- tor-commits - lists.torproject.org

[tor/master] fold in last changes item
by arma＠torproject.org 02 Sep '11

02 Sep '11

commit 589ed3f771f1d77d9d5e366ba60849c2babb49e0 Author: Roger Dingledine <arma(a)torproject.org> Date: Thu Sep 1 20:49:21 2011 -0400 fold in last changes item --- ChangeLog | 2 ++ changes/bug3888 | 4 ---- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/ChangeLog b/ChangeLog index a28b2b2..7084a0f 100644 --- a/ChangeLog +++ b/ChangeLog @@ -57,6 +57,8 @@ Changes in version 0.2.3.3-alpha - 2011-09-01 without flushing it completely. Fixes bug 3805. - Fix a bug where server-side tunneled bufferevent-based directory streams would get closed prematurely. Fixes bug 3814. + - Fix a use-after-free error with per-connection rate-limiting + buckets. Fixes bug 3888. o Major bugfixes (also part of 0.2.2.31-rc): - If we're configured to write our ControlPorts to disk, only write diff --git a/changes/bug3888 b/changes/bug3888 deleted file mode 100644 index d3a5e27..0000000 --- a/changes/bug3888 +++ /dev/null @@ -1,4 +0,0 @@ - o Minor bugfixes (bufferevents, fix on 0.2.3.1-alpha): - - Fix a use-after-free error with per-connection rate-limiting - buckets. Fixes bug 3888. -

1 0

[torbutton/master] Bug #3879: Framed sites broken (yopmail, gmane, gmaps, etc)
by mikeperry＠torproject.org 02 Sep '11

02 Sep '11

commit 709c3fb7ef212e9b1b75db3d0a8512e4b03474ff Author: Mike Perry <mikeperry-git(a)fscked.org> Date: Thu Sep 1 17:28:18 2011 -0700 Bug #3879: Framed sites broken (yopmail, gmane, gmaps, etc) We were accidentally clearing the name attribute of framesets because we got notification before the referrer could be set on the channel. --- src/chrome/content/torbutton.js | 9 ++++++--- 1 files changed, 6 insertions(+), 3 deletions(-) diff --git a/src/chrome/content/torbutton.js b/src/chrome/content/torbutton.js index ee4b03e..6496895 100644 --- a/src/chrome/content/torbutton.js +++ b/src/chrome/content/torbutton.js @@ -4232,9 +4232,12 @@ function torbutton_hookdoc(win, doc, state_change, referrer) { // This keeps window.name clean between fresh urls. // It should also apply to iframes because hookdoc gets called for all // frames and subdocuments. - if (!referrer || referrer.spec == "") { - win.name = null; - win.window.name = null; + // + // The about:blank check handles the 'name' attribute of framesets, which + // get set before the referer is set on the channel. + if ((!referrer || referrer.spec == "") && win.location != "about:blank") { + win.name = ""; + win.window.name = ""; } } catch(e) { torbutton_log(4, "Failed to reset window.name: "+e)

1 0

[tor/master] Handle test case where fgets() sees EOF on the last read
by nickm＠torproject.org 01 Sep '11

01 Sep '11

commit f5df96c94f0c40e4bf55e4b4b843df85e9d90a71 Author: Steven Murdoch <Steven.Murdoch(a)cl.cam.ac.uk> Date: Thu Sep 1 18:17:54 2011 +0100 Handle test case where fgets() sees EOF on the last read On some platforms, with non-blocking IO, on EOF you first get EAGAIN, and then on the second read you get zero bytes and EOF is set. However on others, the EOF flag is set as soon as the last byte is read. This patch fixes the test case in the latter scenario. --- src/test/test_util.c | 10 +++++++--- 1 files changed, 7 insertions(+), 3 deletions(-) diff --git a/src/test/test_util.c b/src/test/test_util.c index 7faa0ae..f9672c1 100644 --- a/src/test/test_util.c +++ b/src/test/test_util.c @@ -1541,14 +1541,18 @@ test_util_spawn_background_partial_read(void *ptr) pos = tor_read_all_handle(process_handle.stdout_pipe, stdout_buf, sizeof(stdout_buf) - 1, &process_handle); + tt_int_op(pos, ==, 0); #else - if (!eof) + if (!eof) { + /* We should have got all the data, but maybe not the EOF flag */ pos = tor_read_all_handle(process_handle.stdout_handle, stdout_buf, sizeof(stdout_buf) - 1, &process_handle, &eof); - tt_assert(eof) + tt_int_op(pos, ==, 0); + tt_assert(eof); + } + /* Otherwise, we got the EOF on the last read */ #endif - tt_int_op(pos, ==, 0); /* Check it terminated correctly */ retval = tor_get_exit_code(process_handle, 1, &exit_code);

1 0

[tor/master] Changes file for bug3888
by nickm＠torproject.org 01 Sep '11

01 Sep '11

commit ab9874dba46b3eae8e201a45926ddeb133903ebc Author: Nick Mathewson <nickm(a)torproject.org> Date: Thu Sep 1 10:45:26 2011 -0400 Changes file for bug3888 --- changes/bug3888 | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/changes/bug3888 b/changes/bug3888 new file mode 100644 index 0000000..d3a5e27 --- /dev/null +++ b/changes/bug3888 @@ -0,0 +1,4 @@ + o Minor bugfixes (bufferevents, fix on 0.2.3.1-alpha): + - Fix a use-after-free error with per-connection rate-limiting + buckets. Fixes bug 3888. +

1 0

[tor/master] Explicitly set bucket_cfg to NULL after we freed it
by nickm＠torproject.org 01 Sep '11

01 Sep '11

commit ebb5f8df368c29dd26080ba057ffb6611aee401b Author: Sebastian Hahn <sebastian(a)torproject.org> Date: Thu Sep 1 12:24:05 2011 +0200 Explicitly set bucket_cfg to NULL after we freed it This should fix bug 3888. --- src/or/connection.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/or/connection.c b/src/or/connection.c index 7615952..1b227d7 100644 --- a/src/or/connection.c +++ b/src/or/connection.c @@ -524,6 +524,7 @@ _connection_free(connection_t *conn) #ifdef USE_BUFFEREVENTS if (conn->type == CONN_TYPE_OR && TO_OR_CONN(conn)->bucket_cfg) { ev_token_bucket_cfg_free(TO_OR_CONN(conn)->bucket_cfg); + TO_OR_CONN(conn)->bucket_cfg = NULL; } #endif

1 0

[tor/master] Merge branch 'bug3888'
by nickm＠torproject.org 01 Sep '11

01 Sep '11

commit 9f144144e383169ba6a0de78fde50c0d1f662116 Merge: cfa9ee5 ab9874d Author: Nick Mathewson <nickm(a)torproject.org> Date: Thu Sep 1 10:45:34 2011 -0400 Merge branch 'bug3888' changes/bug3888 | 4 ++++ src/or/connection.c | 1 + 2 files changed, 5 insertions(+), 0 deletions(-)

1 0

[tor/master] Fix double-closing a stdio stream
by nickm＠torproject.org 01 Sep '11

01 Sep '11

commit cfa9ee5fe708539965f46a31b5d2abe4950179af Author: Steven Murdoch <Steven.Murdoch(a)cl.cam.ac.uk> Date: Thu Sep 1 13:09:38 2011 +0100 Fix double-closing a stdio stream After a stream reached eof, we fclose it, but then test_util_spawn_background_partial_read() reads from it again, which causes an error and thus another fclose(). Some platforms are fine with this, others (e.g. debian-sid-i386) trigger a double-free() error. The actual code used by Tor (log_from_pipe() and tor_check_port_forwarding()) handle this case correctly. --- src/common/util.c | 18 +++++++++++++----- src/common/util.h | 3 ++- src/test/test_util.c | 18 ++++++++++++------ 3 files changed, 27 insertions(+), 12 deletions(-) diff --git a/src/common/util.c b/src/common/util.c index 7cd9dd8..db6b00f 100644 --- a/src/common/util.c +++ b/src/common/util.c @@ -3583,14 +3583,19 @@ tor_read_all_handle(HANDLE h, char *buf, size_t count, * process is NULL, the function will return immediately if there is * nothing more to read. Otherwise data will be read until end of file, or * count bytes are read. Returns the number of bytes read, or -1 on - * error. */ + * error. Sets eof to true if eof is not NULL and the end of the + * file has been reached. */ ssize_t tor_read_all_handle(FILE *h, char *buf, size_t count, - const process_handle_t *process) + const process_handle_t *process, + int *eof) { size_t numread = 0; char *retval; + if (eof) + *eof = 0; + if (count > SIZE_T_CEILING || count > SSIZE_T_MAX) return -1; @@ -3599,7 +3604,10 @@ tor_read_all_handle(FILE *h, char *buf, size_t count, retval = fgets(buf+numread, (int)(count-numread), h); if (NULL == retval) { if (feof(h)) { + log_debug(LD_GENERAL, "fgets() reached end of file"); fclose(h); + if (eof) + *eof = 1; break; } else { if (EAGAIN == errno) { @@ -3620,7 +3628,7 @@ tor_read_all_handle(FILE *h, char *buf, size_t count, numread += strlen(retval); } - log_debug(LD_GENERAL, "fgets read %d bytes from handle", (int)numread); + log_debug(LD_GENERAL, "fgets() read %d bytes from handle", (int)numread); return (ssize_t)numread; } #endif @@ -3635,7 +3643,7 @@ tor_read_all_from_process_stdout(const process_handle_t *process_handle, process_handle); #else return tor_read_all_handle(process_handle->stdout_handle, buf, count, - process_handle); + process_handle, NULL); #endif } @@ -3649,7 +3657,7 @@ tor_read_all_from_process_stderr(const process_handle_t *process_handle, process_handle); #else return tor_read_all_handle(process_handle->stderr_handle, buf, count, - process_handle); + process_handle, NULL); #endif } diff --git a/src/common/util.h b/src/common/util.h index 6211267..c8cce39 100644 --- a/src/common/util.h +++ b/src/common/util.h @@ -391,7 +391,8 @@ ssize_t tor_read_all_handle(HANDLE h, char *buf, size_t count, const process_handle_t *process); #else ssize_t tor_read_all_handle(FILE *h, char *buf, size_t count, - const process_handle_t *process); + const process_handle_t *process, + int *eof); #endif ssize_t tor_read_all_from_process_stdout( const process_handle_t *process_handle, char *buf, size_t count); diff --git a/src/test/test_util.c b/src/test/test_util.c index 06c1be3..7faa0ae 100644 --- a/src/test/test_util.c +++ b/src/test/test_util.c @@ -1484,7 +1484,7 @@ test_util_spawn_background_partial_read(void *ptr) const int expected_status = PROCESS_STATUS_RUNNING; int retval, exit_code; - ssize_t pos; + ssize_t pos = -1; process_handle_t process_handle; char stdout_buf[100], stderr_buf[100]; #ifdef MS_WINDOWS @@ -1499,6 +1499,7 @@ test_util_spawn_background_partial_read(void *ptr) "DONE\n", NULL }; const char *expected_err = "ERR\n"; + int eof = 0; #endif int expected_out_ctr; (void)ptr; @@ -1517,8 +1518,10 @@ test_util_spawn_background_partial_read(void *ptr) pos = tor_read_all_handle(process_handle.stdout_pipe, stdout_buf, sizeof(stdout_buf) - 1, NULL); #else + /* Check that we didn't read the end of file last time */ + tt_assert(!eof); pos = tor_read_all_handle(process_handle.stdout_handle, stdout_buf, - sizeof(stdout_buf) - 1, NULL); + sizeof(stdout_buf) - 1, NULL, &eof); #endif log_info(LD_GENERAL, "tor_read_all_handle() returned %d", (int)pos); @@ -1526,7 +1529,7 @@ test_util_spawn_background_partial_read(void *ptr) if (0 == pos) continue; - tt_int_op(pos, >=, 0); + tt_int_op(pos, >, 0); stdout_buf[pos] = '\0'; tt_str_op(stdout_buf, ==, expected_out[expected_out_ctr]); tt_int_op(pos, ==, strlen(expected_out[expected_out_ctr])); @@ -1539,10 +1542,13 @@ test_util_spawn_background_partial_read(void *ptr) sizeof(stdout_buf) - 1, &process_handle); #else - pos = tor_read_all_handle(process_handle.stdout_handle, stdout_buf, - sizeof(stdout_buf) - 1, - &process_handle); + if (!eof) + pos = tor_read_all_handle(process_handle.stdout_handle, stdout_buf, + sizeof(stdout_buf) - 1, + &process_handle, &eof); + tt_assert(eof) #endif + tt_int_op(pos, ==, 0); /* Check it terminated correctly */ retval = tor_get_exit_code(process_handle, 1, &exit_code);

1 0

[tor/master] clean 0.2.3.3-alpha changelog
by arma＠torproject.org 01 Sep '11

01 Sep '11

commit a500389904c992e492db0c219e5019ed44704c29 Author: Roger Dingledine <arma(a)torproject.org> Date: Thu Sep 1 02:43:11 2011 -0400 clean 0.2.3.3-alpha changelog ready when you are, nick --- ChangeLog | 94 ++++++++++++++++++++++++++++++------------------------------ 1 files changed, 47 insertions(+), 47 deletions(-) diff --git a/ChangeLog b/ChangeLog index bbefe24..a28b2b2 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,59 +1,64 @@ -Changes in version 0.2.3.3-alpha - 2011-0?-?? - Tor 0.2.3.3-alpha adds a new major "stream isolation" feature to - improve Tor's security, and provides client-side support for several - the microdescriptor and optimistic data features introduced earlier in - the 0.2.3.x series. Also, it has numerous critical bugfixes in the - (optional) bufferevent-based networking backend. - - o Major features: +Changes in version 0.2.3.3-alpha - 2011-09-01 + Tor 0.2.3.3-alpha adds a new "stream isolation" feature to improve Tor's + security, and provides client-side support for the microdescriptor + and optimistic data features introduced earlier in the 0.2.3.x + series. It also includes numerous critical bugfixes in the (optional) + bufferevent-based networking backend. + + o Major features (stream isolation): - You can now configure Tor so that streams from different applications are isolated on different circuits, to prevent an - attacker who sees your streams leaving an exit node from linking - your sessions to one another. To do this, choose some way to - distinguish the applications: have them connect to different + attacker who sees your streams as they leave an exit node from + linking your sessions to one another. To do this, choose some way + to distinguish the applications: have them connect to different SocksPorts, or have one of them use SOCKS4 while the other uses SOCKS5, or have them pass different authentication strings to the SOCKS proxy. Then, use the new SocksPort syntax to configure the degree of isolation you need. This implements Proposal 171. - - The microdescriptor system is now on by default for clients. This - allows clients to download a much smaller amount of directory - information. To disable it, set "UseMicrodescriptors 0" in your - torrc file. - - Tor's firewall-helper feature, introduced in 0.2.3.1-alpha, now - supports Windows. - - When using an exit nodes running 0.2.3.x, clients can now - "optimistically" send data before the exit node reports that the - stream has opened. This saves a round trip when starting - connections where the client speaks first. This behavior is - controlled by a (currently disabled) consensus parameter. To turn - it on or off manually, use the "OptimisticData" torrc - option. Implements proposal 181; code by Ian Goldberg. - - o Major bugfixes (bufferevents): - - When using IOCP on windows, we need to enable Libevent windows - threading support. Bugfix on 0.2.3.1-alpha. + - There's a new syntax for specifying multiple client ports (such as + SOCKSPort, TransPort, DNSPort, NATDPort): you can now just declare + multiple *Port entries with full addr:port syntax on each. + The old *ListenAddress format is still supported, but you can't + mix it with the new *Port syntax. + + o Major features (other): + - Enable microdescriptor fetching by default for clients. This allows + clients to download a much smaller amount of directory information. + To disable it (and go back to the old-style consensus and + descriptors), set "UseMicrodescriptors 0" in your torrc file. + - Tor's firewall-helper feature, introduced in 0.2.3.1-alpha (see the + "PortForwarding" config option), now supports Windows. + - When using an exit relay running 0.2.3.x, clients can now + "optimistically" send data before the exit relay reports that + the stream has opened. This saves a round trip when starting + connections where the client speaks first (such as web browsing). + This behavior is controlled by a consensus parameter (currently + disabled). To turn it on or off manually, use the "OptimisticData" + torrc option. Implements proposal 181; code by Ian Goldberg. + + o Major bugfixes (bufferevents, fixes on 0.2.3.1-alpha): + - When using IOCP on Windows, we need to enable Libevent windows + threading support. - The IOCP backend now works even when the user has not specified - the (internal, debbuging-only) _UseFilteringSSLBufferevents option. - Fixes part of bug 3752; bugfix on 0.2.3.1-alpha. + the (internal, debugging-only) _UseFilteringSSLBufferevents option. + Fixes part of bug 3752. - Correctly record the bytes we've read and written when using bufferevents, so that we can include them in our bandwidth history - and advertised bandwidth. Fixes bug 3803; bugfix on 0.2.3.1-alpha. + and advertised bandwidth. Fixes bug 3803. - Apply rate-limiting only at the bottom of a chain of filtering bufferevents. This prevents us from filling up internal read buffers and violating rate-limits when filtering bufferevents - are enabled. Bugfix on 0.2.3.1-alpha; fixes part of bug 3804. + are enabled. Fixes part of bug 3804. - Add high-watermarks to the output buffers for filtered bufferevents. This prevents us from filling up internal write buffers and wasting CPU cycles when filtering bufferevents are - enabled. Bugfix on 0.2.3.1-alpha; fixes part of bug 3804. + enabled. Fixes part of bug 3804. - Correctly notice when data has been written from a bufferevent - without flushing it completely. Bugfix on 0.2.3.1-alpha; fixes - bug 3805. + without flushing it completely. Fixes bug 3805. - Fix a bug where server-side tunneled bufferevent-based directory - streams would get closed prematurely. Fixes 3814, bugfix on - 0.2.3.1-alpha. + streams would get closed prematurely. Fixes bug 3814. - o Major bugfixes (on 0.2.2.x and earlier): [stet] + o Major bugfixes (also part of 0.2.2.31-rc): - If we're configured to write our ControlPorts to disk, only write them after switching UID and creating the data directory. This way, we don't fail when starting up with a nonexistent DataDirectory @@ -61,27 +66,22 @@ Changes in version 0.2.3.3-alpha - 2011-0?-?? bug 3747; bugfix on Tor 0.2.2.26-beta. o Minor features: - - There's a new syntax for specifying multiple client ports (such as - SOCKSPort, TransPort, DNSPort, NATDPort): you can now just declare - multiple *Port entries with full addr:port syntax on each. - The old *ListenAddress format is still supported, but you can't - mix it with the new *Port syntax. - Added a new CONF_CHANGED event so that controllers can be notified of any configuration changes made by other controllers, or by the - user. Implements #1692. + user. Implements ticket 1692. - Use evbuffer_copyout() in inspect_evbuffer(). This fixes a memory leak when using bufferevents, and lets Libevent worry about how to best copy data out of a buffer. - Replace files in stats/ rather than appending to them. Now that we include statistics in extra-info descriptors, it makes no sense to - keep old statistics forever. Implements #2930. + keep old statistics forever. Implements ticket 2930. o Minor features (build compatibility): - Limited, experimental support for building with nmake and MSVC. - Provide a substitute implementation of lround() for MSVC, which apparently lacks it. Patch from Gisle Vanem. - o Minor features: [stet] + o Minor features (also part of 0.2.2.31-rc): - Update to the August 2 2011 Maxmind GeoLite Country database. o Minor bugfixes (on 0.2.3.x-alpha): @@ -98,7 +98,7 @@ Changes in version 0.2.3.3-alpha - 2011-0?-?? the absense of configured logs. Fixes bug 3550; bugfix on 0.2.0.10-alpha. - o Minor bugfixes (on 2.2.x and earlier): [stet] + o Minor bugfixes (also part of 0.2.2.31-rc): - Write several files in text mode, on OSes that distinguish text mode from binary mode (namely, Windows). These files are: 'buffer-stats', 'dirreq-stats', and 'entry-stats' on relays @@ -120,8 +120,8 @@ Changes in version 0.2.3.3-alpha - 2011-0?-?? on 0.2.1.5-alpha. o Code simplifications and refactoring: - - Rewrote the listener-selection logic so that parsing which ports - we want to listen on is now separate form binding to the ports + - Rewrite the listener-selection logic so that parsing which ports + we want to listen on is now separate from binding to the ports we want. o Build changes:

1 0

[torbrowser/maint-2.2] remove -alpha from windows filename and add the right branch for srcdist
by erinn＠torproject.org 01 Sep '11

01 Sep '11

commit b3452f96ff6a45d329fd55d11406e5d5b7cce29a Author: Erinn Clark <erinn(a)torproject.org> Date: Thu Sep 1 04:32:46 2011 +0100 remove -alpha from windows filename and add the right branch for srcdist --- build-scripts/windows.mk | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/build-scripts/windows.mk b/build-scripts/windows.mk index d4d4665..72909dd 100644 --- a/build-scripts/windows.mk +++ b/build-scripts/windows.mk @@ -145,7 +145,7 @@ NAME="Tor Browser" DISTDIR=tbbwin-alpha-dist ## Version and name of the compressed bundle (also used for source) -VERSION=$(RELEASE_VER)-$(BUILD_NUM)-alpha +VERSION=$(RELEASE_VER)-$(BUILD_NUM) DEFAULT_COMPRESSED_BASENAME=tor-browser-$(VERSION) IM_COMPRESSED_BASENAME=tor-im-browser-$(VERSION) DEFAULT_COMPRESSED_NAME=$(DEFAULT_COMPRESSED_BASENAME) @@ -471,7 +471,7 @@ SRCDEST=/tmp SRCDESTPATH=$(SRCDEST)/$(SRCNAME) srcdist: rm -fr $(SRCDESTPATH) - git clone git://git.torproject.org/torbrowser.git \ + git clone -b maint-2.2 git://git.torproject.org/torbrowser.git \ $(SRCDESTPATH) cd $(SRCDEST); tar --exclude src/archived-patches \ --exclude src/current-patches --exclude src/processtest \

1 0

[torbrowser/maint-2.2] remove unnecessary sed from linux makefile
by erinn＠torproject.org 01 Sep '11

01 Sep '11

commit fb0ff547d07a40d7121e4360d26efb9640cff301 Author: Erinn Clark <erinn(a)torproject.org> Date: Thu Sep 1 04:28:54 2011 +0100 remove unnecessary sed from linux makefile --- build-scripts/linux.mk | 2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/build-scripts/linux.mk b/build-scripts/linux.mk index e7eaf94..1df7377 100644 --- a/build-scripts/linux.mk +++ b/build-scripts/linux.mk @@ -465,8 +465,6 @@ patch-firefox-language: ## Fix prefs.js since extensions.checkCompatibility, false doesn't work update-extension-pref: - #ext_ver=$$(sed -n '/em:version/{s,.*="$.*$".*,\1,p;q}' $(BUNDLE)/Data/profile/extensions/langpack-$(LANGCODE)(a)firefox.mozilla.org/install.rdf) \ - sed -i -e "s/SHPONKA/langpack-$(LANGCODE)@firefox.mozilla.org:$$ext_ver/g" $(BUNDLE)/Data/profile/prefs.js sed -i -e "s/SHPONKA/$(LANGCODE)/g" $(BUNDLE)/Data/profile/prefs.js sed -i -e "s/SHPONKA/$(LANGCODE)/g" $(BUNDLE)/App/Firefox/defaults/profile/prefs.js

1 0