/[path-sandbox]/trunk/ChangeLog
Gentoo

Contents of /trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 28 - (hide annotations) (download)
Wed Mar 2 05:48:12 2005 UTC (9 years, 1 month ago) by ferringb
File size: 13700 byte(s)
killed off _init and _fini in favor of
void __attribute__ ((constructor)) init_func and
void __attribute__ ((destructor)) closing_func.  _(init|func) were deprecated.

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