Skip to content
This repository has been archived by the owner on Jul 30, 2021. It is now read-only.

Commit

Permalink
Merge changelogs from the 2.0 branch
Browse files Browse the repository at this point in the history
git-svn-id: http://www.varnish-cache.org/svn/trunk@4606 d4fa192b-c00b-0410-8231-f00ffab90ce4
  • Loading branch information
Tollef Fog Heen committed Mar 10, 2010
1 parent c956cc4 commit fe56044
Show file tree
Hide file tree
Showing 8 changed files with 383 additions and 0 deletions.
10 changes: 10 additions & 0 deletions doc/Makefile.am
@@ -1,6 +1,11 @@
# $Id$

CHANGELOGS = \
changes-2.0.6.html \
changes-2.0.5.html \
changes-2.0.4.html \
changes-2.0.3.html \
changes-2.0.2.html \
changes-2.0.1.html \
changes-2.0.html \
changes-1.1.2.html \
Expand All @@ -9,6 +14,11 @@ CHANGELOGS = \
changes-1.0.4.html

XML = \
changes-2.0.5-2.0.6.xml \
changes-2.0.4-2.0.5.xml \
changes-2.0.3-2.0.4.xml \
changes-2.0.2-2.0.3.xml \
changes-2.0.1-2.0.2.xml \
changes-2.0-2.0.1.xml \
changes-1.1.2-2.0.xml \
changes-1.1.1-1.1.2.xml \
Expand Down
25 changes: 25 additions & 0 deletions doc/changes-2.0.1-2.0.2.xml
@@ -0,0 +1,25 @@
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE group [
<!ENTITY mdash "&#8212;">
]>
<!-- $Id$ -->
<group from="2.0.1" to="2.0.2">
<subsystem>
<name>varnishd</name>

<change type="bug" ref="345">
<para>In high-load situations, when using
ESI, <code>varnishd</code> would sometimes mishandle objects and
crash. This has been worked around. </para>
</change>
</subsystem>

<subsystem>
<name>varnishreplay</name>

<change type="bug">
<para><code>varnishreplay</code> did not work correctly on
Linux, due to a too small stack. This has now been fixed.</para>
</change>
</subsystem>
</group>
98 changes: 98 additions & 0 deletions doc/changes-2.0.3-2.0.4.xml
@@ -0,0 +1,98 @@
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE group [
<!ENTITY mdash "&#8212;">
]>
<!-- $Id$ -->
<group from="2.0.3" to="2.0.4">
<subsystem>
<name>varnishd</name>

<change type="enh">
<para>Make Varnish more portable by pulling in fixes for
Solaris and NetBSD.</para>
</change>

<change type="bug">
<para>Correct description of -a in the manual page.</para>
</change>

<change type="bug">
<para>Ensure we are compiling in C99 mode.</para>
</change>

<change type="bug" ref="458">
<para>If error was called with a null reason, we would crash on
Solaris. Make sure this no longer happens.</para>
</change>

<change type="bug" ref="428">
<para>Varnish used to crash if you asked it to use a
non-existent waiter. This has now been fixed.</para>
</change>

<change type="enh">
<para>Add documentation to the default VCL explaining that
using <code>Connection: close</code> in <code>vcl_close</code>
is generally a good idea.</para>
</change>

<change type="enh">
<para>Add minimal facility for dealing with TELNET option
negotiation by returning WONT to DO and DONT requests.</para>
</change>

<change type="enh">
<para>If the backend is unhealthy, use a graced object if one is
available.</para>
</change>

<change type="enh">
<para>Make <code>server.hostname</code>
and <code>server.identity</code> available to VCL. The latter
can be set with the <code>-i</code> parameter
to <code>varnishd</code>.</para>
</change>

<change type="enh">
<para>Make <code>restart</code> available
from <code>vcl_error</code>.</para>
</change>

<change type="enh">
<para>Previously, only the TTL of an object was considered in
whether it would be marked as cacheable. This has been changed
to take the grace into consideration as well.</para>
</change>

<change type="bug" ref="445">
<para>Previously, if an included ESI fragment had a zero size,
we would send out a zero-sized chunk which signifies
end-of-transmission. We now ignore zero-sized chunks.</para>
</change>

<change type="bug">
<para>We accidentially slept for far too long when we reached
the maximum number of open file descriptors. This has been
corrected and <code>accept_fd_holdoff</code> now works
correctly.</para>
</change>
<change type="bug" ref="476">
<para>Previously, when ESI processing, we did not look at the
full length, but stopped at the first NULL byte. We no longer
do that, enabling ESI processing of binary data.</para>
</change>
</subsystem>

<subsystem>
<name>varnishtest</name>
<change type="bug">
<para>Make sure system "..." returns successfully to ensure
test failures do not go unnoticed.</para>
</change>

<change type="enh">
<para>Make it possible to send NULL bytes through the testing
framework.</para>
</change>
</subsystem>
</group>
12 changes: 12 additions & 0 deletions doc/changes-2.0.3.xml
@@ -0,0 +1,12 @@
<?xml version="1.0" encoding="utf-8"?>
<?xml-stylesheet type="text/xml" href="changes-html.xsl"?>
<!DOCTYPE changelog [
<!ENTITY mdash "&#8212;">
]>
<!-- $Id$ -->
<changelog xmlns:xi="http://www.w3.org/2001/XInclude">
<package>Varnish</package>
<version>2.0.3</version>

<xi:include href="changes-2.0.2-2.0.3.xml"/>
</changelog>
161 changes: 161 additions & 0 deletions doc/changes-2.0.4-2.0.5.xml
@@ -0,0 +1,161 @@
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE group [
<!ENTITY mdash "&#8212;">
]>
<!-- $Id$ -->
<group from="2.0.4" to="2.0.5">
<subsystem>
<name>varnishd</name>

<change type="bug" ref="498">
<para>Handle object workspace overruns better.</para>
</change>

<change type="enh">
<para>Allow turning off ESI processing per request by using
<code>set req.esi = off</code>.</para>
</change>

<change type="enh">
<para>Tell the kernel that we expect to use the mmap-ed file in
a random fashion. On Linux, this turns off/down readahead and
increases performance.</para>
</change>

<change type="enh">
<para>Make it possible to change the maximum number of HTTP
headers we allow by
passing <code>--with-max-header-fields=NUM</code> rather than
changing the code.</para>
</change>

<change type="enh" ref="428">
<para>Implement support for HTTP continuation lines.</para>
</change>

<change type="enh">
<para>Change how connections are closed and only use SO_LINGER
for orderly connection closure. This should hopefully make
worker threads less prone to hangups on network problems.</para>
</change>

<change type="bug" ref="502">
<para>Handle multi-element purges correctly. Previously we
ended up with parse errors when this was done from VCL.</para>
</change>

<change type="bug" ref="506, 549">
<para>Handle illegal responses from the backend better by
serving a 503 page rather than panic-ing.</para>
</change>

<change type="enh">
<para>When we run into an assertion that is not true, Varnish
would previously dump a little bit of information about itself.
Extend that information with a backtrace. Note that this relies
on the varnish binary being unstripped.</para>
</change>

<change type="enh">
<para>Add a session_max parameter that limits the maximum
number of sessions we keep open before we start dropping new
connections summarily.</para>
</change>

<change type="bug">
<para>Try to consume less memory when doing ESI processing by
properly rolling back used workspace after processing an
object. This should make it possible to
turn <code>sess_workspace</code> quite a bit for users with
ESI-heavy pages.</para>
</change>

<change type="enh">
<para>Turn on <code>session_linger</code> by default. Tests
have shown that <code>session_linger</code> helps a fair bit
with performance.</para>
</change>

<change type="enh">
<para>Rewrite the epoll acceptor for better performance. This
should lead to both higher processing rates and maximum number
of connections on Linux.</para>
</change>

<change type="enh">
<para>Add If-None-Match support, this gives significant
bandwidth savings for users with compliant browsers.</para>
</change>

<change type="bug" ref="529">
<para>RFC2616 specifies
that <code>ETag</code>, <code>Content-Location</code>, <code>Expires</code>, <code>Cache-Control</code>
and <code>Vary</code> should be emitted when delivering a
response with the 304 response code.</para>
</change>

<change type="bug" ref="531">
<para>Various fixes which makes Varnish compile and work on AIX.</para>
</change>

<change type="enh">
<para>Turn on TCP_DEFER_ACCEPT on Linux. This should make us
less suspecible to denial of service attacks as well as give us
slightly better performance.</para>
</change>

<change type="enh">
<para>Add an <code>.initial</code> property to the backend
probe specification. This is the number of good probes we
pretend to have seen. The default is one less than .threshold,
which means the first probe will decide if we consider the
backend healthy.</para>
</change>

<change type="enh">
<para>Make it possible to compare strings against other
string-like objects, not just plain strings. This allows you to
compare two headers, for instance.</para>
</change>

<change type="bug">
<para>When support for <code>restart</code>
in <code>vcl_error</code> was added, there was no check to
prevent infinte recursion. This has now been fixed.
</para>
</change>

<change type="enh">
<para>Turn on purge_dups by default. This should make us
consume less memory when there are many bans for the same
pattern added.</para>
</change>

<change type="enh">
<para>Add a new log tag called <code>FetchError</code> which
tries to explain why we could not fetch an object from the
backend.</para>
</change>

<change type="enh">
<para>Change the default <code>srcaddr_ttl</code> to 0. It is
not used by anything and has been removed in the development
version. This will increase performance somewhat.</para>
</change>

</subsystem>

<subsystem>
<name>varnishtop</name>
<change type="bug" ref="514">
<para>varnishtop did not handle variable-length log fields
correctly. This is now fixed.</para>
</change>

<change type="enh">
<para>varnishtop previously did not print the name of the tag,
which made it very hard to understand. We now print out the
tag name.</para>
</change>
</subsystem>
</group>
53 changes: 53 additions & 0 deletions doc/changes-2.0.5-2.0.6.xml
@@ -0,0 +1,53 @@
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE group [
<!ENTITY mdash "&#8212;">
]>
<!-- $Id$ -->
<group from="2.0.5" to="2.0.6">
<subsystem>
<name>varnishd</name>

<change type="bug" ref="585">
<para>2.0.5 had an off-by-one error in the ESI handling causing
includes to fail a large part of the time. This has now been
fixed.</para>
</change>

<change type="bug" ref="584">
<para>Try harder to not confuse backends when sending them
backend probes. We half-closed the connection, something some
backends thought meant we had dropped the connection. Stop
doing so, and add the capability for specifying the expected
response code.</para>
</change>

<change type="bug">
<para>In 2.0.5, session lingering was turned on. This caused
statistics to not be counted often enough in some cases. This
has now been fixed.</para>
</change>

<change type="bug" ref="547">
<para>Avoid triggering an assert if the other end closes the
connection while we are lingering and waiting for another
request from them.</para>
</change>

<change type="bug" ref="577">
<para>When generating backtraces, prefer the built-in backtrace
function if such exists. This fixes a problem compiling 2.0.5
on Solaris.</para>
</change>

<change type="enh" ref="572">
<para>Make it possible to specify the per-thread stack size.
This might be useful on 32 bit systems with their limited
address space.</para>
</change>

<change type="bug" ref="559">
<para>Document the <code>-C</code> option
to <code>varnishd</code>.</para>
</change>
</subsystem>
</group>
12 changes: 12 additions & 0 deletions doc/changes-2.0.5.xml
@@ -0,0 +1,12 @@
<?xml version="1.0" encoding="utf-8"?>
<?xml-stylesheet type="text/xml" href="changes-html.xsl"?>
<!DOCTYPE changelog [
<!ENTITY mdash "&#8212;">
]>
<!-- $Id$ -->
<changelog xmlns:xi="http://www.w3.org/2001/XInclude">
<package>Varnish</package>
<version>2.0.5</version>

<xi:include href="changes-2.0.4-2.0.5.xml"/>
</changelog>

0 comments on commit fe56044

Please sign in to comment.