- tor-commits - lists.torproject.org

[torflow/master] Actually, we should be using min(circ_error, pid_error).
by mikeperry＠torproject.org 02 Dec '11

02 Dec '11

commit 3be33faf271d378fddc4a048538703cf16ffd586 Author: Mike Perry <mikeperry-git(a)fscked.org> Date: Fri Dec 2 01:20:17 2011 -0800 Actually, we should be using min(circ_error,pid_error). --- NetworkScanners/BwAuthority/README.spec.txt | 3 ++- NetworkScanners/BwAuthority/aggregate.py | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/NetworkScanners/BwAuthority/README.spec.txt b/NetworkScanners/BwAuthority/README.spec.txt index d223160..ddaa7cc 100644 --- a/NetworkScanners/BwAuthority/README.spec.txt +++ b/NetworkScanners/BwAuthority/README.spec.txt @@ -517,7 +517,8 @@ respectively. If this error value is negative (meaning the node is failing - more circuits than the network average), we use it as the + more circuits than the network average), we use the smaller + of the circ_error and the original pid_error as the new pid_error. "bwauthbestratio=0" diff --git a/NetworkScanners/BwAuthority/aggregate.py b/NetworkScanners/BwAuthority/aggregate.py index ae6f5be..884b065 100755 --- a/NetworkScanners/BwAuthority/aggregate.py +++ b/NetworkScanners/BwAuthority/aggregate.py @@ -549,7 +549,7 @@ def main(argv): # FIXME: Hrmm, should we only penalize for circ fails, or should # we reward, too? if circ_error < 0: - n.pid_error = circ_error + n.pid_error = min(circ_error,n.pid_error) if n.idhex in prev_votes.vote_map: # If there is a new sample, let's use it for all but guards

1 0

r25241: {website} bring the bridge explanation partway out of the stone age (website/trunk/docs/en)
by Roger Dingledine 02 Dec '11

02 Dec '11

Author: arma Date: 2011-12-02 07:52:49 +0000 (Fri, 02 Dec 2011) New Revision: 25241 Modified: website/trunk/docs/en/bridges.wml Log: bring the bridge explanation partway out of the stone age Modified: website/trunk/docs/en/bridges.wml =================================================================== --- website/trunk/docs/en/bridges.wml 2011-12-02 07:50:12 UTC (rev 25240) +++ website/trunk/docs/en/bridges.wml 2011-12-02 07:52:49 UTC (rev 25241) @@ -25,11 +25,7 @@ <p> The addition of bridges to Tor is a step forward in the blocking resistance race. It is perfectly possible that even if your ISP filters - the Internet, you do not require a bridge to use Tor. Many filtering - programs look for unencrypted Tor directory requests to recognize that - you're using Tor, but Tor version 0.2.0.23-rc and later use encrypted - directory queries by default. This change means that most filtering - programs are now unable to recognize Tor connections. So you should try + the Internet, you do not require a bridge to use Tor. So you should try to use Tor without bridges first, since it might work. </p> @@ -42,8 +38,8 @@ FAQ about problems with running Tor properly</a> when you have issues. If you feel that the issue is clearly blocking, or you'd simply like to try because you're unsure or feeling adventurous, please read on. Ensure - that you're using the <a href="<page download/download>#Dev">latest 0.2.1.x or - 0.2.2.x bundle for your platform</a>. + that you're using the <a href="<page download/download>#Dev">latest 0.2.2.x or + 0.2.3.x bundle for your platform</a>. </p> <p> @@ -207,3 +203,4 @@ </div>  #include <foot.wmi> +

1 0

r25240: {website} Fixed link to Sponsor TODO list (website/trunk/docs/en)
by Runa Sandvik 02 Dec '11

02 Dec '11

Author: runa Date: 2011-12-02 07:50:12 +0000 (Fri, 02 Dec 2011) New Revision: 25240 Modified: website/trunk/docs/en/documentation.wml Log: Fixed link to Sponsor TODO list Modified: website/trunk/docs/en/documentation.wml =================================================================== --- website/trunk/docs/en/documentation.wml 2011-12-01 04:51:48 UTC (rev 25239) +++ website/trunk/docs/en/documentation.wml 2011-12-02 07:50:12 UTC (rev 25240) @@ -145,7 +145,7 @@ <li> Our <a - href="https://trac.torproject.org/projects/tor/wiki/sponsors">sponsor TODO list</a> starts with a + href="https://trac.torproject.org/projects/tor/wiki/org/sponsors">sponsor TODO list</a> starts with a timeline for external promises — things <a href="<page about/sponsors>">our sponsors</a> have paid to see done. It also lists many other tasks and topics we'd like to tackle next.

1 0

[torflow/master] Directly use the circ_error if it is negative.
by mikeperry＠torproject.org 02 Dec '11

02 Dec '11

commit 9e27a77f17bae2f9535373f8b48c2cee6ea2ae4d Author: Mike Perry <mikeperry-git(a)fscked.org> Date: Thu Dec 1 21:23:46 2011 -0800 Directly use the circ_error if it is negative. --- NetworkScanners/BwAuthority/README.spec.txt | 4 +++- NetworkScanners/BwAuthority/aggregate.py | 8 ++++---- 2 files changed, 7 insertions(+), 5 deletions(-) diff --git a/NetworkScanners/BwAuthority/README.spec.txt b/NetworkScanners/BwAuthority/README.spec.txt index db5c125..d223160 100644 --- a/NetworkScanners/BwAuthority/README.spec.txt +++ b/NetworkScanners/BwAuthority/README.spec.txt @@ -516,7 +516,9 @@ to the node, and the average success rate for the entire node class, respectively. - This error value is then added to pid_error. + If this error value is negative (meaning the node is failing + more circuits than the network average), we use it as the + pid_error. "bwauthbestratio=0" If absent, the larger of stream bandwidth vs filtered bandwidth diff --git a/NetworkScanners/BwAuthority/aggregate.py b/NetworkScanners/BwAuthority/aggregate.py index a2372af..ae6f5be 100755 --- a/NetworkScanners/BwAuthority/aggregate.py +++ b/NetworkScanners/BwAuthority/aggregate.py @@ -546,10 +546,10 @@ def main(argv): if cs_junk.use_circ_fails: circ_error = ((1.0-n.circ_fail_rate) - true_circ_avg[n.node_class()]) \ / true_circ_avg[n.node_class()] - # FIXME: Hrmm, should we only penalize for circ successes, or should - # we reward, too? Let's try both for now. - # if circ_error < 0: - n.pid_error += circ_error + # FIXME: Hrmm, should we only penalize for circ fails, or should + # we reward, too? + if circ_error < 0: + n.pid_error = circ_error if n.idhex in prev_votes.vote_map: # If there is a new sample, let's use it for all but guards

1 0

[tor/master] Appease check-spaces
by nickm＠torproject.org 02 Dec '11

02 Dec '11

commit c811b8f3a0a3ec2a071983d6952350f521c8b28d Author: Sebastian Hahn <sebastian(a)torproject.org> Date: Fri Dec 2 06:16:47 2011 +0100 Appease check-spaces --- src/test/test_config.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/test/test_config.c b/src/test/test_config.c index 4ce0fa8..d8161de 100644 --- a/src/test/test_config.c +++ b/src/test/test_config.c @@ -156,7 +156,7 @@ test_config_addressmap(void *arg) strlcpy(address, "www.torproject.org", sizeof(address)); test_assert(!addressmap_rewrite(address, sizeof(address), &expires)); -done: + done: ; }

1 0

[tor/master] Work around a false positive in Coverity.
by nickm＠torproject.org 02 Dec '11

02 Dec '11

commit 95af91565bd846be0b199fbc02d7197bf771e867 Author: Sebastian Hahn <sebastian(a)torproject.org> Date: Fri Dec 2 06:15:31 2011 +0100 Work around a false positive in Coverity. Fixes cid 501 and 502. --- src/common/crypto.c | 7 +++++-- 1 files changed, 5 insertions(+), 2 deletions(-) diff --git a/src/common/crypto.c b/src/common/crypto.c index 62b0bce..2aa3fc9 100644 --- a/src/common/crypto.c +++ b/src/common/crypto.c @@ -1912,9 +1912,12 @@ crypto_get_stored_dynamic_dh_modulus(const char *fname) } /* 'fname' contains the DH parameters stored in base64-ed DER - format. We are only interested in the DH modulus. */ + * format. We are only interested in the DH modulus. + * NOTE: We allocate more storage here than we need. Since we're already + * doing that, we can also add 1 byte extra to appease Coverity's + * scanner. */ - cp = base64_decoded_dh = tor_malloc_zero(strlen(contents)); + cp = base64_decoded_dh = tor_malloc_zero(strlen(contents) + 1); length = base64_decode((char *)base64_decoded_dh, strlen(contents), contents, strlen(contents)); if (length < 0) {

1 0

[tor/master] Merge remote-tracking branch 'sebastian/coverity_strlen_fp'
by nickm＠torproject.org 02 Dec '11

02 Dec '11

commit cd10013218a6b1cc42568eb91ed9372e6caddf1f Merge: 5ffa710 95af915 Author: Nick Mathewson <nickm(a)torproject.org> Date: Fri Dec 2 00:24:33 2011 -0500 Merge remote-tracking branch 'sebastian/coverity_strlen_fp' src/common/crypto.c | 7 +++++-- src/test/test_config.c | 2 +- 2 files changed, 6 insertions(+), 3 deletions(-)

1 0

[tor/master] Don't segfault when checking whether a not-yet-used intro point should expire
by nickm＠torproject.org 02 Dec '11

02 Dec '11

commit 5ffa7102c00385445d3855bbc481668edf62d139 Author: Robert Ransom <rransom.8774(a)gmail.com> Date: Thu Dec 1 15:26:45 2011 -0800 Don't segfault when checking whether a not-yet-used intro point should expire Found by katmagic. Bugfix on the #3460 branch, not yet in any release. --- src/or/rendservice.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/src/or/rendservice.c b/src/or/rendservice.c index a360d5c..177f3bf 100644 --- a/src/or/rendservice.c +++ b/src/or/rendservice.c @@ -1957,7 +1957,8 @@ intro_point_should_expire_now(rend_intro_point_t *intro, return 1; } - if (digestmap_size(intro->accepted_intro_rsa_parts) >= + if (intro->accepted_intro_rsa_parts != NULL && + digestmap_size(intro->accepted_intro_rsa_parts) >= INTRO_POINT_LIFETIME_INTRODUCTIONS) { /* This intro point has been used too many times. Expire it now. */ return 1;

1 0

[torflow/master] Fix variable shadow, and add some logs.
by mikeperry＠torproject.org 02 Dec '11

02 Dec '11

commit c183c9fe3970c9bd022d1d7816860dab62f34d07 Author: Mike Perry <mikeperry-git(a)fscked.org> Date: Thu Dec 1 21:03:06 2011 -0800 Fix variable shadow, and add some logs. --- NetworkScanners/BwAuthority/aggregate.py | 32 +++++++++++++++-------------- 1 files changed, 17 insertions(+), 15 deletions(-) diff --git a/NetworkScanners/BwAuthority/aggregate.py b/NetworkScanners/BwAuthority/aggregate.py index ccf6436..a2372af 100755 --- a/NetworkScanners/BwAuthority/aggregate.py +++ b/NetworkScanners/BwAuthority/aggregate.py @@ -464,32 +464,34 @@ def main(argv): plog("INFO", "PID control enabled") if cs_junk.group_by_class: - for c in ["Guard+Exit", "Guard", "Exit", "Middle"]: - c_nodes = filter(lambda n: n.node_class() == c, nodes.itervalues()) - true_filt_avg[c] = sum(map(lambda n: n.filt_bw, c_nodes))/float(len(c_nodes)) - true_strm_avg[c] = sum(map(lambda n: n.strm_bw, c_nodes))/float(len(c_nodes)) - true_circ_avg[c] = sum(map(lambda n: (1.0-n.circ_fail_rate), + for cl in ["Guard+Exit", "Guard", "Exit", "Middle"]: + c_nodes = filter(lambda n: n.node_class() == cl, nodes.itervalues()) + true_filt_avg[cl] = sum(map(lambda n: n.filt_bw, c_nodes))/float(len(c_nodes)) + true_strm_avg[cl] = sum(map(lambda n: n.strm_bw, c_nodes))/float(len(c_nodes)) + true_circ_avg[cl] = sum(map(lambda n: (1.0-n.circ_fail_rate), c_nodes))/float(len(c_nodes)) - plog("INFO", "Network true_filt_avg["+c+"]: "+str(true_filt_avg[c])) - plog("INFO", "Network true_circ_avg["+c+"]: "+str(true_circ_avg[c])) + plog("INFO", "Network true_filt_avg["+cl+"]: "+str(true_filt_avg[cl])) + plog("INFO", "Network true_circ_avg["+cl+"]: "+str(true_circ_avg[cl])) else: filt_avg = sum(map(lambda n: n.filt_bw, nodes.itervalues()))/float(len(nodes)) strm_avg = sum(map(lambda n: n.strm_bw, nodes.itervalues()))/float(len(nodes)) circ_avg = sum(map(lambda n: (1.0-n.circ_fail_rate), nodes.itervalues()))/float(len(nodes)) - for c in ["Guard+Exit", "Guard", "Exit", "Middle"]: - true_filt_avg[c] = filt_avg - true_strm_avg[c] = strm_avg - true_circ_avg[c] = circ_avg + for cl in ["Guard+Exit", "Guard", "Exit", "Middle"]: + true_filt_avg[cl] = filt_avg + true_strm_avg[cl] = strm_avg + true_circ_avg[cl] = circ_avg + plog("INFO", "Network true_filt_avg: "+str(true_filt_avg["Middle"])) + plog("INFO", "Network true_circ_avg: "+str(true_circ_avg["Middle"])) else: plog("INFO", "PID control disabled") filt_avg = sum(map(lambda n: n.filt_bw*(1.0-n.circ_fail_rate), nodes.itervalues()))/float(len(nodes)) strm_avg = sum(map(lambda n: n.strm_bw*(1.0-n.circ_fail_rate), nodes.itervalues()))/float(len(nodes)) - for c in ["Guard+Exit", "Guard", "Exit", "Middle"]: - true_filt_avg[c] = filt_avg - true_strm_avg[c] = strm_avg + for cl in ["Guard+Exit", "Guard", "Exit", "Middle"]: + true_filt_avg[cl] = filt_avg + true_strm_avg[cl] = strm_avg prev_votes = None @@ -525,7 +527,7 @@ def main(argv): tot_net_bw = 0 for n in nodes.itervalues(): n.fbw_ratio = n.filt_bw/true_filt_avg[n.node_class()] - n.sbw_ratio = n.strm_bw/true_strm_avg[n.node_clasS()] + n.sbw_ratio = n.strm_bw/true_strm_avg[n.node_class()] if cs_junk.bwauth_pid_control: if cs_junk.use_desc_bw:

1 0

[torflow/master] Fix an exception that can happen with very large bandwidths.
by mikeperry＠torproject.org 02 Dec '11

02 Dec '11

commit 325bfcc0f000808c5528f90d69996ab0a186aa10 Author: Mike Perry <mikeperry-git(a)fscked.org> Date: Thu Dec 1 16:56:46 2011 -0800 Fix an exception that can happen with very large bandwidths. We have a cap+log warn below that we will hit in this case instead of crashing. --- NetworkScanners/BwAuthority/aggregate.py | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/NetworkScanners/BwAuthority/aggregate.py b/NetworkScanners/BwAuthority/aggregate.py index 46dce2e..19697eb 100755 --- a/NetworkScanners/BwAuthority/aggregate.py +++ b/NetworkScanners/BwAuthority/aggregate.py @@ -345,7 +345,7 @@ def main(argv): ns_list = c.get_network_status() for n in ns_list: if n.bandwidth == None: n.bandwidth = -1 - ns_list.sort(lambda x, y: y.bandwidth - x.bandwidth) + ns_list.sort(lambda x, y: int(y.bandwidth/10000.0 - x.bandwidth/10000.0)) for n in ns_list: if n.bandwidth == -1: n.bandwidth = None got_ns_bw = False

1 0