<h1 class="SECT1"><a name="WHATSNEW" id="WHATSNEW">3. What's New in this
Release</a></h1>
- <p><span class="APPLICATION">Privoxy 3.0.24</span> stable contains a
- couple of new features but is mainly a bug-fix release. Two of the fixed
- bugs are security issues and may be used to remotely trigger crashes on
- platforms that carefully check memory accesses (most don't).</p>
+ <p><span class="APPLICATION">Privoxy 3.0.25</span> beta introduces
+ client-based tags and includes a couple of minor improvements. It will be
+ followed by a stable release in the near future.</p>
<ul>
<li>
- <p>Security fixes (denial of service):</p>
+ <p>Bug fixes:</p>
<ul>
<li>
- <p>Prevent invalid reads in case of corrupt chunk-encoded
- content. CVE-2016-1982. Bug discovered with afl-fuzz and
- AddressSanitizer.</p>
+ <p>Always use the current toggle state for new requests.
+ Previously new requests on reused connections inherited the
+ toggle state from the previous request even though the toggle
+ state could have changed. Reported by Robert Klemme.</p>
</li>
<li>
- <p>Remove empty Host headers in client requests. Previously they
- would result in invalid reads. CVE-2016-1983. Bug discovered with
- afl-fuzz and AddressSanitizer.</p>
+ <p>Fixed two buffer-overflows in the (deprecated) static pcre
+ code. These bugs are not considered security issues as the input
+ is trusted. Found with afl-fuzz and ASAN.</p>
</li>
</ul>
</li>
<li>
- <p>Bug fixes:</p>
+ <p>General improvements:</p>
<ul>
<li>
- <p>When using socks5t, send the request body optimistically as
- well. Previously the request body wasn't guaranteed to be sent at
- all and the error message incorrectly blamed the server. Fixes
- #1686 reported by Peter Müller and G4JC.</p>
+ <p>Added support for client-specific tags which allow Privoxy
+ admins to pre-define tags that are set for all requests from
+ clients that previously opted in through the CGI interface. They
+ are useful in multi-user setups where admins may want to allow
+ users to disable certain actions and filters for themselves
+ without affecting others. In single-user setups they are useful
+ to allow more fine-grained toggling. For example to disable
+ request blocking while still crunching cookies, or to disable
+ experimental filters only. This is an experimental feature, the
+ syntax and behaviour may change in future versions. Sponsored by
+ Robert Klemme.</p>
</li>
<li>
- <p>Fixed buffer scaling in execute_external_filter() that could
- lead to crashes. Submitted by Yang Xia in #892.</p>
+ <p>Dynamic filters and taggers now support a $listen-address
+ variable which contains the address the request came in on. For
+ external filters the variable is called $PRIVOXY_LISTEN_ADDRESS.
+ Original patch contributed by pursievro.</p>
</li>
<li>
- <p>Fixed crashes when executing external filters on platforms
- like Mac OS X. Reported by Jonathan McKenzie on
- ijbswa-users@.</p>
+ <p>Add client-header-tagger 'listen-address'.</p>
</li>
<li>
- <p>Properly parse ACL directives with ports when compiled with
- HAVE_RFC2553. Previously the port wasn't removed from the host
- and in case of 'permit-access 127.0.0.1 example.org:80' Privoxy
- would try (and fail) to resolve "example.org:80" instead of
- example.org. Reported by Pak Chan on ijbswa-users@.</p>
+ <p>Include the listen-address in the log message when logging new
+ requests. Patch contributed by pursievro.</p>
</li>
<li>
- <p>Check requests more carefully before serving them forcefully
- when blocks aren't enforced. Privoxy always adds the force token
- at the beginning of the path, but would previously accept it
- anywhere in the request line. This could result in requests being
- served that should be blocked. For example in case of pages that
- were loaded with force and contained JavaScript to create
- additionally requests that embed the origin URL (thus inheriting
- the force prefix). The bug is not considered a security issue and
- the fix does not make it harder for remote sites to intentionally
- circumvent blocks if Privoxy isn't configured to enforce them.
- Fixes #1695 reported by Korda.</p>
+ <p>Turn invalid max-client-connections values into fatal
+ errors.</p>
</li>
<li>
- <p>Normalize the request line in intercepted requests to make
- rewriting the destination more convenient. Previously rewrites
- for intercepted requests were expected to fail unless $hostport
- was being used, but they failed "the wrong way" and would result
- in an out-of-memory message (vanilla host patterns) or a crash
- (extended host patterns). Reported by "Guybrush Threepwood" in
- #1694.</p>
+ <p>The show-status page now shows whether or not dates before
+ 1970 and after 2038 are expected to be handled properly. This is
+ mainly useful for Privoxy-Regression-Test but could also come
+ handy when dealing with time-related support requests.</p>
</li>
<li>
- <p>Enable socket lingering for the correct socket. Previously it
- was repeatedly enabled for the listen socket instead of for the
- accepted socket. The bug was found by code inspection and did not
- cause any (reported) issues.</p>
+ <p>On Mac OS X the thread id in log messages are more likely to
+ be unique now.</p>
</li>
<li>
- <p>Detect and reject parameters for parameter-less actions.
- Previously they were silently ignored.</p>
+ <p>When complaining about missing filters, the filter type is
+ logged as well.</p>
</li>
<li>
- <p>Fixed invalid reads in internal and outdated pcre code. Found
- with afl-fuzz and AddressSanitizer.</p>
+ <p>A couple of harmless coverity warnings were silenced (CID
+ #161202, CID #161203, CID #161211).</p>
</li>
+ </ul>
+ </li>
- <li>
- <p>Prevent invalid read when loading invalid action files. Found
- with afl-fuzz and AddressSanitizer.</p>
- </li>
+ <li>
+ <p>Action file improvements:</p>
+ <ul>
<li>
- <p>Windows build: Use the correct function to close the event
- handle. It's unclear if this bug had a negative impact on
- Privoxy's behaviour. Reported by Jarry Xu in #891.</p>
+ <p>Filtering is disabled for Range requests to let download
+ resumption and Windows updates work with the default
+ configuration.</p>
</li>
<li>
- <p>In case of invalid forward-socks5(t) directives, use the
- correct directive name in the error messages. Previously they
- referred to forward-socks4t failures. Reported by Joel Verhagen
- in #889.</p>
+ <p>Unblock ".ardmediathek.de/". Reported by ThTomate in #932.</p>
</li>
</ul>
</li>
<li>
- <p>General improvements:</p>
+ <p>Documentation improvements:</p>
<ul>
<li>
- <p>Set NO_DELAY flag for the accepting socket. This significantly
- reduces the latency if the operating system is not configured to
- set the flag by default. Reported by Johan Sintorn in #894.</p>
+ <p>Add FAQ entry for crashes caused by memory limits.</p>
</li>
<li>
- <p>Allow to build with mingw x86_64. Submitted by Rustam
- Abdullaev in #135.</p>
+ <p>Remove obsolete FAQ entry about a bug in PHP 4.2.3.</p>
</li>
<li>
- <p>Introduce the new forwarding type 'forward-webserver'.
- Currently it is only supported by the forward-override{} action
- and there's no config directive with the same name. The
- forwarding type is similar to 'forward', but the request line
- only contains the path instead of the complete URL.</p>
+ <p>Mention the new mailing lists were appropriate. As the
+ archives have not been migrated, continue to mention the archives
+ at SF in the contacting section for now.</p>
</li>
<li>
- <p>The CGI editor no longer treats 'standard.action' special.
- Nowadays the official "standards" are part of default.action and
- there's no obvious reason to disallow editing them through the
- cgi editor anyway (if the user decided that the lack of
- authentication isn't an issue in her environment).</p>
+ <p>Note that the templates should be adjusted if Privoxy is
+ running as intercepting proxy without getting all requests.</p>
</li>
<li>
- <p>Improved error messages when rejecting intercepted requests
- with unknown destination.</p>
+ <p>A bunch of links were converted to https://.</p>
</li>
<li>
- <p>A couple of log messages now include the number of active
- threads.</p>
+ <p>Rephrase onion service paragraph to make it more obvious that
+ Tor is involved and that the whole website (and not just the
+ homepage) is available as onion service.</p>
</li>
<li>
- <p>Removed non-standard Proxy-Agent headers in HTTP snipplets to
- make testing more convenient.</p>
+ <p>Streamline the "More information" section on the homepage
+ further by additionally ditching the link to the 'See also'
+ section of the user manual. The section contains mostly links
+ that are directly reachable from the homepage already and the
+ rest is not significant enough to get a link from the
+ homepage.</p>
</li>
<li>
- <p>Include the error code for pcre errors Privoxy does not
- recognize.</p>
+ <p>Change the add-header{} example to set the DNT header and use
+ a complete section to make copy and pasting more convenient. Add
+ a comment to make it obvious that adding the header is not
+ recommended for obvious reasons. Using the DNT header as example
+ was suggested by Leo Wzukw.</p>
</li>
<li>
- <p>Config directives with numerical arguments are checked more
- carefully.</p>
+ <p>Streamline the support-and-service template Instead of linking
+ to the various support trackers (whose URLs hopefully change
+ soon), link to the contact section of the user manual to increase
+ the chances that users actually read it.</p>
</li>
<li>
- <p>Privoxy's malloc() wrapper has been changed to prevent
- zero-size allocations which should only occur as the result of
- bugs.</p>
+ <p>Add a FAQ entry for tainted sockets.</p>
</li>
<li>
- <p>Various cosmetic changes.</p>
+ <p>More sections in the documentation have stable URLs now.</p>
</li>
- </ul>
- </li>
- <li>
- <p>Action file improvements:</p>
+ <li>
+ <p>FAQ: Explain why 'ping config.privoxy.org' is not expected to
+ reach a local Privoxy installation.</p>
+ </li>
- <ul>
<li>
- <p>Unblock ".deutschlandradiokultur.de/". Reported by u302320 in
- #924.</p>
+ <p>Note that donations done through Zwiebelfreunde e.V. currently
+ can't be checked automatically.</p>
</li>
<li>
- <p>Add two fast-redirect exceptions for "yandex.ru".</p>
+ <p>Updated section regarding starting Privoxy under OS X.</p>
</li>
<li>
- <p>Disable filter{banners-by-size} for ".plasmaservice.de/".</p>
+ <p>Use dedicated start instructions for FreeBSD and
+ ElectroBSD.</p>
</li>
<li>
- <p>Unblock "klikki.fi/adv/".</p>
+ <p>Removed release instructions for AIX. They haven't been
+ working for years and unsurprisingly nobody seems to care.</p>
</li>
<li>
- <p>Block requests for "resources.infolinks.com/". Reported by
- "Black Rider" on ijbswa-users@.</p>
+ <p>Removed obsolete reference to the solaris-dist target.</p>
</li>
<li>
- <p>Block a bunch of criteo domains. Reported by Black Rider.</p>
+ <p>Updated the release instructions for FreeBSD.</p>
</li>
<li>
- <p>Block "abs.proxistore.com/abe/". Reported by Black Rider.</p>
+ <p>Removed unfinished release instructions for Amiga OS and HP-UX
+ 11.</p>
</li>
<li>
- <p>Disable filter{banners-by-size} for
- ".black-mosquito.org/".</p>
+ <p>Added a pointer to the Cygwin Time Machine for getting the
+ last release of Cygwin version 1.5 to use for building Privoxy on
+ Windows.</p>
</li>
<li>
- <p>Disable fast-redirects for "disqus.com/".</p>
+ <p>Various typos have been fixed.</p>
</li>
</ul>
</li>
<li>
- <p>Documentation improvements:</p>
+ <p>Infrastructure improvements:</p>
<ul>
<li>
- <p>FAQ: Explicitly point fingers at ASUS as an example of a
- company that has been reported to force malware based on Privoxy
- upon its customers.</p>
+ <p>The website is no longer hosted at SourceForge and can be
+ reached through https now.</p>
</li>
<li>
- <p>Correctly document the action type for a bunch of
- "multi-value" actions that were incorrectly documented to be
- "parameterized". Reported by Gregory Seidman on
- ijbswa-users@.</p>
+ <p>The mailing lists at SourceForge have been deprecated, you can
+ subscribe to the new ones at: https://lists.privoxy.org/</p>
</li>
<li>
- <p>Fixed the documented type of the forward-override{} action
- which is obviously 'parameterized'.</p>
+ <p>Migrating the remaining services from SourceForge is work in
+ progress (TODO list item #53).</p>
</li>
</ul>
</li>
<li>
- <p>Website improvements:</p>
+ <p>Build system improvements:</p>
<ul>
<li>
- <p>Users who don't trust binaries served by SourceForge can get
- them from a mirror. Migrating away from SourceForge is planned
- for 2016 (TODO list item #53).</p>
+ <p>Add configure argument to optimistically redefine FD_SETSIZE
+ with the intent to change the maximum number of client
+ connections Privoxy can handle. Only works with some libcs.
+ Sponsored by Robert Klemme.</p>
+ </li>
+
+ <li>
+ <p>Let the tarball-dist target skip files in ".git".</p>
+ </li>
+
+ <li>
+ <p>Let the tarball-dist target work in cwds other than
+ current.</p>
+ </li>
+
+ <li>
+ <p>Make the 'clean' target faster when run from a git
+ repository.</p>
+ </li>
+
+ <li>
+ <p>Include tools in the generic distribution.</p>
+ </li>
+
+ <li>
+ <p>Let the gen-dist target work in cwds other than current.</p>
+ </li>
+
+ <li>
+ <p>Sort find output that is used for distribution tarballs to get
+ reproducible results.</p>
+ </li>
+
+ <li>
+ <p>Don't add '-src' to the name of the tar ball generated by the
+ gen-dist target. The package isn't a source distribution but a
+ binary package. While at it, use a variable for the name to
+ reduce the chances that the various references get out of sync
+ and fix the gen-upload target which was looking in the wrong
+ directory.</p>
+ </li>
+
+ <li>
+ <p>Add regression-tests.action to the files that are
+ distributed.</p>
+ </li>
+
+ <li>
+ <p>The gen-dist target which was broken since 2002 (r1.92) has
+ been fixed.</p>
+ </li>
+
+ <li>
+ <p>Remove genclspec.sh which has been obsolete since 2009.</p>
+ </li>
+
+ <li>
+ <p>Remove obsolete reference to Redhat spec file.</p>
+ </li>
+
+ <li>
+ <p>Remove the obsolete announce target which has been commented
+ out years ago.</p>
+ </li>
+
+ <li>
+ <p>Let rsync skip files if the checksums match.</p>
+ </li>
+ </ul>
+ </li>
+
+ <li>
+ <p>Privoxy-Regression-Test:</p>
+
+ <ul>
+ <li>
+ <p>Add a "Default level offset" directive which can be used to
+ change the default level by a given value. This directive affects
+ all tests located after it until the end of the file or a another
+ "Default level offset" directive is reached. The purpose of this
+ directive is to make it more convenient to skip similar tests in
+ a given file without having to remove or disable the tests
+ completely.</p>
+ </li>
+
+ <li>
+ <p>Let test level 17 depend on FEATURE_64_BIT_TIME_T instead of
+ FEATURE_PTHREAD which has no direct connection to the time_t
+ size.</p>
+ </li>
+
+ <li>
+ <p>Fix indentation in perldoc examples.</p>
+ </li>
+
+ <li>
+ <p>Don't overlook directives in the first line of the action
+ file.</p>
+ </li>
+
+ <li>
+ <p>Bump version to 0.7.</p>
</li>
<li>
- <p>The website is now available as onion service
- (http://jvauzb4sb3bwlsnc.onion/).</p>
+ <p>Fix detection of the Privoxy version now that https:// is used
+ for the website.</p>
</li>
</ul>
</li>