/[path-sandbox]/trunk/ChangeLog
Gentoo

Contents of /trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 33 - (show annotations) (download)
Wed Mar 9 07:34:30 2005 UTC (9 years, 8 months ago) by ferringb
File size: 14758 byte(s)
correct symbols.h building when libdir !+ "/lib"

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