentoo.org> create-localdecls,
196 libsandbox.c :
197 Make sure SB_PATH_MAX will not wrap. Fix two possible memory leaks.
198
199 22 Jun 2003; Martin Schlemmer <azarah@gentoo.org> libsandbox.c, canonicalize.c
200 create-localdecls :
201 When checking path names of files accessed, we need to canonicalize it, else
202 it may be a symlink in a 'write allowed' directory pointing to a file in a
203 directory we should not have write access to.
204
205 With something like coreutils-5.0, we have two problems:
206 1) One of the tests checks if getcwd() can return a path longer than
207 PATH_MAX. This test then tries to create a dir which even while
208 created local (mkdir("conftest2")), it ends up being resolved with
209 a name that is much larger than PATH_MAX. The problem now is that
210 canonicalize() have undefined behaviour when the path was too long
211 (returned wrongly truncated paths, etc), and pass the wrong path to
212 before_syscall() (causing the bogus sandbox violations).
213 2) The ecanonicalize() function we used, along with the canonicalize()
214 function did not support longer than PATH_MAX. This is partly a
215 cause for 1), but the error checking (rather lack of it) of calls
216 to erealpath() in canonicalize() was the prime reason for 1).
217
218 As we do not use this canonicalized name to call the function, we resolve this
219 by fixing canonicalize() to do better error checking, and ecanonicalize() as
220 well as all functions in libsandbox.c to use a PATH_MAX of 'PATH_MAX * 2'.
221 While they will resolve paths properly now, and can check if a write/read is
222 allowed, the functions called from the sandboxed environment will still work
223 as expected.
224
225 This should resolve bug #21766.
226
227 06 Apr 2003; Martin Schlemmer <azarah@gentoo.org> libsandbox.c :
228 For some reason sandbox fails with a 'open_wr' if you run 'locale -a' under
229 it (bug #16298).
230
231 Problem is that for some reason locale fopen's locale.alias with mode "rm".
232
233 -------------------------------------------------------
234 nosferatu root # grep fopen locale.log
235 fopen("/usr/share/locale/locale.alias", "rm"ACCESS DENIED open_wr: /usr/share/locale/locale.alias
236 nosferatu root #
237 --------------------------------------------------------
238
239 I checked the source of locale, but it have fopen with mode 'r', so
240 not sure where the "rm" mode comes from. Anyhow, changed the check in
241 before_syscall_open_char() to also see mode "rm" as readonly.
242
243 23 Feb 2003; Martin Schlemmer <azarah@gentoo.org> create-localdecls :
244
245 Add glibc-2.3 support.
246
247 22 Feb 2003; Martin Schlemmer <azarah@gentoo.org> sandbox.c :
248
249 Some /etc/ld.so.preload fixes. Just changed the #if defines to cover all
250 operations releated to preload, as well as only try to modify ld.so.preload
251 if we can. Also modify to write the pid to /tmp/sandboxpids.tmp even when
252 not using ld.so.preload. Fix to not write this instance of sandbox's pid
253 to /tmp/sandboxpids.tmp on exit if this is not the last sandbox running.
254
255 22 Feb 2003; Nicholas Jones <carpaski@gentoo.org> Makefile :
256
257 Changed the LD to CC for hppa.
258
259 22 Feb 2003; Nicholas Jones <carpaski@gentoo.org> create-localdecls :
260
261 Killed the previous changes I made.
262
263 17 Feb 2003; Nicholas Jones <carpaski@gentoo.org> create-localdecls :
264
265 Added parisc to BROKEN_RTLD_ARCHLIST to see if it we can fix the relocation probs.
266
267 09 Jan 2003; J Robert Ray <jrray@gentoo.org> sandbox.c :
268
269 Don't segfault if $HOME isn't set, set $HOME to "/" instead. Fixes bug 10868.
270
271 16 Dec 2002; Martin Schlemmer <azarah@gentoo.org> create-localdecls :
272
273 Fix memory leak for mips, bug #12236. Thanks to Torgeir Hansen <torgeir@trenger.ro>
274 for this fix.
275
276 4 Dec 2002; J Robert Ray <jrray@gentoo.org> sandbox.h sandbox_futils.c :
277
278 sandbox_futils defined a dirname() function that was masking the same
279 function in glibc and was broken (e.g.: SANDBOX_DIR was being set to
280 '/usr/lib/portage/bi/'). Fixed function to return expected results and
281 renamed it to sb_dirname() to no longer mask the glibc function. Closes bug
282 11231.
283
284 4 Dec 2002; Martin Schlemmer <azarah@gentoo.org> :
285
286 Fix a segfault in libsandbox.c if canonicalize() was called with
287 first parameter = NULL.
288
289 1 Sep 2002; Martin Schlemmer <azarah@gentoo.org> :
290
291 Fix my braindead 'return 1;' in a void function. Updated sandbox.c,
292 cleanup() for this.
293
294 Change cleanup() in sandbox.c not to exit with fail status if
295 the pidsfile is missing. We really should still display sandbox
296 violations if they occured.
297
298 31 Aug 2002; Martin Schlemmer <azarah@gentoo.org> :
299
300 Update cleanup() in sandbox.c to remove the PIDSFILE if this is
301 the last sandbox running.
302
303 25 Aug 2002; Martin Schlemmer <azarah@gentoo.org> :
304
305 Major cleanups to mainly libsandbox.c again.
306
307 22 Aug 2002; Martin Schlemmer <azarah@gentoo.org> :
308
309 Add copyrights to sandbox.h and sandbox_futils.h. If wrong, the
310 parties involved should please contact me so that we can fix it.
311
312 Add opendir wrapper to libsandbox.c.
313
314 21 Aug 2002; Martin Schlemmer <azarah@gentoo.org> :
315
316 Do some more cleanups to ecanonicalize(), as it dropped filenames in
317 rare cases (after my symlink cleanups), and caused glibc to bork.
318 These fixes went into canonicalize.c.
319
320 20 Aug 2002; Martin Schlemmer <azarah@gentoo.org> :
321
322 Fix spawn_shell() and main() in sandbox.c to properly return fail
323 status.
324
325 19 Aug 2002; Martin Schlemmer <azarah@gentoo.org> :
326
327 The new canonicalize() function in libsandbox.c also resolved symlinks,
328 which caused on cleaning sandbox errors if the symlink pointed to a
329 file in the live root. Ripped out canonicalize() and realpath() from
330 glibc; removed the symlink stuff, and changed them to ecanonicalize()
331 and erealpath().
332
333 18 Aug 2002; Martin Schlemmer <azarah@gentoo.org> :
334
335 Ripped out all the wrappers, and implemented those of InstallWatch.
336 Losts of cleanups and bugfixes. Implement a execve that forces
337 $LIBSANDBOX in $LD_PRELOAD. We can now thus do away with the feared
338 /etc/ld.so.preload (*g*) ... Made the needed changes to sandbox.c,
339 sandbox.h and sandbox_futils.c. Rewrote the Makefile for most
340 parts; it now have an install target.
341
342 Reformat the whole thing to look somewhat like the reworked sandbox.c
343 and new sandbox.h and sandbox_futils.c from:
344
345 Brad House <brad@mainstreetsoftworks.com>.
346
347 Additional Copyrights now due to the InstallWatch code:
348
349 Copyright (C) 1998-9 Pancrazio `Ezio' de Mauro <p@demauro.net>

Properties

Name Value
svn:eol-style native
svn:keywords Author Date Id Revision

  ViewVC Help
Powered by ViewVC 1.1.20