Redis 2.6.14

Sentinel: parse new INFO replication output correctly.
Sentinel was not able to detect slaves when connected to a very recent version of Redis master since a previos non-backward compatible change to INFO broken the parsing of the slaves ip:port INFO output. This fixes issue #1164
2013-06-20 10:36:40 +02:00 · 2013-06-20 10:24:43 +02:00 · 2013-06-19 18:53:17 +02:00 · 2013-06-19 18:29:10 +02:00 · 2013-06-04 15:56:26 +02:00 · 2013-05-14 11:22:50 +02:00
53 changed files with 1129 additions and 605 deletions
@@ -1,8 +1,13 @@
 .*.swp
 *.o
-*.rdb
 *.log
-redis-*
+dump.rdb
+redis-benchmark
+redis-check-aof
+redis-check-dump
+redis-cli
+redis-sentinel
+redis-server
 doc-tools
 release
 misc/*
@@ -14,6 +14,56 @@ HIGH:     There is a critical bug that may affect a subset of users. Upgrade!
 CRITICAL: There is a critical bug affecting MOST USERS. Upgrade ASAP.
 --------------------------------------------------------------------------------

+--[ Redis 2.6.14 ] Release date: 20 Jun 2013
+
+UPGRADE URGENCY: HIGH because of the following two issues:
+    * Lua scripting + Replication + AOF in slaves problem (see Issue #1164).
+    * AOF + expires possible race condition (see Issue #1079).
+
+* [FIX] AOF bug: expire could be removed from key on AOF rewrite.
+* [FIX] Allow writes from scripts called by AOF loading in read-only slaves.
+* [FIX] Sentinel: parse new verison of INFO replication output correctly.
+* [NEW] Reset masterauth if an empty string is configured.
+
+--[ Redis 2.6.13 ]
+
+UPGRADE URGENCY: MODERATE, nothing very critical but upgrading is suggested
+                 if you experienced:
+
+                 1) Strange issues with Lua scripting.
+                 2) Not reconfigured reappearing master using Sentinel.
+                 3) Server continusly trying to save on save error.
+
+                 This version of Redis may also help with AOF and slow / busy
+                 disks and latency issues.
+
+* [FIX] Throttle BGSAVE attempt on saving error.
+* [FIX] redis-cli: raise error on bad command line switch.
+* [FIX] Redis/Jemalloc Gitignore were too aggressive.
+* [FIX] Test: fix RDB test checking file permissions.
+* [FIX] Sentinel: always redirect on master->slave transition.
+* [FIX] Lua updated to version 5.1.5. Fixes rare scripting issues.
+* [NEW] AOF: improved latency figures with slow/busy disks.
+* [NEW] Sentinel: turn old master into a slave when it comes back.
+* [NEW] More explicit panic message on out of memory.
+* [NEW] redis-cli: --latency-history mode implemented.
+
+--[ Redis 2.6.12 ]
+
+UPGRADE URGENCY: MODERATE, nothing very critical but a few non trivial bugs.
+
+* [BUGFIX]   redis-cli --bigkeys: don't crash with empty DB.
+* [BUGFIX]   stop-writes-on-bgsave-error now works in redis.conf
+* [BUGFIX]   Don't crash at startup if RDB is there but can't be opened.
+* [BUGFIX]   Initial value for master_link_down_since_seconds is now huge.
+* [BUGFIX]   Allow SELECT while loading the DB.
+* [BUGFIX]   Don't replicate/AOF an empty MULTI/EXEC if the transaction
+             is empty or containing just read-only commands.
+* [BUGFIX]   EXPIRE should not be able to resurrect keys (see issue #1026).
+* [IMPROVED] Extended SET back ported from Redis 2.8 / unstable
+             See http://redis.io/commands/set for more information.
+* [IMPROVED] Test suite improved.
+
 --[ Redis 2.6.11 ]

 UPGRADE URGENCY: LOW, however updating is encouraged if you have many instances
@@ -23,6 +73,8 @@ UPGRADE URGENCY: LOW, however updating is encouraged if you have many instances
 * [BUGFIX]   redis-cli: use keepalive socket option for improved reliability.
 * [BUGFIX]   Allow AUTH while loading the DB in memory.
 * [BUGFIX]   Don't segfault on unbalanced quotes while parsing config file.
+* [BUGFIX]   Don't segfault if command gets propagated to AOF / replication
+             link as another command name that was renamed in redis.conf
 * [IMPROVED] serverCron() frequency is now a runtime parameter (was REDIS_HZ).
 * [IMPROVED] Use a lot less CPU when idle, even with many configured DBs.

@@ -26,9 +26,20 @@ After building Redis is a good idea to test it, using:

    % make test

-NOTE: if after building Redis with a 32 bit target you need to rebuild it
-      with a 64 bit target you need to perform a "make clean" in the root
-      directory of the Redis distribution.
+Fixing problems building 32 bit binaries
+---------
+
+If after building Redis with a 32 bit target you need to rebuild it
+with a 64 bit target, or the other way around, you need to perform a
+"make distclean" in the root directory of the Redis distribution.
+
+In case of build errors when trying to build a 32 bit binary of Redis, try
+the following steps:
+
+* Install the packages libc6-dev-i386 (also try g++-multilib).
+* Try using the following command line instead of "make 32bit":
+
+    make CFLAGS="-m32 -march=native" LDFLAGS="-m32"

 Allocator
 ---------
@@ -2,24 +2,19 @@
 /config.stamp
 /config.log
 /config.status
-/configure
 /doc/html.xsl
 /doc/manpages.xsl
 /doc/jemalloc.xml
-/doc/jemalloc.html
-/doc/jemalloc.3
 /lib/
 /Makefile
 /include/jemalloc/internal/jemalloc_internal\.h
 /include/jemalloc/internal/size_classes\.h
 /include/jemalloc/jemalloc\.h
 /include/jemalloc/jemalloc_defs\.h
-/test/jemalloc_test\.h
 /src/*.[od]
 /test/*.[od]
 /test/*.out
 /test/[a-zA-Z_]*
 !test/*.c
 !test/*.exp
-/VERSION
 /bin/jemalloc.sh
@@ -9,7 +9,7 @@ For details and rationale, see http://www.lua.org/license.html .

 ===============================================================================

-Copyright (C) 1994-2008 Lua.org, PUC-Rio.
+Copyright (C) 1994-2012 Lua.org, PUC-Rio.

 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal
@@ -48,7 +48,7 @@ TO_MAN= lua.1 luac.1

 # Lua version and release.
 V= 5.1
-R= 5.1.4
+R= 5.1.5

 all:	$(PLAT)

@@ -3,11 +3,10 @@
 <HEAD>
 <TITLE>Lua 5.1 Reference Manual - contents</TITLE>
 <LINK REL="stylesheet" TYPE="text/css" HREF="lua.css">
-<META HTTP-EQUIV="content-type" CONTENT="text/html; charset=iso-8859-1">
+<META HTTP-EQUIV="content-type" CONTENT="text/html; charset=utf-8">
 <STYLE TYPE="text/css">
 ul {
 	list-style-type: none ;
-	list-style-position: outside ;
 }
 </STYLE>
 </HEAD>
@@ -20,7 +19,13 @@ ul {
 Lua 5.1 Reference Manual
 </H1>

-This is an online version of
+<P>
+The reference manual is the official definition of the Lua language.
+For a complete introduction to Lua programming, see the book
+<A HREF="http://www.lua.org/docs.html#pil">Programming in Lua</A>.
+
+<P>
+This manual is also available as a book:
 <BLOCKQUOTE>
 <A HREF="http://www.amazon.com/exec/obidos/ASIN/8590379833/lua-indexmanual-20">
 <IMG SRC="cover.png" ALT="" TITLE="buy from Amazon" BORDER=1 ALIGN="left" HSPACE=12>
@@ -29,128 +34,119 @@ This is an online version of
 <BR>by R. Ierusalimschy, L. H. de Figueiredo, W. Celes
 <BR>Lua.org, August 2006
 <BR>ISBN 85-903798-3-3
-<BR><A HREF="http://www.amazon.com/exec/obidos/ASIN/8590379833/lua-indexmanual-20">
-<IMG SRC="amazon.gif" ALT="[Buy from Amazon]" BORDER=0></A>
 <BR CLEAR="all">
 </BLOCKQUOTE>
-<P>

-Buy a copy of this book and 
+<P>
+<A HREF="http://www.amazon.com/exec/obidos/ASIN/8590379833/lua-indexmanual-20">Buy a copy</A>
+of this book and
 <A HREF="http://www.lua.org/donations.html">help to support</A>
 the Lua project.
-<P>

-The reference manual is the official definition of the Lua language.
-For a complete introduction to Lua programming, see the book
-<A HREF="http://www.lua.org/docs.html#books">Programming in Lua</A>.
 <P>
-
 <A HREF="manual.html">start</A>
 &middot;
 <A HREF="#contents">contents</A>
 &middot;
 <A HREF="#index">index</A>
 &middot;
-<A HREF="http://www.lua.org/manual/5.1/pt/">português</A>
-&middot;
-<A HREF="http://www.lua.org/manual/5.1/es/">español</A>
+<A HREF="http://www.lua.org/manual/">other versions</A>
 <HR>
 <SMALL>
-Copyright &copy; 2006-2008 Lua.org, PUC-Rio.
+Copyright &copy; 2006&ndash;2012 Lua.org, PUC-Rio.
 Freely available under the terms of the
-<a href="http://www.lua.org/license.html#5">Lua license</a>.
+<A HREF="http://www.lua.org/license.html">Lua license</A>.
 </SMALL>
-<P>

 <H2><A NAME="contents">Contents</A></H2>
 <UL style="padding: 0">
-<LI><A HREF="manual.html">1 - Introduction</A>
+<LI><A HREF="manual.html">1 &ndash; Introduction</A>
 <P>
-<LI><A HREF="manual.html#2">2 - The Language</A>
+<LI><A HREF="manual.html#2">2 &ndash; The Language</A>
 <UL>
-<LI><A HREF="manual.html#2.1">2.1 - Lexical Conventions</A>
-<LI><A HREF="manual.html#2.2">2.2 - Values and Types</A>
+<LI><A HREF="manual.html#2.1">2.1 &ndash; Lexical Conventions</A>
+<LI><A HREF="manual.html#2.2">2.2 &ndash; Values and Types</A>
 <UL>
-<LI><A HREF="manual.html#2.2.1">2.2.1 - Coercion</A>
+<LI><A HREF="manual.html#2.2.1">2.2.1 &ndash; Coercion</A>
 </UL>
-<LI><A HREF="manual.html#2.3">2.3 - Variables</A>
-<LI><A HREF="manual.html#2.4">2.4 - Statements</A>
+<LI><A HREF="manual.html#2.3">2.3 &ndash; Variables</A>
+<LI><A HREF="manual.html#2.4">2.4 &ndash; Statements</A>
 <UL>
-<LI><A HREF="manual.html#2.4.1">2.4.1 - Chunks</A>
-<LI><A HREF="manual.html#2.4.2">2.4.2 - Blocks</A>
-<LI><A HREF="manual.html#2.4.3">2.4.3 - Assignment</A>
-<LI><A HREF="manual.html#2.4.4">2.4.4 - Control Structures</A>
-<LI><A HREF="manual.html#2.4.5">2.4.5 - For Statement</A>
-<LI><A HREF="manual.html#2.4.6">2.4.6 - Function Calls as Statements</A>
-<LI><A HREF="manual.html#2.4.7">2.4.7 - Local Declarations</A>
+<LI><A HREF="manual.html#2.4.1">2.4.1 &ndash; Chunks</A>
+<LI><A HREF="manual.html#2.4.2">2.4.2 &ndash; Blocks</A>
+<LI><A HREF="manual.html#2.4.3">2.4.3 &ndash; Assignment</A>
+<LI><A HREF="manual.html#2.4.4">2.4.4 &ndash; Control Structures</A>
+<LI><A HREF="manual.html#2.4.5">2.4.5 &ndash; For Statement</A>
+<LI><A HREF="manual.html#2.4.6">2.4.6 &ndash; Function Calls as Statements</A>
+<LI><A HREF="manual.html#2.4.7">2.4.7 &ndash; Local Declarations</A>
 </UL>
-<LI><A HREF="manual.html#2.5">2.5 - Expressions</A>
+<LI><A HREF="manual.html#2.5">2.5 &ndash; Expressions</A>
 <UL>
-<LI><A HREF="manual.html#2.5.1">2.5.1 - Arithmetic Operators</A>
-<LI><A HREF="manual.html#2.5.2">2.5.2 - Relational Operators</A>
-<LI><A HREF="manual.html#2.5.3">2.5.3 - Logical Operators</A>
-<LI><A HREF="manual.html#2.5.4">2.5.4 - Concatenation</A>
-<LI><A HREF="manual.html#2.5.5">2.5.5 - The Length Operator</A>
-<LI><A HREF="manual.html#2.5.6">2.5.6 - Precedence</A>
-<LI><A HREF="manual.html#2.5.7">2.5.7 - Table Constructors</A>
-<LI><A HREF="manual.html#2.5.8">2.5.8 - Function Calls</A>
-<LI><A HREF="manual.html#2.5.9">2.5.9 - Function Definitions</A>
+<LI><A HREF="manual.html#2.5.1">2.5.1 &ndash; Arithmetic Operators</A>
+<LI><A HREF="manual.html#2.5.2">2.5.2 &ndash; Relational Operators</A>
+<LI><A HREF="manual.html#2.5.3">2.5.3 &ndash; Logical Operators</A>
+<LI><A HREF="manual.html#2.5.4">2.5.4 &ndash; Concatenation</A>
+<LI><A HREF="manual.html#2.5.5">2.5.5 &ndash; The Length Operator</A>
+<LI><A HREF="manual.html#2.5.6">2.5.6 &ndash; Precedence</A>
+<LI><A HREF="manual.html#2.5.7">2.5.7 &ndash; Table Constructors</A>
+<LI><A HREF="manual.html#2.5.8">2.5.8 &ndash; Function Calls</A>
+<LI><A HREF="manual.html#2.5.9">2.5.9 &ndash; Function Definitions</A>
 </UL>
-<LI><A HREF="manual.html#2.6">2.6 - Visibility Rules</A>
-<LI><A HREF="manual.html#2.7">2.7 - Error Handling</A>
-<LI><A HREF="manual.html#2.8">2.8 - Metatables</A>
-<LI><A HREF="manual.html#2.9">2.9 - Environments</A>
-<LI><A HREF="manual.html#2.10">2.10 - Garbage Collection</A>
+<LI><A HREF="manual.html#2.6">2.6 &ndash; Visibility Rules</A>
+<LI><A HREF="manual.html#2.7">2.7 &ndash; Error Handling</A>
+<LI><A HREF="manual.html#2.8">2.8 &ndash; Metatables</A>
+<LI><A HREF="manual.html#2.9">2.9 &ndash; Environments</A>
+<LI><A HREF="manual.html#2.10">2.10 &ndash; Garbage Collection</A>
 <UL>
-<LI><A HREF="manual.html#2.10.1">2.10.1 - Garbage-Collection Metamethods</A>
-<LI><A HREF="manual.html#2.10.2">2.10.2 - Weak Tables</A>
+<LI><A HREF="manual.html#2.10.1">2.10.1 &ndash; Garbage-Collection Metamethods</A>
+<LI><A HREF="manual.html#2.10.2">2.10.2 &ndash; Weak Tables</A>
 </UL>
-<LI><A HREF="manual.html#2.11">2.11 - Coroutines</A>
+<LI><A HREF="manual.html#2.11">2.11 &ndash; Coroutines</A>
 </UL>
 <P>
-<LI><A HREF="manual.html#3">3 - The Application Program Interface</A>
+<LI><A HREF="manual.html#3">3 &ndash; The Application Program Interface</A>
 <UL>
-<LI><A HREF="manual.html#3.1">3.1 - The Stack</A>
-<LI><A HREF="manual.html#3.2">3.2 - Stack Size</A>
-<LI><A HREF="manual.html#3.3">3.3 - Pseudo-Indices</A>
-<LI><A HREF="manual.html#3.4">3.4 - C Closures</A>
-<LI><A HREF="manual.html#3.5">3.5 - Registry</A>
-<LI><A HREF="manual.html#3.6">3.6 - Error Handling in C</A>
-<LI><A HREF="manual.html#3.7">3.7 - Functions and Types</A>
-<LI><A HREF="manual.html#3.8">3.8 - The Debug Interface</A>
+<LI><A HREF="manual.html#3.1">3.1 &ndash; The Stack</A>
+<LI><A HREF="manual.html#3.2">3.2 &ndash; Stack Size</A>
+<LI><A HREF="manual.html#3.3">3.3 &ndash; Pseudo-Indices</A>
+<LI><A HREF="manual.html#3.4">3.4 &ndash; C Closures</A>
+<LI><A HREF="manual.html#3.5">3.5 &ndash; Registry</A>
+<LI><A HREF="manual.html#3.6">3.6 &ndash; Error Handling in C</A>
+<LI><A HREF="manual.html#3.7">3.7 &ndash; Functions and Types</A>
+<LI><A HREF="manual.html#3.8">3.8 &ndash; The Debug Interface</A>
 </UL>
 <P>
-<LI><A HREF="manual.html#4">4 - The Auxiliary Library</A>
+<LI><A HREF="manual.html#4">4 &ndash; The Auxiliary Library</A>
 <UL>
-<LI><A HREF="manual.html#4.1">4.1 - Functions and Types</A>
+<LI><A HREF="manual.html#4.1">4.1 &ndash; Functions and Types</A>
 </UL>
 <P>
-<LI><A HREF="manual.html#5">5 - Standard Libraries</A>
+<LI><A HREF="manual.html#5">5 &ndash; Standard Libraries</A>
 <UL>
-<LI><A HREF="manual.html#5.1">5.1 - Basic Functions</A>
-<LI><A HREF="manual.html#5.2">5.2 - Coroutine Manipulation</A>
-<LI><A HREF="manual.html#5.3">5.3 - Modules</A>
-<LI><A HREF="manual.html#5.4">5.4 - String Manipulation</A>
+<LI><A HREF="manual.html#5.1">5.1 &ndash; Basic Functions</A>
+<LI><A HREF="manual.html#5.2">5.2 &ndash; Coroutine Manipulation</A>
+<LI><A HREF="manual.html#5.3">5.3 &ndash; Modules</A>
+<LI><A HREF="manual.html#5.4">5.4 &ndash; String Manipulation</A>
 <UL>
-<LI><A HREF="manual.html#5.4.1">5.4.1 - Patterns</A>
+<LI><A HREF="manual.html#5.4.1">5.4.1 &ndash; Patterns</A>
 </UL>
-<LI><A HREF="manual.html#5.5">5.5 - Table Manipulation</A>
-<LI><A HREF="manual.html#5.6">5.6 - Mathematical Functions</A>
-<LI><A HREF="manual.html#5.7">5.7 - Input and Output Facilities</A>
-<LI><A HREF="manual.html#5.8">5.8 - Operating System Facilities</A>
-<LI><A HREF="manual.html#5.9">5.9 - The Debug Library</A>
+<LI><A HREF="manual.html#5.5">5.5 &ndash; Table Manipulation</A>
+<LI><A HREF="manual.html#5.6">5.6 &ndash; Mathematical Functions</A>
+<LI><A HREF="manual.html#5.7">5.7 &ndash; Input and Output Facilities</A>
+<LI><A HREF="manual.html#5.8">5.8 &ndash; Operating System Facilities</A>
+<LI><A HREF="manual.html#5.9">5.9 &ndash; The Debug Library</A>
 </UL>
 <P>
-<LI><A HREF="manual.html#6">6 - Lua Stand-alone</A>
+<LI><A HREF="manual.html#6">6 &ndash; Lua Stand-alone</A>
 <P>
-<LI><A HREF="manual.html#7">7 - Incompatibilities with the Previous Version</A>
+<LI><A HREF="manual.html#7">7 &ndash; Incompatibilities with the Previous Version</A>
 <UL>
-<LI><A HREF="manual.html#7.1">7.1 - Changes in the Language</A>
-<LI><A HREF="manual.html#7.2">7.2 - Changes in the Libraries</A>
-<LI><A HREF="manual.html#7.3">7.3 - Changes in the API</A>
+<LI><A HREF="manual.html#7.1">7.1 &ndash; Changes in the Language</A>
+<LI><A HREF="manual.html#7.2">7.2 &ndash; Changes in the Libraries</A>
+<LI><A HREF="manual.html#7.3">7.3 &ndash; Changes in the API</A>
 </UL>
 <P>
-<LI><A HREF="manual.html#8">8 - The Complete Syntax of Lua</A>
+<LI><A HREF="manual.html#8">8 &ndash; The Complete Syntax of Lua</A>
 </UL>

 <H2><A NAME="index">Index</A></H2>
@@ -160,6 +156,8 @@ Freely available under the terms of the
 <H3><A NAME="functions">Lua functions</A></H3>
 <A HREF="manual.html#pdf-_G">_G</A><BR>
 <A HREF="manual.html#pdf-_VERSION">_VERSION</A><BR>
+<P>
+
 <A HREF="manual.html#pdf-assert">assert</A><BR>
 <A HREF="manual.html#pdf-collectgarbage">collectgarbage</A><BR>
 <A HREF="manual.html#pdf-dofile">dofile</A><BR>
@@ -487,12 +485,12 @@ Freely available under the terms of the
 <P>

 <HR>
-<SMALL>
+<SMALL CLASS="footer">
 Last update:
-Sat Jan 19 13:24:29 BRST 2008
+Mon Feb 13 18:53:32 BRST 2012
 </SMALL>
 <!--
-Last change: revised for Lua 5.1.3
+Last change: revised for Lua 5.1.5
 -->

 </BODY>
@@ -1,17 +1,37 @@
 body {
 	color: #000000 ;
 	background-color: #FFFFFF ;
-	font-family: sans-serif ;
+	font-family: Helvetica, Arial, sans-serif ;
 	text-align: justify ;
-	margin-right: 20px ;
-	margin-left: 20px ;
+	margin-right: 30px ;
+	margin-left: 30px ;
 }

 h1, h2, h3, h4 {
+	font-family: Verdana, Geneva, sans-serif ;
 	font-weight: normal ;
 	font-style: italic ;
 }

+h2 {
+	padding-top: 0.4em ;
+	padding-bottom: 0.4em ;
+	padding-left: 30px ;
+	padding-right: 30px ;
+	margin-left: -30px ;
+	background-color: #E0E0FF ;
+}
+
+h3 {
+	padding-left: 0.5em ;
+	border-left: solid #E0E0FF 1em ;
+}
+
+table h3 {
+	padding-left: 0px ;
+	border-left: none ;
+}
+
 a:link {
 	color: #000080 ;
 	background-color: inherit ;
@@ -39,3 +59,25 @@ hr {
 	background-color: #a0a0a0 ;
 }

+:target {
+	background-color: #F8F8F8 ;
+	padding: 8px ;
+	border: solid #a0a0a0 2px ;
+}
+
+.footer {
+	color: gray ;
+	font-size: small ;
+}
+
+input[type=text] {
+	border: solid #a0a0a0 2px ;
+	border-radius: 2em ;
+	-moz-border-radius: 2em ;
+	background-image: url('images/search.png') ;
+	background-repeat: no-repeat;
+	background-position: 4px center ;
+	padding-left: 20px ;
+	height: 2em ;
+}
+
@@ -1,13 +1,24 @@
 h3 code {
 	font-family: inherit ;
+	font-size: inherit ;
 }

-pre {
-	font-size: 105% ;
+pre, code {
+	font-size: 12pt ;
 }

 span.apii {
 	float: right ;
 	font-family: inherit ;
+	font-style: normal ;
+	font-size: small ;
+	color: gray ;
 }

+p+h1, ul+h1 {
+	padding-top: 0.4em ;
+	padding-bottom: 0.4em ;
+	padding-left: 30px ;
+	margin-left: -30px ;
+	background-color: #E0E0FF ;
+}
@@ -19,9 +19,9 @@ Lua 5.1 Reference Manual
 by Roberto Ierusalimschy, Luiz Henrique de Figueiredo, Waldemar Celes
 <p>
 <small>
-Copyright &copy; 2006-2008 Lua.org, PUC-Rio.
+Copyright &copy; 2006&ndash;2012 Lua.org, PUC-Rio.
 Freely available under the terms of the
-<a href="http://www.lua.org/license.html#5">Lua license</a>.
+<a href="http://www.lua.org/license.html">Lua license</a>.
 </small>
 <hr>
 <p>
@@ -29,11 +29,13 @@ Freely available under the terms of the
 <a href="contents.html#contents">contents</A>
 &middot;
 <a href="contents.html#index">index</A>
+&middot;
+<A HREF="http://www.lua.org/manual/">other versions</A>

 <!-- ====================================================================== -->
 <p>

-<!-- $Id: manual.of,v 1.48 2008/08/18 15:24:20 roberto Exp $ -->
+<!-- $Id: manual.of,v 1.49.1.2 2012/01/13 20:23:26 roberto Exp $ -->



@@ -5645,7 +5647,7 @@ Currently, Lua has the following standard libraries:

 <ul>

-<li>basic library,</li> which includes the coroutine sub-library;
+<li>basic library, which includes the coroutine sub-library;</li>

 <li>package library;</li>

@@ -5709,7 +5711,7 @@ when absent, it defaults to "assertion failed!"


 <p>
-<hr><h3><a name="pdf-collectgarbage"><code>collectgarbage (opt [, arg])</code></a></h3>
+<hr><h3><a name="pdf-collectgarbage"><code>collectgarbage ([opt [, arg]])</code></a></h3>


 <p>
@@ -5718,6 +5720,11 @@ It performs different functions according to its first argument, <code>opt</code

 <ul>

+<li><b>"collect":</b>
+performs a full garbage-collection cycle.
+This is the default option.
+</li>
+
 <li><b>"stop":</b>
 stops the garbage collector.
 </li>
@@ -5726,10 +5733,6 @@ stops the garbage collector.
 restarts the garbage collector.
 </li>

-<li><b>"collect":</b>
-performs a full garbage-collection cycle.
-</li>
-
 <li><b>"count":</b>
 returns the total memory in use by Lua (in Kbytes).
 </li>
@@ -5760,7 +5763,7 @@ Returns the previous value for <em>step</em>.


 <p>
-<hr><h3><a name="pdf-dofile"><code>dofile (filename)</code></a></h3>
+<hr><h3><a name="pdf-dofile"><code>dofile ([filename])</code></a></h3>
 Opens the named file and executes its contents as a Lua chunk.
 When called without arguments,
 <code>dofile</code> executes the contents of the standard input (<code>stdin</code>).
@@ -8421,7 +8424,7 @@ Otherwise, it returns the name of the upvalue.


 <p>
-<hr><h3><a name="pdf-debug.traceback"><code>debug.traceback ([thread,] [message] [, level])</code></a></h3>
+<hr><h3><a name="pdf-debug.traceback"><code>debug.traceback ([thread,] [message [, level]])</code></a></h3>


 <p>
@@ -8789,12 +8792,12 @@ Here is the complete syntax of Lua in extended BNF.


 <HR>
-<SMALL>
+<SMALL CLASS="footer">
 Last update:
-Mon Aug 18 13:25:46 BRT 2008
+Mon Feb 13 18:54:19 BRST 2012
 </SMALL>
 <!--
-Last change: revised for Lua 5.1.4
+Last change: revised for Lua 5.1.5
 -->

 </body></html>
@@ -12,7 +12,7 @@
 Documentation
 </H1>

-This is the documentation included in the source distribution of Lua 5.1.4.
+This is the documentation included in the source distribution of Lua 5.1.5.

 <UL>
 <LI><A HREF="contents.html">Reference manual</A>
@@ -33,7 +33,7 @@ especially the
 <HR>
 <SMALL>
 Last update:
-Tue Aug 12 14:46:07 BRT 2008
+Fri Feb  3 09:44:42 BRST 2012
 </SMALL>

 </BODY>
@@ -5,7 +5,7 @@
 # grep '^V=' ../Makefile
 V= 5.1
 # grep '^R=' ../Makefile
-R= 5.1.4
+R= 5.1.5

 # grep '^INSTALL_.*=' ../Makefile | sed 's/INSTALL_TOP/prefix/'
 prefix= /usr/local
@@ -48,7 +48,7 @@ o:	$(ALL_O)
 a:	$(ALL_A)

 $(LUA_A): $(CORE_O) $(LIB_O)
-	$(AR) $@ $?
+	$(AR) $@ $(CORE_O) $(LIB_O)	# DLL needs all object files
 	$(RANLIB) $@

 $(LUA_T): $(LUA_O) $(LUA_A)
@@ -631,7 +631,7 @@ static void base_open (lua_State *L) {
  luaL_register(L, "_G", base_funcs);
  lua_pushliteral(L, LUA_VERSION);
  lua_setglobal(L, "_VERSION");  /* set global _VERSION */
-  /* `ipairs' and `pairs' need auxliliary functions as upvalues */
+  /* `ipairs' and `pairs' need auxiliary functions as upvalues */
  auxopen(L, "ipairs", luaB_ipairs, ipairsaux);
  auxopen(L, "pairs", luaB_pairs, luaB_next);
  /* `newproxy' needs a weaktable as upvalue */
@@ -1,5 +1,5 @@
 /*
-** $Id: lcode.c,v 2.25.1.3 2007/12/28 15:32:23 roberto Exp $
+** $Id: lcode.c,v 2.25.1.5 2011/01/31 14:53:16 roberto Exp $
 ** Code generator for Lua
 ** See Copyright Notice in lua.h
 */
@@ -544,10 +544,6 @@ void luaK_goiftrue (FuncState *fs, expdesc *e) {
      pc = NO_JUMP;  /* always true; do nothing */
      break;
    }
-    case VFALSE: {
-      pc = luaK_jump(fs);  /* always jump */
-      break;
-    }
    case VJMP: {
      invertjump(fs, e);
      pc = e->u.s.info;
@@ -572,10 +568,6 @@ static void luaK_goiffalse (FuncState *fs, expdesc *e) {
      pc = NO_JUMP;  /* always false; do nothing */
      break;
    }
-    case VTRUE: {
-      pc = luaK_jump(fs);  /* always jump */
-      break;
-    }
    case VJMP: {
      pc = e->u.s.info;
      break;
@@ -1,5 +1,5 @@
 /*
-** $Id: ldblib.c,v 1.104.1.3 2008/01/21 13:11:21 roberto Exp $
+** $Id: ldblib.c,v 1.104.1.4 2009/08/04 18:50:18 roberto Exp $
 ** Interface from Lua to its debug API
 ** See Copyright Notice in lua.h
 */
@@ -45,6 +45,7 @@ static int db_setmetatable (lua_State *L) {


 static int db_getfenv (lua_State *L) {
+  luaL_checkany(L, 1);
  lua_getfenv(L, 1);
  return 1;
 }
@@ -1,5 +1,5 @@
 /*
-** $Id: ldo.c,v 2.38.1.3 2008/01/18 22:31:22 roberto Exp $
+** $Id: ldo.c,v 2.38.1.4 2012/01/18 02:27:10 roberto Exp $
 ** Stack and Call structure of Lua
 ** See Copyright Notice in lua.h
 */
@@ -217,6 +217,7 @@ static StkId adjust_varargs (lua_State *L, Proto *p, int actual) {
    int nvar = actual - nfixargs;  /* number of extra arguments */
    lua_assert(p->is_vararg & VARARG_HASARG);
    luaC_checkGC(L);
+    luaD_checkstack(L, p->maxstacksize);
    htab = luaH_new(L, nvar, 1);  /* create `arg' table */
    for (i=0; i<nvar; i++)  /* put extra arguments into `arg' table */
      setobj2n(L, luaH_setnum(L, htab, i+1), L->top - nvar + i);
@@ -1,5 +1,5 @@
 /*
-** $Id: lgc.c,v 2.38.1.1 2007/12/27 13:02:25 roberto Exp $
+** $Id: lgc.c,v 2.38.1.2 2011/03/18 18:05:38 roberto Exp $
 ** Garbage Collector
 ** See Copyright Notice in lua.h
 */
@@ -627,7 +627,6 @@ void luaC_step (lua_State *L) {
    }
  }
  else {
-    lua_assert(g->totalbytes >= g->estimate);
    setthreshold(g);
  }
 }
@@ -1,5 +1,5 @@
 /*
-** $Id: liolib.c,v 2.73.1.3 2008/01/18 17:47:43 roberto Exp $
+** $Id: liolib.c,v 2.73.1.4 2010/05/14 15:33:51 roberto Exp $
 ** Standard I/O (and system) library
 ** See Copyright Notice in lua.h
 */
@@ -276,7 +276,10 @@ static int read_number (lua_State *L, FILE *f) {
    lua_pushnumber(L, d);
    return 1;
  }
-  else return 0;  /* read fails */
+  else {
+    lua_pushnil(L);  /* "result" to be removed */
+    return 0;  /* read fails */
+  }
 }


@@ -1,5 +1,5 @@
 /*
-** $Id: llex.c,v 2.20.1.1 2007/12/27 13:02:25 roberto Exp $
+** $Id: llex.c,v 2.20.1.2 2009/11/23 14:58:22 roberto Exp $
 ** Lexical Analyzer
 ** See Copyright Notice in lua.h
 */
@@ -118,8 +118,10 @@ TString *luaX_newstring (LexState *ls, const char *str, size_t l) {
  lua_State *L = ls->L;
  TString *ts = luaS_newlstr(L, str, l);
  TValue *o = luaH_setstr(L, ls->fs->h, ts);  /* entry for `str' */
-  if (ttisnil(o))
+  if (ttisnil(o)) {
    setbvalue(o, 1);  /* make sure `str' will not be collected */
+    luaC_checkGC(L);
+  }
  return ts;
 }

@@ -1,5 +1,5 @@
 /*
-** $Id: loadlib.c,v 1.52.1.3 2008/08/06 13:29:28 roberto Exp $
+** $Id: loadlib.c,v 1.52.1.4 2009/09/09 13:17:16 roberto Exp $
 ** Dynamic library loader for Lua
 ** See Copyright Notice in lua.h
 **
@@ -639,7 +639,7 @@ LUALIB_API int luaopen_package (lua_State *L) {
  lua_pushvalue(L, -1);
  lua_replace(L, LUA_ENVIRONINDEX);
  /* create `loaders' table */
-  lua_createtable(L, 0, sizeof(loaders)/sizeof(loaders[0]) - 1);
+  lua_createtable(L, sizeof(loaders)/sizeof(loaders[0]) - 1, 0);
  /* fill it with pre-defined loaders */
  for (i=0; loaders[i] != NULL; i++) {
    lua_pushcfunction(L, loaders[i]);
@@ -1,5 +1,5 @@
 /*
-** $Id: lparser.c,v 2.42.1.3 2007/12/28 15:32:23 roberto Exp $
+** $Id: lparser.c,v 2.42.1.4 2011/10/21 19:31:42 roberto Exp $
 ** Lua Parser
 ** See Copyright Notice in lua.h
 */
@@ -374,9 +374,9 @@ static void close_func (LexState *ls) {
  lua_assert(luaG_checkcode(f));
  lua_assert(fs->bl == NULL);
  ls->fs = fs->prev;
-  L->top -= 2;  /* remove table and prototype from the stack */
  /* last token read was anchored in defunct function; must reanchor it */
  if (fs) anchor_token(ls);
+  L->top -= 2;  /* remove table and prototype from the stack */
 }


@@ -1,5 +1,5 @@
 /*
-** $Id: lstrlib.c,v 1.132.1.4 2008/07/11 17:27:21 roberto Exp $
+** $Id: lstrlib.c,v 1.132.1.5 2010/05/14 15:34:19 roberto Exp $
 ** Standard library for string operations and pattern-matching
 ** See Copyright Notice in lua.h
 */
@@ -754,6 +754,7 @@ static void addintlen (char *form) {


 static int str_format (lua_State *L) {
+  int top = lua_gettop(L);
  int arg = 1;
  size_t sfl;
  const char *strfrmt = luaL_checklstring(L, arg, &sfl);
@@ -768,7 +769,8 @@ static int str_format (lua_State *L) {
    else { /* format item */
      char form[MAX_FORMAT];  /* to store the format (`%...') */
      char buff[MAX_ITEM];  /* to store the formatted item */
-      arg++;
+      if (++arg > top)
+        luaL_argerror(L, arg, "no value");
      strfrmt = scanformat(L, strfrmt, form);
      switch (*strfrmt++) {
        case 'c': {
@@ -1,5 +1,5 @@
 /*
-** $Id: lua.h,v 1.218.1.5 2008/08/06 13:30:12 roberto Exp $
+** $Id: lua.h,v 1.218.1.7 2012/01/13 20:36:20 roberto Exp $
 ** Lua - An Extensible Extension Language
 ** Lua.org, PUC-Rio, Brazil (http://www.lua.org)
 ** See Copyright Notice at the end of this file
@@ -17,9 +17,9 @@


 #define LUA_VERSION	"Lua 5.1"
-#define LUA_RELEASE	"Lua 5.1.4"
+#define LUA_RELEASE	"Lua 5.1.5"
 #define LUA_VERSION_NUM	501
-#define LUA_COPYRIGHT	"Copyright (C) 1994-2008 Lua.org, PUC-Rio"
+#define LUA_COPYRIGHT	"Copyright (C) 1994-2012 Lua.org, PUC-Rio"
 #define LUA_AUTHORS 	"R. Ierusalimschy, L. H. de Figueiredo & W. Celes"


@@ -362,7 +362,7 @@ struct lua_Debug {


 /******************************************************************************
-* Copyright (C) 1994-2008 Lua.org, PUC-Rio.  All rights reserved.
+* Copyright (C) 1994-2012 Lua.org, PUC-Rio.  All rights reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining
 * a copy of this software and associated documentation files (the
@@ -1,5 +1,5 @@
 /*
-** $Id: lvm.c,v 2.63.1.3 2007/12/28 15:32:23 roberto Exp $
+** $Id: lvm.c,v 2.63.1.5 2011/08/17 20:43:11 roberto Exp $
 ** Lua virtual machine
 ** See Copyright Notice in lua.h
 */
@@ -133,6 +133,7 @@ void luaV_gettable (lua_State *L, const TValue *t, TValue *key, StkId val) {

 void luaV_settable (lua_State *L, const TValue *t, TValue *key, StkId val) {
  int loop;
+  TValue temp;
  for (loop = 0; loop < MAXTAGLOOP; loop++) {
    const TValue *tm;
    if (ttistable(t)) {  /* `t' is a table? */
@@ -141,6 +142,7 @@ void luaV_settable (lua_State *L, const TValue *t, TValue *key, StkId val) {
      if (!ttisnil(oldval) ||  /* result is no nil? */
          (tm = fasttm(L, h->metatable, TM_NEWINDEX)) == NULL) { /* or no TM? */
        setobj2t(L, oldval, val);
+        h->flags = 0;
        luaC_barriert(L, h, val);
        return;
      }
@@ -152,7 +154,9 @@ void luaV_settable (lua_State *L, const TValue *t, TValue *key, StkId val) {
      callTM(L, tm, t, key, val);
      return;
    }
-    t = tm;  /* else repeat with `tm' */ 
+    /* else repeat with `tm' */
+    setobj(L, &temp, tm);  /* avoid pointing inside table (may rehash) */
+    t = &temp;
  }
  luaG_runerror(L, "loop in settable");
 }
@@ -580,6 +580,12 @@ client-output-buffer-limit pubsub 32mb 8mb 60
 # 100 only in environments where very low latency is required.
 hz 10

+# When a child rewrites the AOF file, if the following option is enabled
+# the file will be fsync-ed every 32 MB of data generated. This is useful
+# in order to commit the file to the disk more incrementally and avoid
+# big latency spikes.
+aof-rewrite-incremental-fsync yes
+
 ################################## INCLUDES ###################################

 # Include one or more other config files here.  This is useful if you
@@ -855,6 +855,8 @@ int rewriteAppendOnlyFile(char *filename) {
    }

    rioInitWithFile(&aof,fp);
+    if (server.aof_rewrite_incremental_fsync)
+        rioSetAutoSync(&aof,REDIS_AOF_AUTOSYNC_BYTES);
    for (j = 0; j < server.dbnum; j++) {
        char selectcmd[] = "*2\r\n$6\r\nSELECT\r\n";
        redisDb *db = server.db+j;
@@ -882,6 +884,9 @@ int rewriteAppendOnlyFile(char *filename) {

            expiretime = getExpire(db,&key);

+            /* If this key is already expired skip it */
+            if (expiretime != -1 && expiretime < now) continue;
+
            /* Save the key and associated value */
            if (o->type == REDIS_STRING) {
                /* Emit a SET command */
@@ -904,8 +909,6 @@ int rewriteAppendOnlyFile(char *filename) {
            /* Save the expire time */
            if (expiretime != -1) {
                char cmd[]="*3\r\n$9\r\nPEXPIREAT\r\n";
-                /* If this key is already expired skip it */
-                if (expiretime < now) continue;
                if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr;
                if (rioWriteBulkObject(&aof,&key) == 0) goto werr;
                if (rioWriteBulkLongLong(&aof,expiretime) == 0) goto werr;
@@ -58,8 +58,8 @@ static int getBitOffsetFromArgument(redisClient *c, robj *o, size_t *offset) {
 /* Count number of bits set in the binary array pointed by 's' and long
 * 'count' bytes. The implementation of this function is required to
 * work with a input string length up to 512 MB. */
-long popcount(void *s, long count) {
-    long bits = 0;
+size_t popcount(void *s, long count) {
+    size_t bits = 0;
    unsigned char *p;
    uint32_t *p4 = s;
    static const unsigned char bitsinbyte[256] = {0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4,1,2,2,3,2,3,3,4,2,3,3,4,3,4,4,5,1,2,2,3,2,3,3,4,2,3,3,4,3,4,4,5,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,1,2,2,3,2,3,3,4,2,3,3,4,3,4,4,5,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,3,4,4,5,4,5,5,6,4,5,5,6,5,6,6,7,1,2,2,3,2,3,3,4,2,3,3,4,3,4,4,5,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,3,4,4,5,4,5,5,6,4,5,5,6,5,6,6,7,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,3,4,4,5,4,5,5,6,4,5,5,6,5,6,6,7,3,4,4,5,4,5,5,6,4,5,5,6,5,6,6,7,4,5,5,6,5,6,6,7,5,6,6,7,6,7,7,8};
@@ -317,6 +317,12 @@ void loadServerConfigFromString(char *config) {
                   argc == 2)
        {
            server.aof_rewrite_min_size = memtoll(argv[1],NULL);
+        } else if (!strcasecmp(argv[0],"aof-rewrite-incremental-fsync") &&
+                   argc == 2)
+        {
+            if ((server.aof_rewrite_incremental_fsync = yesnotoi(argv[1])) == -1) {
+                err = "argument must be 'yes' or 'no'"; goto loaderr;
+            }
        } else if (!strcasecmp(argv[0],"requirepass") && argc == 2) {
            if (strlen(argv[1]) > REDIS_AUTHPASS_MAX_LEN) {
                err = "Password is longer than REDIS_AUTHPASS_MAX_LEN";
@@ -487,7 +493,7 @@ void configSetCommand(redisClient *c) {
        server.requirepass = ((char*)o->ptr)[0] ? zstrdup(o->ptr) : NULL;
    } else if (!strcasecmp(c->argv[2]->ptr,"masterauth")) {
        zfree(server.masterauth);
-        server.masterauth = zstrdup(o->ptr);
+        server.masterauth = ((char*)o->ptr)[0] ? zstrdup(o->ptr) : NULL;
    } else if (!strcasecmp(c->argv[2]->ptr,"maxmemory")) {
        if (getLongLongFromObject(o,&ll) == REDIS_ERR ||
            ll < 0) goto badfmt;
@@ -566,6 +572,11 @@ void configSetCommand(redisClient *c) {
    } else if (!strcasecmp(c->argv[2]->ptr,"auto-aof-rewrite-min-size")) {
        if (getLongLongFromObject(o,&ll) == REDIS_ERR || ll < 0) goto badfmt;
        server.aof_rewrite_min_size = ll;
+    } else if (!strcasecmp(c->argv[2]->ptr,"aof-rewrite-incremental-fsync")) {
+        int yn = yesnotoi(o->ptr);
+
+        if (yn == -1) goto badfmt;
+        server.aof_rewrite_incremental_fsync = yn;
    } else if (!strcasecmp(c->argv[2]->ptr,"save")) {
        int vlen, j;
        sds *v = sdssplitlen(o->ptr,sdslen(o->ptr)," ",1,&vlen);
@@ -789,7 +800,7 @@ void configGetCommand(redisClient *c) {
    /* String values */
    config_get_string_field("dbfilename",server.rdb_filename);
    config_get_string_field("requirepass",server.requirepass);
-    config_get_string_field("masterauth",server.requirepass);
+    config_get_string_field("masterauth",server.masterauth);
    config_get_string_field("bind",server.bindaddr);
    config_get_string_field("unixsocket",server.unixsocket);
    config_get_string_field("logfile",server.logfile);
@@ -847,6 +858,8 @@ void configGetCommand(redisClient *c) {
    config_get_bool_field("activerehashing", server.activerehashing);
    config_get_bool_field("repl-disable-tcp-nodelay",
            server.repl_disable_tcp_nodelay);
+    config_get_bool_field("aof-rewrite-incremental-fsync",
+            server.aof_rewrite_incremental_fsync);

    /* Everything we can't handle with macros follows. */

@@ -540,7 +540,6 @@ int expireIfNeeded(redisDb *db, robj *key) {
 * unit is either UNIT_SECONDS or UNIT_MILLISECONDS, and is only used for
 * the argv[2] parameter. The basetime is always specified in milliseconds. */
 void expireGenericCommand(redisClient *c, long long basetime, int unit) {
-    dictEntry *de;
    robj *key = c->argv[1], *param = c->argv[2];
    long long when; /* unix time in milliseconds when the key will expire. */

@@ -550,11 +549,12 @@ void expireGenericCommand(redisClient *c, long long basetime, int unit) {
    if (unit == UNIT_SECONDS) when *= 1000;
    when += basetime;

-    de = dictFind(c->db->dict,key->ptr);
-    if (de == NULL) {
+    /* No key, return zero. */
+    if (lookupKeyRead(c->db,key) == NULL) {
        addReply(c,shared.czero);
        return;
    }
+
    /* EXPIRE with negative TTL, or EXPIREAT with a timestamp into the past
     * should never be executed as a DEL when load the AOF or in the context
     * of a slave instance.
@@ -330,9 +330,14 @@ void debugCommand(redisClient *c) {

        usleep(utime);
        addReply(c,shared.ok);
+    } else if (!strcasecmp(c->argv[1]->ptr,"set-active-expire") &&
+               c->argc == 3)
+    {
+        server.active_expire_enabled = atoi(c->argv[2]->ptr);
+        addReply(c,shared.ok);
    } else {
-        addReplyError(c,
-            "Syntax error, try DEBUG [SEGFAULT|OBJECT <key>|SWAPIN <key>|SWAPOUT <key>|RELOAD]");
+        addReplyErrorFormat(c, "Unknown DEBUG subcommand or wrong number of arguments for '%s'",
+            (char*)c->argv[1]->ptr);
    }
 }

@@ -382,9 +387,12 @@ void redisLogObjectDebugInfo(robj *o) {
    redisLog(REDIS_WARNING,"Object encoding: %d", o->encoding);
    redisLog(REDIS_WARNING,"Object refcount: %d", o->refcount);
    if (o->type == REDIS_STRING && o->encoding == REDIS_ENCODING_RAW) {
-        redisLog(REDIS_WARNING,"Object raw string len: %d", sdslen(o->ptr));
-        if (sdslen(o->ptr) < 4096)
-            redisLog(REDIS_WARNING,"Object raw string content: \"%s\"", (char*)o->ptr);
+        redisLog(REDIS_WARNING,"Object raw string len: %zu", sdslen(o->ptr));
+        if (sdslen(o->ptr) < 4096) {
+            sds repr = sdscatrepr(sdsempty(),o->ptr,sdslen(o->ptr));
+            redisLog(REDIS_WARNING,"Object raw string content: %s", repr);
+            sdsfree(repr);
+        }
    } else if (o->type == REDIS_LIST) {
        redisLog(REDIS_WARNING,"List length: %d", (int) listTypeLength(o));
    } else if (o->type == REDIS_SET) {
@@ -69,6 +69,11 @@ struct commandHelp {
    "Pop a value from a list, push it to another list and return it; or block until one is available",
    2,
    "2.2.0" },
+    { "CLIENT GETNAME",
+    "-",
+    "Get the current connection name",
+    9,
+    "2.6.9" },
    { "CLIENT KILL",
    "ip:port",
    "Kill the connection of a client",
@@ -79,6 +84,11 @@ struct commandHelp {
    "Get the list of client connections",
    9,
    "2.4.0" },
+    { "CLIENT SETNAME",
+    "connection-name",
+    "Set the current connection name",
+    9,
+    "2.6.9" },
    { "CONFIG GET",
    "parameter",
    "Get the value of a configuration parameter",
@@ -280,7 +290,7 @@ struct commandHelp {
    1,
    "2.6.0" },
    { "INFO",
-    "-",
+    "[section]",
    "Get information and statistics about the server",
    9,
    "1.0.0" },
@@ -525,7 +535,7 @@ struct commandHelp {
    8,
    "1.0.0" },
    { "SET",
-    "key value",
+    "key value [EX seconds] [PX milliseconds] [NX|XX]",
    "Set the string value of a key",
    1,
    "1.0.0" },
@@ -665,7 +675,7 @@ struct commandHelp {
    7,
    "2.2.0" },
    { "ZADD",
-    "key score member [score] [member]",
+    "key score member [score member ...]",
    "Add one or more members to a sorted set, or update its score if it already exists",
    4,
    "1.2.0" },
@@ -103,13 +103,11 @@ void discardCommand(redisClient *c) {

 /* Send a MULTI command to all the slaves and AOF file. Check the execCommand
 * implementation for more information. */
-void execCommandReplicateMulti(redisClient *c) {
+void execCommandPropagateMulti(redisClient *c) {
    robj *multistring = createStringObject("MULTI",5);

-    if (server.aof_state != REDIS_AOF_OFF)
-        feedAppendOnlyFile(server.multiCommand,c->db->id,&multistring,1);
-    if (listLength(server.slaves))
-        replicationFeedSlaves(server.slaves,c->db->id,&multistring,1);
+    propagate(server.multiCommand,c->db->id,&multistring,1,
+              REDIS_PROPAGATE_AOF|REDIS_PROPAGATE_REPL);
    decrRefCount(multistring);
 }

@@ -118,6 +116,7 @@ void execCommand(redisClient *c) {
    robj **orig_argv;
    int orig_argc;
    struct redisCommand *orig_cmd;
+    int must_propagate = 0; /* Need to propagate MULTI/EXEC to AOF / slaves? */

    if (!(c->flags & REDIS_MULTI)) {
        addReplyError(c,"EXEC without MULTI");
@@ -133,19 +132,10 @@ void execCommand(redisClient *c) {
    if (c->flags & (REDIS_DIRTY_CAS|REDIS_DIRTY_EXEC)) {
        addReply(c, c->flags & REDIS_DIRTY_EXEC ? shared.execaborterr :
                                                  shared.nullmultibulk);
-        freeClientMultiState(c);
-        initClientMultiState(c);
-        c->flags &= ~(REDIS_MULTI|REDIS_DIRTY_CAS|REDIS_DIRTY_EXEC);
-        unwatchAllKeys(c);
+        discardTransaction(c);
        goto handle_monitor;
    }

-    /* Replicate a MULTI request now that we are sure the block is executed.
-     * This way we'll deliver the MULTI/..../EXEC block as a whole and
-     * both the AOF and the replication link will have the same consistency
-     * and atomicity guarantees. */
-    execCommandReplicateMulti(c);
-
    /* Exec all the queued commands */
    unwatchAllKeys(c); /* Unwatch ASAP otherwise we'll waste CPU cycles */
    orig_argv = c->argv;
@@ -156,6 +146,16 @@ void execCommand(redisClient *c) {
        c->argc = c->mstate.commands[j].argc;
        c->argv = c->mstate.commands[j].argv;
        c->cmd = c->mstate.commands[j].cmd;
+
+        /* Propagate a MULTI request once we encounter the first write op.
+         * This way we'll deliver the MULTI/..../EXEC block as a whole and
+         * both the AOF and the replication link will have the same consistency
+         * and atomicity guarantees. */
+        if (!must_propagate && !(c->cmd->flags & REDIS_CMD_READONLY)) {
+            execCommandPropagateMulti(c);
+            must_propagate = 1;
+        }
+
        call(c,REDIS_CALL_FULL);

        /* Commands may alter argc/argv, restore mstate. */
@@ -166,13 +166,10 @@ void execCommand(redisClient *c) {
    c->argv = orig_argv;
    c->argc = orig_argc;
    c->cmd = orig_cmd;
-    freeClientMultiState(c);
-    initClientMultiState(c);
-    c->flags &= ~(REDIS_MULTI|REDIS_DIRTY_CAS|REDIS_DIRTY_EXEC);
-    /* Make sure the EXEC command is always replicated / AOF, since we
-     * always send the MULTI command (we can't know beforehand if the
-     * next operations will contain at least a modification to the DB). */
-    server.dirty++;
+    discardTransaction(c);
+    /* Make sure the EXEC command will be propagated as well if MULTI
+     * was already propagated. */
+    if (must_propagate) server.dirty++;

 handle_monitor:
    /* Send EXEC to clients waiting data from MONITOR. We do it here
@@ -722,6 +722,7 @@ int rdbSaveBackground(char *filename) {
    if (server.rdb_child_pid != -1) return REDIS_ERR;

    server.dirty_before_bgsave = server.dirty;
+    server.lastbgsave_try = time(NULL);

    start = ustime();
    if ((childpid = fork()) == 0) {
@@ -1067,11 +1068,8 @@ int rdbLoad(char *filename) {
    FILE *fp;
    rio rdb;

-    fp = fopen(filename,"r");
-    if (!fp) {
-        errno = ENOENT;
-        return REDIS_ERR;
-    }
+    if ((fp = fopen(filename,"r")) == NULL) return REDIS_ERR;
+
    rioInitWithFile(&rdb,fp);
    if (server.rdb_checksum)
        rdb.update_cksum = rioGenericUpdateChecksum;
@@ -42,6 +42,7 @@
 #include <sys/time.h>
 #include <assert.h>
 #include <fcntl.h>
+#include <limits.h>

 #include "hiredis.h"
 #include "sds.h"
@@ -71,11 +72,13 @@ static struct config {
    int monitor_mode;
    int pubsub_mode;
    int latency_mode;
+    int latency_history;
    int cluster_mode;
    int cluster_reissue_command;
    int slave_mode;
    int pipe_mode;
    int getrdb_mode;
+    int stat_mode;
    char *rdb_filename;
    int bigkeys;
    int stdinarg; /* get last arg from stdin. (-x option) */
@@ -630,6 +633,36 @@ static int cliSendCommand(int argc, char **argv, int repeat) {
    return REDIS_OK;
 }

+/* Send the INFO command, reconnecting the link if needed. */
+static redisReply *reconnectingInfo(void) {
+    redisContext *c = context;
+    redisReply *reply = NULL;
+    int tries = 0;
+
+    assert(!c->err);
+    while(reply == NULL) {
+        while (c->err & (REDIS_ERR_IO | REDIS_ERR_EOF)) {
+            printf("Reconnecting (%d)...\r", ++tries);
+            fflush(stdout);
+
+            redisFree(c);
+            c = redisConnect(config.hostip,config.hostport);
+            usleep(1000000);
+        }
+
+        reply = redisCommand(c,"INFO");
+        if (c->err && !(c->err & (REDIS_ERR_IO | REDIS_ERR_EOF))) {
+            fprintf(stderr, "Error: %s\n", c->errstr);
+            exit(1);
+        } else if (tries > 0) {
+            printf("\n");
+        }
+    }
+
+    context = c;
+    return reply;
+}
+
 /*------------------------------------------------------------------------------
 * User interface
 *--------------------------------------------------------------------------- */
@@ -668,8 +701,13 @@ static int parseOptions(int argc, char **argv) {
            config.output = OUTPUT_CSV;
        } else if (!strcmp(argv[i],"--latency")) {
            config.latency_mode = 1;
+        } else if (!strcmp(argv[i],"--latency-history")) {
+            config.latency_mode = 1;
+            config.latency_history = 1;
        } else if (!strcmp(argv[i],"--slave")) {
            config.slave_mode = 1;
+        } else if (!strcmp(argv[i],"--stat")) {
+            config.stat_mode = 1;
        } else if (!strcmp(argv[i],"--rdb") && !lastarg) {
            config.getrdb_mode = 1;
            config.rdb_filename = argv[++i];
@@ -690,7 +728,15 @@ static int parseOptions(int argc, char **argv) {
            sdsfree(version);
            exit(0);
        } else {
-            break;
+            if (argv[i][0] == '-') {
+                fprintf(stderr,
+                    "Unrecognized option or bad number of args for: '%s'\n",
+                    argv[i]);
+                exit(1);
+            } else {
+                /* Likely the command name, stop here. */
+                break;
+            }
        }
    }
    return i;
@@ -719,26 +765,29 @@ static void usage() {
 "redis-cli %s\n"
 "\n"
 "Usage: redis-cli [OPTIONS] [cmd [arg [arg ...]]]\n"
-"  -h <hostname>    Server hostname (default: 127.0.0.1)\n"
-"  -p <port>        Server port (default: 6379)\n"
-"  -s <socket>      Server socket (overrides hostname and port)\n"
-"  -a <password>    Password to use when connecting to the server\n"
-"  -r <repeat>      Execute specified command N times\n"
-"  -i <interval>    When -r is used, waits <interval> seconds per command.\n"
-"                   It is possible to specify sub-second times like -i 0.1\n"
-"  -n <db>          Database number\n"
-"  -x               Read last argument from STDIN\n"
-"  -d <delimiter>   Multi-bulk delimiter in for raw formatting (default: \\n)\n"
-"  -c               Enable cluster mode (follow -ASK and -MOVED redirections)\n"
-"  --raw            Use raw formatting for replies (default when STDOUT is not a tty)\n"
-"  --latency        Enter a special mode continuously sampling latency\n"
-"  --slave          Simulate a slave showing commands received from the master\n"
-"  --rdb <filename> Transfer an RDB dump from remote server to local file.\n"
-"  --pipe           Transfer raw Redis protocol from stdin to server\n"
-"  --bigkeys        Sample Redis keys looking for big keys\n"
-"  --eval <file>    Send an EVAL command using the Lua script at <file>\n"
-"  --help           Output this help and exit\n"
-"  --version        Output version and exit\n"
+"  -h <hostname>     Server hostname (default: 127.0.0.1)\n"
+"  -p <port>         Server port (default: 6379)\n"
+"  -s <socket>       Server socket (overrides hostname and port)\n"
+"  -a <password>     Password to use when connecting to the server\n"
+"  -r <repeat>       Execute specified command N times\n"
+"  -i <interval>     When -r is used, waits <interval> seconds per command.\n"
+"                    It is possible to specify sub-second times like -i 0.1\n"
+"  -n <db>           Database number\n"
+"  -x                Read last argument from STDIN\n"
+"  -d <delimiter>    Multi-bulk delimiter in for raw formatting (default: \\n)\n"
+"  -c                Enable cluster mode (follow -ASK and -MOVED redirections)\n"
+"  --raw             Use raw formatting for replies (default when STDOUT is\n"
+"                    not a tty)\n"
+"  --latency         Enter a special mode continuously sampling latency\n"
+"  --latency-history Like --latency but tracking latency changes over time.\n"
+"                    Default time interval is 15 sec. Change it using -i.\n"
+"  --slave           Simulate a slave showing commands received from the master\n"
+"  --rdb <filename>  Transfer an RDB dump from remote server to local file.\n"
+"  --pipe            Transfer raw Redis protocol from stdin to server\n"
+"  --bigkeys         Sample Redis keys looking for big keys\n"
+"  --eval <file>     Send an EVAL command using the Lua script at <file>\n"
+"  --help            Output this help and exit\n"
+"  --version         Output version and exit\n"
 "\n"
 "Examples:\n"
 "  cat /etc/passwd | redis-cli -x set mypasswd\n"
@@ -909,10 +958,16 @@ static int evalMode(int argc, char **argv) {
    return cliSendCommand(argc+3-got_comma, argv2, config.repeat);
 }

+#define LATENCY_SAMPLE_RATE 10 /* milliseconds. */
+#define LATENCY_HISTORY_DEFAULT_INTERVAL 15000 /* milliseconds. */
 static void latencyMode(void) {
    redisReply *reply;
    long long start, latency, min = 0, max = 0, tot = 0, count = 0;
+    long long history_interval =
+        config.interval ? config.interval/1000 :
+                          LATENCY_HISTORY_DEFAULT_INTERVAL;
    double avg;
+    long long history_start = mstime();

    if (!context) exit(1);
    while(1) {
@@ -937,7 +992,13 @@ static void latencyMode(void) {
        printf("\x1b[0G\x1b[2Kmin: %lld, max: %lld, avg: %.2f (%lld samples)",
            min, max, avg, count);
        fflush(stdout);
-        usleep(10000);
+        if (config.latency_history && mstime()-history_start > history_interval)
+        {
+            printf(" -- %.2f seconds range\n", (float)(mstime()-history_start)/1000);
+            history_start = mstime();
+            min = max = tot = count = 0;
+        }
+        usleep(LATENCY_SAMPLE_RATE * 1000);
    }
 }

@@ -1205,7 +1266,11 @@ static void findBigKeys(void) {
            fprintf(stderr, "RANDOMKEY error: %s\n",
                reply1->str);
            exit(1);
+        } else if (reply1->type == REDIS_REPLY_NIL) {
+            fprintf(stderr, "It looks like the database is empty!\n");
+            exit(1);
        }
+
        /* Get the key type */
        reply2 = redisCommand(context,"TYPE %s",reply1->str);
        assert(reply2 && reply2->type == REDIS_REPLY_STATUS);
@@ -1261,6 +1326,145 @@ static void findBigKeys(void) {
    }
 }

+/* Return the specified INFO field from the INFO command output "info".
+ * A new buffer is allocated for the result, that needs to be free'd.
+ * If the field is not found NULL is returned. */
+static char *getInfoField(char *info, char *field) {
+    char *p = strstr(info,field);
+    char *n1, *n2;
+    char *result;
+
+    if (!p) return NULL;
+    p += strlen(field)+1;
+    n1 = strchr(p,'\r');
+    n2 = strchr(p,',');
+    if (n2 && n2 < n1) n1 = n2;
+    result = malloc(sizeof(char)*(n1-p)+1);
+    memcpy(result,p,(n1-p));
+    result[n1-p] = '\0';
+    return result;
+}
+
+/* Like the above function but automatically convert the result into
+ * a long. On error (missing field) LONG_MIN is returned. */
+static long getLongInfoField(char *info, char *field) {
+    char *value = getInfoField(info,field);
+    long l;
+
+    if (!value) return LONG_MIN;
+    l = strtol(value,NULL,10);
+    free(value);
+    return l;
+}
+
+/* Convert number of bytes into a human readable string of the form:
+ * 100B, 2G, 100M, 4K, and so forth. */
+void bytesToHuman(char *s, long long n) {
+    double d;
+
+    if (n < 0) {
+        *s = '-';
+        s++;
+        n = -n;
+    }
+    if (n < 1024) {
+        /* Bytes */
+        sprintf(s,"%lluB",n);
+        return;
+    } else if (n < (1024*1024)) {
+        d = (double)n/(1024);
+        sprintf(s,"%.2fK",d);
+    } else if (n < (1024LL*1024*1024)) {
+        d = (double)n/(1024*1024);
+        sprintf(s,"%.2fM",d);
+    } else if (n < (1024LL*1024*1024*1024)) {
+        d = (double)n/(1024LL*1024*1024);
+        sprintf(s,"%.2fG",d);
+    }
+}
+
+static void statMode() {
+    redisReply *reply;
+    long aux, requests = 0;
+    int i = 0;
+
+    while(1) {
+        char buf[64];
+        int j;
+
+        reply = reconnectingInfo();
+        if (reply->type == REDIS_REPLY_ERROR) {
+            printf("ERROR: %s\n", reply->str);
+            exit(1);
+        }
+
+        if ((i++ % 20) == 0) {
+            printf(
+"------- data ------ --------------------- load -------------------- - child -\n"
+"keys       mem      clients blocked requests            connections          \n");
+        }
+
+        /* Keys */
+        aux = 0;
+        for (j = 0; j < 20; j++) {
+            long k;
+
+            sprintf(buf,"db%d:keys",j);
+            k = getLongInfoField(reply->str,buf);
+            if (k == LONG_MIN) continue;
+            aux += k;
+        }
+        sprintf(buf,"%ld",aux);
+        printf("%-11s",buf);
+
+        /* Used memory */
+        aux = getLongInfoField(reply->str,"used_memory");
+        bytesToHuman(buf,aux);
+        printf("%-8s",buf);
+
+        /* Clients */
+        aux = getLongInfoField(reply->str,"connected_clients");
+        sprintf(buf,"%ld",aux);
+        printf(" %-8s",buf);
+
+        /* Blocked (BLPOPPING) Clients */
+        aux = getLongInfoField(reply->str,"blocked_clients");
+        sprintf(buf,"%ld",aux);
+        printf("%-8s",buf);
+
+        /* Requets */
+        aux = getLongInfoField(reply->str,"total_commands_processed");
+        sprintf(buf,"%ld (+%ld)",aux,requests == 0 ? 0 : aux-requests);
+        printf("%-19s",buf);
+        requests = aux;
+
+        /* Connections */
+        aux = getLongInfoField(reply->str,"total_connections_received");
+        sprintf(buf,"%ld",aux);
+        printf(" %-12s",buf);
+
+        /* Children */
+        aux = getLongInfoField(reply->str,"bgsave_in_progress");
+        aux |= getLongInfoField(reply->str,"aof_rewrite_in_progress") << 1;
+        switch(aux) {
+        case 0: break;
+        case 1:
+            printf("SAVE");
+            break;
+        case 2:
+            printf("AOF");
+            break;
+        case 3:
+            printf("SAVE+AOF");
+            break;
+        }
+
+        printf("\n");
+        freeReplyObject(reply);
+        usleep(config.interval);
+    }
+}
+
 int main(int argc, char **argv) {
    int firstarg;

@@ -1275,6 +1479,7 @@ int main(int argc, char **argv) {
    config.monitor_mode = 0;
    config.pubsub_mode = 0;
    config.latency_mode = 0;
+    config.latency_history = 0;
    config.cluster_mode = 0;
    config.slave_mode = 0;
    config.getrdb_mode = 0;
@@ -1325,6 +1530,13 @@ int main(int argc, char **argv) {
        findBigKeys();
    }

+    /* Stat mode */
+    if (config.stat_mode) {
+        if (cliConnect(0) == REDIS_ERR) exit(1);
+        if (config.interval == 0) config.interval = 1000000;
+        statMode();
+    }
+
    /* Start interactive mode when no command is provided */
    if (argc == 0 && !config.eval) {
        /* Note that in repl mode we don't abort on connection error.
@@ -114,7 +114,7 @@ struct redisCommand *commandTable;
 */
 struct redisCommand redisCommandTable[] = {
    {"get",getCommand,2,"r",0,NULL,1,1,1,0,0},
-    {"set",setCommand,3,"wm",0,noPreloadGetKeys,1,1,1,0,0},
+    {"set",setCommand,-3,"wm",0,noPreloadGetKeys,1,1,1,0,0},
    {"setnx",setnxCommand,3,"wm",0,noPreloadGetKeys,1,1,1,0,0},
    {"setex",setexCommand,4,"wm",0,noPreloadGetKeys,1,1,1,0,0},
    {"psetex",psetexCommand,4,"wm",0,noPreloadGetKeys,1,1,1,0,0},
@@ -197,7 +197,7 @@ struct redisCommand redisCommandTable[] = {
    {"mset",msetCommand,-3,"wm",0,NULL,1,-1,2,0,0},
    {"msetnx",msetnxCommand,-3,"wm",0,NULL,1,-1,2,0,0},
    {"randomkey",randomkeyCommand,1,"rR",0,NULL,0,0,0,0,0},
-    {"select",selectCommand,2,"r",0,NULL,0,0,0,0,0},
+    {"select",selectCommand,2,"rl",0,NULL,0,0,0,0,0},
    {"move",moveCommand,3,"w",0,NULL,1,1,1,0,0},
    {"rename",renameCommand,3,"w",0,renameGetKeys,1,2,1,0,0},
    {"renamenx",renamenxCommand,3,"w",0,renameGetKeys,1,2,1,0,0},
@@ -236,7 +236,7 @@ struct redisCommand redisCommandTable[] = {
    {"unsubscribe",unsubscribeCommand,-1,"rpslt",0,NULL,0,0,0,0,0},
    {"psubscribe",psubscribeCommand,-2,"rpslt",0,NULL,0,0,0,0,0},
    {"punsubscribe",punsubscribeCommand,-1,"rpslt",0,NULL,0,0,0,0,0},
-    {"publish",publishCommand,3,"pflt",0,NULL,0,0,0,0,0},
+    {"publish",publishCommand,3,"pfltr",0,NULL,0,0,0,0,0},
    {"watch",watchCommand,-2,"rs",0,noPreloadGetKeys,1,-1,1,0,0},
    {"unwatch",unwatchCommand,1,"rs",0,NULL,0,0,0,0,0},
    {"restore",restoreCommand,4,"awm",0,NULL,1,1,1,0,0},
@@ -825,7 +825,8 @@ void clientsCron(void) {
 void databasesCron(void) {
    /* Expire keys by random sampling. Not required for slaves
     * as master will synthesize DELs for us. */
-    if (server.masterhost == NULL) activeExpireCycle();
+    if (server.active_expire_enabled && server.masterhost == NULL)
+        activeExpireCycle();

    /* Perform hash tables rehashing if needed, but only if there are no
     * other processes saving the DB on disk. Otherwise rehashing is bad
@@ -994,8 +995,16 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) {
         for (j = 0; j < server.saveparamslen; j++) {
            struct saveparam *sp = server.saveparams+j;

+            /* Save if we reached the given amount of changes,
+             * the given amount of seconds, and if the latest bgsave was
+             * successful or if, in case of an error, at least
+             * REDIS_BGSAVE_RETRY_DELAY seconds already elapsed. */
            if (server.dirty >= sp->changes &&
-                server.unixtime-server.lastsave > sp->seconds) {
+                server.unixtime-server.lastsave > sp->seconds &&
+                (server.unixtime-server.lastbgsave_try >
+                 REDIS_BGSAVE_RETRY_DELAY ||
+                 server.lastbgsave_status == REDIS_OK))
+            {
                redisLog(REDIS_NOTICE,"%d changes in %d seconds. Saving...",
                    sp->changes, sp->seconds);
                rdbSaveBackground(server.rdb_filename);
@@ -1156,6 +1165,7 @@ void initServerConfig() {
    server.verbosity = REDIS_NOTICE;
    server.maxidletime = REDIS_MAXIDLETIME;
    server.tcpkeepalive = 0;
+    server.active_expire_enabled = 1;
    server.client_max_querybuf_len = REDIS_MAX_QUERYBUF_LEN;
    server.saveparams = NULL;
    server.loading = 0;
@@ -1179,12 +1189,14 @@ void initServerConfig() {
    server.aof_fd = -1;
    server.aof_selected_db = -1; /* Make sure the first time will not match */
    server.aof_flush_postponed_start = 0;
+    server.aof_rewrite_incremental_fsync = 1;
    server.pidfile = zstrdup("/var/run/redis.pid");
    server.rdb_filename = zstrdup("dump.rdb");
    server.aof_filename = zstrdup("appendonly.aof");
    server.requirepass = NULL;
    server.rdb_compression = 1;
    server.rdb_checksum = 1;
+    server.stop_writes_on_bgsave_err = 1;
    server.activerehashing = 1;
    server.maxclients = REDIS_MAX_CLIENTS;
    server.bpop_blocked_clients = 0;
@@ -1221,7 +1233,7 @@ void initServerConfig() {
    server.repl_syncio_timeout = REDIS_REPL_SYNCIO_TIMEOUT;
    server.repl_serve_stale_data = 1;
    server.repl_slave_ro = 1;
-    server.repl_down_since = time(NULL);
+    server.repl_down_since = 0; /* Never connected, repl is down since EVER. */
    server.repl_disable_tcp_nodelay = 0;
    server.slave_priority = REDIS_DEFAULT_SLAVE_PRIORITY;

@@ -1372,7 +1384,8 @@ void initServer() {
    server.aof_child_pid = -1;
    aofRewriteBufferReset();
    server.aof_buf = sdsempty();
-    server.lastsave = time(NULL);
+    server.lastsave = time(NULL); /* At startup we consider the DB saved. */
+    server.lastbgsave_try = 0;    /* At startup we never tried to BGSAVE. */
    server.rdb_save_time_last = -1;
    server.rdb_save_time_start = -1;
    server.dirty = 0;
@@ -1392,7 +1405,6 @@ void initServer() {
    server.ops_sec_last_sample_ops = 0;
    server.unixtime = time(NULL);
    server.lastbgsave_status = REDIS_OK;
-    server.stop_writes_on_bgsave_err = 1;
    if(aeCreateTimeEvent(server.el, 1, serverCron, NULL, NULL) == AE_ERR) {
        redisPanic("create time event failed");
        exit(1);
@@ -2603,7 +2615,7 @@ void loadDataFromDisk(void) {
            redisLog(REDIS_NOTICE,"DB loaded from disk: %.3f seconds",
                (float)(ustime()-start)/1000000);
        } else if (errno != ENOENT) {
-            redisLog(REDIS_WARNING,"Fatal error loading the DB. Exiting.");
+            redisLog(REDIS_WARNING,"Fatal error loading the DB: %s. Exiting.",strerror(errno));
            exit(1);
        }
    }
@@ -2612,7 +2624,7 @@ void loadDataFromDisk(void) {
 void redisOutOfMemoryHandler(size_t allocation_size) {
    redisLog(REDIS_WARNING,"Out Of Memory allocating %zu bytes!",
        allocation_size);
-    redisPanic("OOM");
+    redisPanic("Redis aborting for OUT OF MEMORY");
 }

 int main(int argc, char **argv) {
@@ -94,6 +94,7 @@
 #define REDIS_REPL_PING_SLAVE_PERIOD 10
 #define REDIS_RUN_ID_SIZE 40
 #define REDIS_OPS_SEC_SAMPLES 16
+#define REDIS_BGSAVE_RETRY_DELAY 5 /* Wait a few secs before trying again. */

 /* Protocol and I/O related defines */
 #define REDIS_MAX_QUERYBUF_LEN  (1024*1024*1024) /* 1GB max query buffer. */
@@ -101,6 +102,7 @@
 #define REDIS_REPLY_CHUNK_BYTES (16*1024) /* 16k output buffer */
 #define REDIS_INLINE_MAX_SIZE   (1024*64) /* Max size of inline reads */
 #define REDIS_MBULK_BIG_ARG     (1024*32)
+#define REDIS_AOF_AUTOSYNC_BYTES (1024*1024*32) /* fdatasync every 32MB */

 /* Hash table parameters */
 #define REDIS_HT_MINFILL        10      /* Minimal hash table fill 10% */
@@ -557,6 +559,7 @@ struct redisServer {
    int verbosity;                  /* Loglevel in redis.conf */
    int maxidletime;                /* Client timeout in seconds */
    int tcpkeepalive;               /* Set SO_KEEPALIVE if non-zero. */
+    int active_expire_enabled;      /* Can be disabled for testing purposes. */
    size_t client_max_querybuf_len; /* Limit for client query buffer length */
    int dbnum;                      /* Total number of configured DBs */
    int daemonize;                  /* True if running as a daemon */
@@ -582,6 +585,7 @@ struct redisServer {
    time_t aof_rewrite_time_start;  /* Current AOF rewrite start time. */
    int aof_lastbgrewrite_status;   /* REDIS_OK or REDIS_ERR */
    unsigned long aof_delayed_fsync;  /* delayed AOF fsync() counter */
+    int aof_rewrite_incremental_fsync;/* fsync incrementally while rewriting? */
    /* RDB persistence */
    long long dirty;                /* Changes to DB from the last save */
    long long dirty_before_bgsave;  /* Used to restore dirty on failed BGSAVE */
@@ -592,6 +596,7 @@ struct redisServer {
    int rdb_compression;            /* Use compression in RDB? */
    int rdb_checksum;               /* Use RDB checksum? */
    time_t lastsave;                /* Unix time of last successful save */
+    time_t lastbgsave_try;          /* Unix time of last attempted bgsave */
    time_t rdb_save_time_last;      /* Time used by last RDB save run. */
    time_t rdb_save_time_start;     /* Current RDB save start time. */
    int lastbgsave_status;          /* REDIS_OK or REDIS_ERR */
@@ -48,8 +48,11 @@
 #include "fmacros.h"
 #include <string.h>
 #include <stdio.h>
+#include <unistd.h>
 #include "rio.h"
 #include "util.h"
+#include "config.h"
+#include "redis.h"

 uint64_t crc64(uint64_t crc, const unsigned char *s, uint64_t l);

@@ -76,7 +79,18 @@ static off_t rioBufferTell(rio *r) {

 /* Returns 1 or 0 for success/failure. */
 static size_t rioFileWrite(rio *r, const void *buf, size_t len) {
-    return fwrite(buf,len,1,r->io.file.fp);
+    size_t retval;
+
+    retval = fwrite(buf,len,1,r->io.file.fp);
+    r->io.file.buffered += len;
+
+    if (r->io.file.autosync &&
+        r->io.file.buffered >= r->io.file.autosync)
+    {
+        aof_fsync(fileno(r->io.file.fp));
+        r->io.file.buffered = 0;
+    }
+    return retval;
 }

 /* Returns 1 or 0 for success/failure. */
@@ -110,6 +124,8 @@ static const rio rioFileIO = {
 void rioInitWithFile(rio *r, FILE *fp) {
    *r = rioFileIO;
    r->io.file.fp = fp;
+    r->io.file.buffered = 0;
+    r->io.file.autosync = 0;
 }

 void rioInitWithBuffer(rio *r, sds s) {
@@ -124,6 +140,19 @@ void rioGenericUpdateChecksum(rio *r, const void *buf, size_t len) {
    r->cksum = crc64(r->cksum,buf,len);
 }

+/* Set the file-based rio object to auto-fsync every 'bytes' file written.
+ * By default this is set to zero that means no automatic file sync is
+ * performed.
+ *
+ * This feature is useful in a few contexts since when we rely on OS write
+ * buffers sometimes the OS buffers way too much, resulting in too many
+ * disk I/O concentrated in very little time. When we fsync in an explicit
+ * way instead the I/O pressure is more distributed across time. */
+void rioSetAutoSync(rio *r, off_t bytes) {
+    redisAssert(r->read == rioFileIO.read);
+    r->io.file.autosync = bytes;
+}
+
 /* ------------------------------ Higher level interface ---------------------------
 * The following higher level functions use lower level rio.c functions to help
 * generating the Redis protocol for the Append Only File. */
@@ -43,10 +43,11 @@ struct _rio {
    size_t (*read)(struct _rio *, void *buf, size_t len);
    size_t (*write)(struct _rio *, const void *buf, size_t len);
    off_t (*tell)(struct _rio *);
-    /* The update_cksum method if not NULL is used to compute the checksum of all the
-     * data that was read or written so far. The method should be designed so that
-     * can be called with the current checksum, and the buf and len fields pointing
-     * to the new block of data to add to the checksum computation. */
+    /* The update_cksum method if not NULL is used to compute the checksum of
+     * all the data that was read or written so far. The method should be
+     * designed so that can be called with the current checksum, and the buf
+     * and len fields pointing to the new block of data to add to the checksum
+     * computation. */
    void (*update_cksum)(struct _rio *, const void *buf, size_t len);

    /* The current checksum */
@@ -60,6 +61,8 @@ struct _rio {
        } buffer;
        struct {
            FILE *fp;
+            off_t buffered; /* Bytes written since last fsync. */
+            off_t autosync; /* fsync after 'autosync' bytes written. */
        } file;
    } io;
 };
@@ -96,5 +99,6 @@ size_t rioWriteBulkLongLong(rio *r, long long l);
 size_t rioWriteBulkDouble(rio *r, double d);

 void rioGenericUpdateChecksum(rio *r, const void *buf, size_t len);
+void rioSetAutoSync(rio *r, off_t bytes);

 #endif
@@ -258,6 +258,7 @@ int luaRedisGenericCommand(lua_State *lua, int raise_error) {
                "Write commands not allowed after non deterministic commands");
            goto cleanup;
        } else if (server.masterhost && server.repl_slave_ro &&
+                   !server.loading &&
                   !(server.lua_caller->flags & REDIS_MASTER))
        {
            luaPushError(lua, shared.roslaveerr->ptr);
@@ -74,6 +74,8 @@ typedef struct sentinelAddr {
 #define SRI_RECONF_DONE (1<<13)     /* Slave synchronized with new master. */
 #define SRI_FORCE_FAILOVER (1<<14)  /* Force failover with master up. */
 #define SRI_SCRIPT_KILL_SENT (1<<15) /* SCRIPT KILL already sent on -BUSY */
+#define SRI_DEMOTE (1<<16)   /* If the instance claims to be a master, demote
+                                it into a slave sending SLAVEOF. */

 #define SENTINEL_INFO_PERIOD 10000
 #define SENTINEL_PING_PERIOD 1000
@@ -403,7 +405,7 @@ void initSentinel(void) {
    /* Initialize various data structures. */
    sentinel.masters = dictCreate(&instancesDictType,NULL);
    sentinel.tilt = 0;
-    sentinel.tilt_start_time = mstime();
+    sentinel.tilt_start_time = 0;
    sentinel.previous_time = mstime();
    sentinel.running_scripts = 0;
    sentinel.scripts_queue = listCreate();
@@ -1130,7 +1132,6 @@ int sentinelResetMastersByPattern(char *pattern, int flags) {
 * TODO: make this reset so that original sentinels are re-added with
 * same ip / port / runid.
 */
-
 int sentinelResetMasterAndChangeAddress(sentinelRedisInstance *master, char *ip, int port) {
    sentinelAddr *oldaddr, *newaddr;

@@ -1139,12 +1140,26 @@ int sentinelResetMasterAndChangeAddress(sentinelRedisInstance *master, char *ip,
    sentinelResetMaster(master,SENTINEL_NO_FLAGS);
    oldaddr = master->addr;
    master->addr = newaddr;
+    master->o_down_since_time = 0;
+    master->s_down_since_time = 0;
+
    /* Release the old address at the end so we are safe even if the function
     * gets the master->addr->ip and master->addr->port as arguments. */
    releaseSentinelAddr(oldaddr);
    return REDIS_OK;
 }

+/* Return non-zero if there was no SDOWN or ODOWN error associated to this
+ * instance in the latest 'ms' milliseconds. */
+int sentinelRedisInstanceNoDownFor(sentinelRedisInstance *ri, mstime_t ms) {
+    mstime_t most_recent;
+
+    most_recent = ri->s_down_since_time;
+    if (ri->o_down_since_time > most_recent)
+        most_recent = ri->o_down_since_time;
+    return most_recent == 0 || (mstime() - most_recent) > ms;
+}
+
 /* ============================ Config handling ============================= */
 char *sentinelHandleConfiguration(char **argv, int argc) {
    sentinelRedisInstance *ri;
@@ -1382,26 +1397,39 @@ void sentinelRefreshInstanceInfo(sentinelRedisInstance *ri, const char *info) {
            }
        }

-        /* slave0:<ip>,<port>,<state> */
+        /* old versions: slave0:<ip>,<port>,<state>
+         * new versions: slave0:ip=127.0.0.1,port=9999,... */
        if ((ri->flags & SRI_MASTER) &&
            sdslen(l) >= 7 &&
            !memcmp(l,"slave",5) && isdigit(l[5]))
        {
            char *ip, *port, *end;

-            ip = strchr(l,':'); if (!ip) continue;
-            ip++; /* Now ip points to start of ip address. */
-            port = strchr(ip,','); if (!port) continue;
-            *port = '\0'; /* nul term for easy access. */
-            port++; /* Now port points to start of port number. */
-            end = strchr(port,','); if (!end) continue;
-            *end = '\0'; /* nul term for easy access. */
+            if (strstr(l,"ip=") == NULL) {
+                /* Old format. */
+                ip = strchr(l,':'); if (!ip) continue;
+                ip++; /* Now ip points to start of ip address. */
+                port = strchr(ip,','); if (!port) continue;
+                *port = '\0'; /* nul term for easy access. */
+                port++; /* Now port points to start of port number. */
+                end = strchr(port,','); if (!end) continue;
+                *end = '\0'; /* nul term for easy access. */
+            } else {
+                /* New format. */
+                ip = strstr(l,"ip="); if (!ip) continue;
+                ip += 3; /* Now ip points to start of ip address. */
+                port = strstr(l,"port="); if (!port) continue;
+                port += 5; /* Now port points to start of port number. */
+                /* Nul term both fields for easy access. */
+                end = strchr(ip,','); if (end) *end = '\0';
+                end = strchr(port,','); if (end) *end = '\0';
+            }

            /* Check if we already have this slave into our table,
             * otherwise add it. */
            if (sentinelRedisInstanceLookupSlave(ri,ip,atoi(port)) == NULL) {
                if ((slave = createSentinelRedisInstance(NULL,SRI_SLAVE,ip,
-                            atoi(port), ri->quorum,ri)) != NULL)
+                            atoi(port), ri->quorum, ri)) != NULL)
                {
                    sentinelEvent(REDIS_NOTICE,"+slave",slave,"%@");
                }
@@ -1446,44 +1474,80 @@ void sentinelRefreshInstanceInfo(sentinelRedisInstance *ri, const char *info) {
    ri->info_refresh = mstime();
    sdsfreesplitres(lines,numlines);

-    /* ---------------------------- Acting half ----------------------------- */
-    if (sentinel.tilt) return;
+    /* ---------------------------- Acting half -----------------------------
+     * Some things will not happen if sentinel.tilt is true, but some will
+     * still be processed. */

-    /* Act if a master turned into a slave. */
-    if ((ri->flags & SRI_MASTER) && role == SRI_SLAVE) {
-        if ((first_runid || runid_changed) && ri->slave_master_host) {
-            /* If it is the first time we receive INFO from it, but it's
-             * a slave while it was configured as a master, we want to monitor
-             * its master instead. */
-            sentinelEvent(REDIS_WARNING,"+redirect-to-master",ri,
-                "%s %s %d %s %d",
-                ri->name, ri->addr->ip, ri->addr->port,
-                ri->slave_master_host, ri->slave_master_port);
-            sentinelResetMasterAndChangeAddress(ri,ri->slave_master_host,
-                                                   ri->slave_master_port);
-            return;
-        }
+    /* When what we believe is our master, turned into a slave, the wiser
+     * thing we can do is to follow the events and redirect to the new
+     * master, always. */
+    if ((ri->flags & SRI_MASTER) && role == SRI_SLAVE && ri->slave_master_host)
+    {
+        sentinelEvent(REDIS_WARNING,"+redirect-to-master",ri,
+            "%s %s %d %s %d",
+            ri->name, ri->addr->ip, ri->addr->port,
+            ri->slave_master_host, ri->slave_master_port);
+        sentinelResetMasterAndChangeAddress(ri,ri->slave_master_host,
+                                               ri->slave_master_port);
+        return; /* Don't process anything after this event. */
    }

-    /* Act if a slave turned into a master. */
+    /* Handle slave -> master role switch. */
    if ((ri->flags & SRI_SLAVE) && role == SRI_MASTER) {
-        if (!(ri->master->flags & SRI_FAILOVER_IN_PROGRESS) &&
-            (runid_changed || first_runid))
+        if (!sentinel.tilt && ri->flags & SRI_DEMOTE) {
+            /* If this sentinel was partitioned from the slave's master,
+             * or tilted recently, wait some time before to act,
+             * so that DOWN and roles INFO will be refreshed. */
+            mstime_t wait_time = SENTINEL_INFO_PERIOD*2 +
+                                 ri->master->down_after_period*2;
+
+            if (!sentinelRedisInstanceNoDownFor(ri->master,wait_time) ||
+                (mstime()-sentinel.tilt_start_time) < wait_time)
+                return;
+
+            /* Old master returned back? Turn it into a slave ASAP if
+             * we can reach what we believe is the new master now, and
+             * have a recent role information for it.
+             *
+             * Note: we'll clear the DEMOTE flag only when we have the
+             * acknowledge that it's a slave again. */
+            if (ri->master->flags & SRI_MASTER &&
+                (ri->master->flags & (SRI_S_DOWN|SRI_O_DOWN)) == 0 &&
+                (mstime() - ri->master->info_refresh) < SENTINEL_INFO_PERIOD*2)
+            {
+                int retval;
+                retval = redisAsyncCommand(ri->cc,
+                    sentinelDiscardReplyCallback, NULL, "SLAVEOF %s %d",
+                        ri->master->addr->ip,
+                        ri->master->addr->port);
+                if (retval == REDIS_OK)
+                    sentinelEvent(REDIS_NOTICE,"+demote-old-slave",ri,"%@");
+            } else {
+                /* Otherwise if there are not the conditions to demote, we
+                 * no longer trust the DEMOTE flag and remove it. */
+                ri->flags &= ~SRI_DEMOTE;
+                sentinelEvent(REDIS_NOTICE,"-demote-flag-cleared",ri,"%@");
+            }
+        } else if (!(ri->master->flags & SRI_FAILOVER_IN_PROGRESS) &&
+                    (runid_changed || first_runid))
        {
            /* If a slave turned into master but:
             *
             * 1) Failover not in progress.
-             * 2) RunID hs changed, or its the first time we see an INFO output.
+             * 2) RunID has changed or its the first time we see an INFO output.
             * 
             * We assume this is a reboot with a wrong configuration.
-             * Log the event and remove the slave. */
+             * Log the event and remove the slave. Note that this is processed
+             * in tilt mode as well, otherwise we lose the information that the
+             * runid changed (reboot?) and when the tilt mode ends a fake
+             * failover will be detected. */
            int retval;

            sentinelEvent(REDIS_WARNING,"-slave-restart-as-master",ri,"%@ #removing it from the attached slaves");
            retval = dictDelete(ri->master->slaves,ri->name);
            redisAssert(retval == REDIS_OK);
            return;
-        } else if (ri->flags & SRI_PROMOTED) {
+        } else if (!sentinel.tilt && ri->flags & SRI_PROMOTED) {
            /* If this is a promoted slave we can change state to the
             * failover state machine. */
            if ((ri->master->flags & SRI_FAILOVER_IN_PROGRESS) &&
@@ -1499,11 +1563,12 @@ void sentinelRefreshInstanceInfo(sentinelRedisInstance *ri, const char *info) {
                sentinelCallClientReconfScript(ri->master,SENTINEL_LEADER,
                    "start",ri->master->addr,ri->addr);
            }
-        } else if (!(ri->master->flags & SRI_FAILOVER_IN_PROGRESS) ||
-                    ((ri->master->flags & SRI_FAILOVER_IN_PROGRESS) &&
-                     (ri->master->flags & SRI_I_AM_THE_LEADER) &&
-                     ri->master->failover_state ==
-                     SENTINEL_FAILOVER_STATE_WAIT_START))
+        } else if (!sentinel.tilt && (
+                    !(ri->master->flags & SRI_FAILOVER_IN_PROGRESS) ||
+                     ((ri->master->flags & SRI_FAILOVER_IN_PROGRESS) &&
+                      (ri->master->flags & SRI_I_AM_THE_LEADER) &&
+                       ri->master->failover_state ==
+                       SENTINEL_FAILOVER_STATE_WAIT_START)))
        {
            /* No failover in progress? Then it is the start of a failover
             * and we are an observer.
@@ -1523,6 +1588,7 @@ void sentinelRefreshInstanceInfo(sentinelRedisInstance *ri, const char *info) {
            ri->master->failover_state_change_time = mstime();
            ri->master->promoted_slave = ri;
            ri->flags |= SRI_PROMOTED;
+            ri->flags &= ~SRI_DEMOTE;
            sentinelCallClientReconfScript(ri->master,SENTINEL_OBSERVER,
                "start", ri->master->addr,ri->addr);
            /* We are an observer, so we can only assume that the leader
@@ -1534,6 +1600,10 @@ void sentinelRefreshInstanceInfo(sentinelRedisInstance *ri, const char *info) {
        }
    }

+    /* None of the following conditions are processed when in tilt mode, so
+     * return asap. */
+    if (sentinel.tilt) return;
+
    /* Detect if the slave that is in the process of being reconfigured
     * changed state. */
    if ((ri->flags & SRI_SLAVE) && role == SRI_SLAVE &&
@@ -1564,6 +1634,13 @@ void sentinelRefreshInstanceInfo(sentinelRedisInstance *ri, const char *info) {
            ri->failover_state_change_time = mstime();
        }
    }
+
+    /* Detect if the old master was demoted as slave and generate the
+     * +slave event. */
+    if (role == SRI_SLAVE && ri->flags & SRI_DEMOTE) {
+        sentinelEvent(REDIS_NOTICE,"+slave",ri,"%@");
+        ri->flags &= ~SRI_DEMOTE;
+    }
 }

 void sentinelInfoReplyCallback(redisAsyncContext *c, void *reply, void *privdata) {
@@ -1835,6 +1912,7 @@ void addReplySentinelRedisInstance(redisClient *c, sentinelRedisInstance *ri) {
    if (ri->flags & SRI_RECONF_SENT) flags = sdscat(flags,"reconf_sent,");
    if (ri->flags & SRI_RECONF_INPROG) flags = sdscat(flags,"reconf_inprog,");
    if (ri->flags & SRI_RECONF_DONE) flags = sdscat(flags,"reconf_done,");
+    if (ri->flags & SRI_DEMOTE) flags = sdscat(flags,"demote,");

    if (sdslen(flags) != 0) flags = sdsrange(flags,0,-2); /* remove last "," */
    addReplyBulkCString(c,flags);
@@ -2592,6 +2670,7 @@ sentinelRedisInstance *sentinelSelectSlave(sentinelRedisInstance *master) {
        mstime_t info_validity_time = mstime()-SENTINEL_INFO_VALIDITY_TIME;

        if (slave->flags & (SRI_S_DOWN|SRI_O_DOWN|SRI_DISCONNECTED)) continue;
+        if (slave->flags & SRI_DEMOTE) continue; /* Old master not yet ready. */
        if (slave->last_avail_time < info_validity_time) continue;
        if (slave->slave_priority == 0) continue;

@@ -2837,12 +2916,28 @@ void sentinelFailoverReconfNextSlave(sentinelRedisInstance *master) {
 void sentinelFailoverSwitchToPromotedSlave(sentinelRedisInstance *master) {
    sentinelRedisInstance *ref = master->promoted_slave ?
                                 master->promoted_slave : master;
+    sds old_master_ip;
+    int old_master_port;

    sentinelEvent(REDIS_WARNING,"+switch-master",master,"%s %s %d %s %d",
        master->name, master->addr->ip, master->addr->port,
        ref->addr->ip, ref->addr->port);

+    old_master_ip = sdsdup(master->addr->ip);
+    old_master_port = master->addr->port;
    sentinelResetMasterAndChangeAddress(master,ref->addr->ip,ref->addr->port);
+    /* If this is a real switch, that is, we have master->promoted_slave not
+     * NULL, then we want to add the old master as a slave of the new master,
+     * but flagging it with SRI_DEMOTE so that we know we'll need to send
+     * SLAVEOF once the old master is reachable again. */
+    if (master != ref) {
+        /* Add the new slave, but don't generate a Sentinel event as it will
+         * happen later when finally the instance will claim to be a slave
+         * in the INFO output. */
+        createSentinelRedisInstance(NULL,SRI_SLAVE|SRI_DEMOTE,
+                    old_master_ip, old_master_port, master->quorum, master);
+    }
+    sdsfree(old_master_ip);
 }

 void sentinelFailoverStateMachine(sentinelRedisInstance *ri) {
@@ -42,7 +42,27 @@ static int checkStringLength(redisClient *c, long long size) {
    return REDIS_OK;
 }

-void setGenericCommand(redisClient *c, int nx, robj *key, robj *val, robj *expire, int unit) {
+/* The setGenericCommand() function implements the SET operation with different
+ * options and variants. This function is called in order to implement the
+ * following commands: SET, SETEX, PSETEX, SETNX.
+ *
+ * 'flags' changes the behavior of the command (NX or XX, see belove).
+ *
+ * 'expire' represents an expire to set in form of a Redis object as passed
+ * by the user. It is interpreted according to the specified 'unit'.
+ *
+ * 'ok_reply' and 'abort_reply' is what the function will reply to the client
+ * if the operation is performed, or when it is not because of NX or
+ * XX flags.
+ *
+ * If ok_reply is NULL "+OK" is used.
+ * If abort_reply is NULL, "$-1" is used. */
+
+#define REDIS_SET_NO_FLAGS 0
+#define REDIS_SET_NX (1<<0)     /* Set if key not exists. */
+#define REDIS_SET_XX (1<<1)     /* Set if key exists. */
+
+void setGenericCommand(redisClient *c, int flags, robj *key, robj *val, robj *expire, int unit, robj *ok_reply, robj *abort_reply) {
    long long milliseconds = 0; /* initialized to avoid any harmness warning */

    if (expire) {
@@ -55,34 +75,68 @@ void setGenericCommand(redisClient *c, int nx, robj *key, robj *val, robj *expir
        if (unit == UNIT_SECONDS) milliseconds *= 1000;
    }

-    if (nx && lookupKeyWrite(c->db,key) != NULL) {
-        addReply(c,shared.czero);
+    if ((flags & REDIS_SET_NX && lookupKeyWrite(c->db,key) != NULL) ||
+        (flags & REDIS_SET_XX && lookupKeyWrite(c->db,key) == NULL))
+    {
+        addReply(c, abort_reply ? abort_reply : shared.nullbulk);
        return;
    }
    setKey(c->db,key,val);
    server.dirty++;
    if (expire) setExpire(c->db,key,mstime()+milliseconds);
-    addReply(c, nx ? shared.cone : shared.ok);
+    addReply(c, ok_reply ? ok_reply : shared.ok);
 }

+/* SET key value [NX] [XX] [EX <seconds>] [PX <milliseconds>] */
 void setCommand(redisClient *c) {
+    int j;
+    robj *expire = NULL;
+    int unit = UNIT_SECONDS;
+    int flags = REDIS_SET_NO_FLAGS;
+
+    for (j = 3; j < c->argc; j++) {
+        char *a = c->argv[j]->ptr;
+        robj *next = (j == c->argc-1) ? NULL : c->argv[j+1];
+
+        if ((a[0] == 'n' || a[0] == 'N') &&
+            (a[1] == 'x' || a[1] == 'X') && a[2] == '\0') {
+            flags |= REDIS_SET_NX;
+        } else if ((a[0] == 'x' || a[0] == 'X') &&
+                   (a[1] == 'x' || a[1] == 'X') && a[2] == '\0') {
+            flags |= REDIS_SET_XX;
+        } else if ((a[0] == 'e' || a[0] == 'E') &&
+                   (a[1] == 'x' || a[1] == 'X') && a[2] == '\0' && next) {
+            unit = UNIT_SECONDS;
+            expire = next;
+            j++;
+        } else if ((a[0] == 'p' || a[0] == 'P') &&
+                   (a[1] == 'x' || a[1] == 'X') && a[2] == '\0' && next) {
+            unit = UNIT_MILLISECONDS;
+            expire = next;
+            j++;
+        } else {
+            addReply(c,shared.syntaxerr);
+            return;
+        }
+    }
+
    c->argv[2] = tryObjectEncoding(c->argv[2]);
-    setGenericCommand(c,0,c->argv[1],c->argv[2],NULL,0);
+    setGenericCommand(c,flags,c->argv[1],c->argv[2],expire,unit,NULL,NULL);
 }

 void setnxCommand(redisClient *c) {
    c->argv[2] = tryObjectEncoding(c->argv[2]);
-    setGenericCommand(c,1,c->argv[1],c->argv[2],NULL,0);
+    setGenericCommand(c,REDIS_SET_NX,c->argv[1],c->argv[2],NULL,0,shared.cone,shared.czero);
 }

 void setexCommand(redisClient *c) {
    c->argv[3] = tryObjectEncoding(c->argv[3]);
-    setGenericCommand(c,0,c->argv[1],c->argv[3],c->argv[2],UNIT_SECONDS);
+    setGenericCommand(c,REDIS_SET_NO_FLAGS,c->argv[1],c->argv[3],c->argv[2],UNIT_SECONDS,NULL,NULL);
 }

 void psetexCommand(redisClient *c) {
    c->argv[3] = tryObjectEncoding(c->argv[3]);
-    setGenericCommand(c,0,c->argv[1],c->argv[3],c->argv[2],UNIT_MILLISECONDS);
+    setGenericCommand(c,REDIS_SET_NO_FLAGS,c->argv[1],c->argv[3],c->argv[2],UNIT_MILLISECONDS,NULL,NULL);
 }

 int getGenericCommand(redisClient *c) {
@@ -1 +1 @@
-#define REDIS_VERSION "2.6.11"
+#define REDIS_VERSION "2.6.14"
@@ -1,344 +1,23 @@
-# Redis configuration file example
+# Redis configuration for testing.

-# Note on units: when memory size is needed, it is possible to specifiy
-# it in the usual form of 1k 5GB 4M and so forth:
-#
-# 1k => 1000 bytes
-# 1kb => 1024 bytes
-# 1m => 1000000 bytes
-# 1mb => 1024*1024 bytes
-# 1g => 1000000000 bytes
-# 1gb => 1024*1024*1024 bytes
-#
-# units are case insensitive so 1GB 1Gb 1gB are all the same.
-
-# By default Redis does not run as a daemon. Use 'yes' if you need it.
-# Note that Redis will write a pid file in /var/run/redis.pid when daemonized.
 daemonize no
-
-# When running daemonized, Redis writes a pid file in /var/run/redis.pid by
-# default. You can specify a custom pid file location here.
 pidfile /var/run/redis.pid
-
-# Accept connections on the specified port, default is 6379.
 port 6379
-
-# If you want you can bind a single interface, if the bind option is not
-# specified all the interfaces will listen for incoming connections.
-#
-# bind 127.0.0.1
-
-# Specify the path for the unix socket that will be used to listen for
-# incoming connections. There is no default, so Redis will not listen
-# on a unix socket when not specified.
-#
-# unixsocket /tmp/redis.sock
-
-# Close the connection after a client is idle for N seconds (0 to disable)
 timeout 0
-
-# Set server verbosity to 'debug'
-# it can be one of:
-# debug (a lot of information, useful for development/testing)
-# verbose (many rarely useful info, but not a mess like the debug level)
-# notice (moderately verbose, what you want in production probably)
-# warning (only very important / critical messages are logged)
 loglevel verbose
-
-# Specify the log file name. Also 'stdout' can be used to force
-# Redis to log on the standard output. Note that if you use standard
-# output for logging but daemonize, logs will be sent to /dev/null
 logfile stdout
-
-# To enable logging to the system logger, just set 'syslog-enabled' to yes,
-# and optionally update the other syslog parameters to suit your needs.
-# syslog-enabled no
-
-# Specify the syslog identity.
-# syslog-ident redis
-
-# Specify the syslog facility.  Must be USER or between LOCAL0-LOCAL7.
-# syslog-facility local0
-
-# Set the number of databases. The default database is DB 0, you can select
-# a different one on a per-connection basis using SELECT <dbid> where
-# dbid is a number between 0 and 'databases'-1
 databases 16

-################################ SNAPSHOTTING  #################################
-#
-# Save the DB on disk:
-#
-#   save <seconds> <changes>
-#
-#   Will save the DB if both the given number of seconds and the given
-#   number of write operations against the DB occurred.
-#
-#   In the example below the behaviour will be to save:
-#   after 900 sec (15 min) if at least 1 key changed
-#   after 300 sec (5 min) if at least 10 keys changed
-#   after 60 sec if at least 10000 keys changed
-#
-#   Note: you can disable saving at all commenting all the "save" lines.
-
 save 900 1
 save 300 10
 save 60 10000

-# Compress string objects using LZF when dump .rdb databases?
-# For default that's set to 'yes' as it's almost always a win.
-# If you want to save some CPU in the saving child set it to 'no' but
-# the dataset will likely be bigger if you have compressible values or keys.
 rdbcompression yes
-
-# The filename where to dump the DB
 dbfilename dump.rdb
-
-# The working directory.
-#
-# The DB will be written inside this directory, with the filename specified
-# above using the 'dbfilename' configuration directive.
-# 
-# Also the Append Only File will be created inside this directory.
-# 
-# Note that you must specify a directory here, not a file name.
 dir ./

-################################# REPLICATION #################################
-
-# Master-Slave replication. Use slaveof to make a Redis instance a copy of
-# another Redis server. Note that the configuration is local to the slave
-# so for example it is possible to configure the slave to save the DB with a
-# different interval, or to listen to another port, and so on.
-#
-# slaveof <masterip> <masterport>
-
-# If the master is password protected (using the "requirepass" configuration
-# directive below) it is possible to tell the slave to authenticate before
-# starting the replication synchronization process, otherwise the master will
-# refuse the slave request.
-#
-# masterauth <master-password>
-
-# When a slave lost the connection with the master, or when the replication
-# is still in progress, the slave can act in two different ways:
-#
-# 1) if slave-serve-stale-data is set to 'yes' (the default) the slave will
-#    still reply to client requests, possibly with out of data data, or the
-#    data set may just be empty if this is the first synchronization.
-#
-# 2) if slave-serve-stale data is set to 'no' the slave will reply with
-#    an error "SYNC with master in progress" to all the kind of commands
-#    but to INFO and SLAVEOF.
-#
 slave-serve-stale-data yes
-
-################################## SECURITY ###################################
-
-# Require clients to issue AUTH <PASSWORD> before processing any other
-# commands.  This might be useful in environments in which you do not trust
-# others with access to the host running redis-server.
-#
-# This should stay commented out for backward compatibility and because most
-# people do not need auth (e.g. they run their own servers).
-# 
-# Warning: since Redis is pretty fast an outside user can try up to
-# 150k passwords per second against a good box. This means that you should
-# use a very strong password otherwise it will be very easy to break.
-#
-# requirepass foobared
-
-# Command renaming.
-#
-# It is possilbe to change the name of dangerous commands in a shared
-# environment. For instance the CONFIG command may be renamed into something
-# of hard to guess so that it will be still available for internal-use
-# tools but not available for general clients.
-#
-# Example:
-#
-# rename-command CONFIG b840fc02d524045429941cc15f59e41cb7be6c52
-#
-# It is also possilbe to completely kill a command renaming it into
-# an empty string:
-#
-# rename-command CONFIG ""
-
-################################### LIMITS ####################################
-
-# Set the max number of connected clients at the same time. By default there
-# is no limit, and it's up to the number of file descriptors the Redis process
-# is able to open. The special value '0' means no limits.
-# Once the limit is reached Redis will close all the new connections sending
-# an error 'max number of clients reached'.
-#
-# maxclients 128
-
-# Don't use more memory than the specified amount of bytes.
-# When the memory limit is reached Redis will try to remove keys with an
-# EXPIRE set. It will try to start freeing keys that are going to expire
-# in little time and preserve keys with a longer time to live.
-# Redis will also try to remove objects from free lists if possible.
-#
-# If all this fails, Redis will start to reply with errors to commands
-# that will use more memory, like SET, LPUSH, and so on, and will continue
-# to reply to most read-only commands like GET.
-#
-# WARNING: maxmemory can be a good idea mainly if you want to use Redis as a
-# 'state' server or cache, not as a real DB. When Redis is used as a real
-# database the memory usage will grow over the weeks, it will be obvious if
-# it is going to use too much memory in the long run, and you'll have the time
-# to upgrade. With maxmemory after the limit is reached you'll start to get
-# errors for write operations, and this may even lead to DB inconsistency.
-#
-# maxmemory <bytes>
-
-# MAXMEMORY POLICY: how Redis will select what to remove when maxmemory
-# is reached? You can select among five behavior:
-# 
-# volatile-lru -> remove the key with an expire set using an LRU algorithm
-# allkeys-lru -> remove any key accordingly to the LRU algorithm
-# volatile-random -> remove a random key with an expire set
-# allkeys->random -> remove a random key, any key
-# volatile-ttl -> remove the key with the nearest expire time (minor TTL)
-# noeviction -> don't expire at all, just return an error on write operations
-# 
-# Note: with all the kind of policies, Redis will return an error on write
-#       operations, when there are not suitable keys for eviction.
-#
-#       At the date of writing this commands are: set setnx setex append
-#       incr decr rpush lpush rpushx lpushx linsert lset rpoplpush sadd
-#       sinter sinterstore sunion sunionstore sdiff sdiffstore zadd zincrby
-#       zunionstore zinterstore hset hsetnx hmset hincrby incrby decrby
-#       getset mset msetnx exec sort
-#
-# The default is:
-#
-# maxmemory-policy volatile-lru
-
-# LRU and minimal TTL algorithms are not precise algorithms but approximated
-# algorithms (in order to save memory), so you can select as well the sample
-# size to check. For instance for default Redis will check three keys and
-# pick the one that was used less recently, you can change the sample size
-# using the following configuration directive.
-#
-# maxmemory-samples 3
-
-############################## APPEND ONLY MODE ###############################
-
-# By default Redis asynchronously dumps the dataset on disk. If you can live
-# with the idea that the latest records will be lost if something like a crash
-# happens this is the preferred way to run Redis. If instead you care a lot
-# about your data and don't want to that a single record can get lost you should
-# enable the append only mode: when this mode is enabled Redis will append
-# every write operation received in the file appendonly.aof. This file will
-# be read on startup in order to rebuild the full dataset in memory.
-#
-# Note that you can have both the async dumps and the append only file if you
-# like (you have to comment the "save" statements above to disable the dumps).
-# Still if append only mode is enabled Redis will load the data from the
-# log file at startup ignoring the dump.rdb file.
-#
-# IMPORTANT: Check the BGREWRITEAOF to check how to rewrite the append
-# log file in background when it gets too big.
-
 appendonly no
-
-# The name of the append only file (default: "appendonly.aof")
-# appendfilename appendonly.aof
-
-# The fsync() call tells the Operating System to actually write data on disk
-# instead to wait for more data in the output buffer. Some OS will really flush 
-# data on disk, some other OS will just try to do it ASAP.
-#
-# Redis supports three different modes:
-#
-# no: don't fsync, just let the OS flush the data when it wants. Faster.
-# always: fsync after every write to the append only log . Slow, Safest.
-# everysec: fsync only if one second passed since the last fsync. Compromise.
-#
-# The default is "everysec" that's usually the right compromise between
-# speed and data safety. It's up to you to understand if you can relax this to
-# "no" that will will let the operating system flush the output buffer when
-# it wants, for better performances (but if you can live with the idea of
-# some data loss consider the default persistence mode that's snapshotting),
-# or on the contrary, use "always" that's very slow but a bit safer than
-# everysec.
-#
-# If unsure, use "everysec".
-
-# appendfsync always
 appendfsync everysec
-# appendfsync no
-
-# When the AOF fsync policy is set to always or everysec, and a background
-# saving process (a background save or AOF log background rewriting) is
-# performing a lot of I/O against the disk, in some Linux configurations
-# Redis may block too long on the fsync() call. Note that there is no fix for
-# this currently, as even performing fsync in a different thread will block
-# our synchronous write(2) call.
-#
-# In order to mitigate this problem it's possible to use the following option
-# that will prevent fsync() from being called in the main process while a
-# BGSAVE or BGREWRITEAOF is in progress.
-#
-# This means that while another child is saving the durability of Redis is
-# the same as "appendfsync none", that in pratical terms means that it is
-# possible to lost up to 30 seconds of log in the worst scenario (with the
-# default Linux settings).
-# 
-# If you have latency problems turn this to "yes". Otherwise leave it as
-# "no" that is the safest pick from the point of view of durability.
 no-appendfsync-on-rewrite no
-
-############################### ADVANCED CONFIG ###############################
-
-# Hashes are encoded in a special way (much more memory efficient) when they
-# have at max a given numer of elements, and the biggest element does not
-# exceed a given threshold. You can configure this limits with the following
-# configuration directives.
-hash-max-ziplist-entries 64
-hash-max-ziplist-value 512
-
-# Similarly to hashes, small lists are also encoded in a special way in order
-# to save a lot of space. The special representation is only used when
-# you are under the following limits:
-list-max-ziplist-entries 512
-list-max-ziplist-value 64
-
-# Sets have a special encoding in just one case: when a set is composed
-# of just strings that happens to be integers in radix 10 in the range
-# of 64 bit signed integers.
-# The following configuration setting sets the limit in the size of the
-# set in order to use this special memory saving encoding.
-set-max-intset-entries 512
-
-# Active rehashing uses 1 millisecond every 100 milliseconds of CPU time in
-# order to help rehashing the main Redis hash table (the one mapping top-level
-# keys to values). The hash table implementation redis uses (see dict.c)
-# performs a lazy rehashing: the more operation you run into an hash table
-# that is rhashing, the more rehashing "steps" are performed, so if the
-# server is idle the rehashing is never complete and some more memory is used
-# by the hash table.
-# 
-# The default is to use this millisecond 10 times every second in order to
-# active rehashing the main dictionaries, freeing memory when possible.
-#
-# If unsure:
-# use "activerehashing no" if you have hard latency requirements and it is
-# not a good thing in your environment that Redis can reply form time to time
-# to queries with 2 milliseconds delay.
-#
-# use "activerehashing yes" if you don't have such hard requirements but
-# want to free memory asap when possible.
 activerehashing yes
-
-################################## INCLUDES ###################################
-
-# Include one or more other config files here.  This is useful if you
-# have a standard template that goes to all redis server but also need
-# to customize a few per-server settings.  Include files can include
-# other files, so use this wisely.
-#
-# include /path/to/local.conf
-# include /path/to/other.conf
@@ -23,3 +23,76 @@ start_server [list overrides [list "dir" $server_path "dbfilename" "encodings.rd
 }
 }

+set server_path [tmpdir "server.rdb-startup-test"]
+
+start_server [list overrides [list "dir" $server_path]] {
+    test {Server started empty with non-existing RDB file} {
+        r debug digest
+    } {0000000000000000000000000000000000000000}
+    # Save an RDB file, needed for the next test.
+    r save
+}
+
+start_server [list overrides [list "dir" $server_path]] {
+    test {Server started empty with empty RDB file} {
+        r debug digest
+    } {0000000000000000000000000000000000000000}
+}
+
+# Helper function to start a server and kill it, just to check the error
+# logged.
+set defaults {}
+proc start_server_and_kill_it {overrides code} {
+    upvar defaults defaults srv srv server_path server_path
+    set config [concat $defaults $overrides]
+    set srv [start_server [list overrides $config]]
+    uplevel 1 $code
+    kill_server $srv
+}
+
+# Make the RDB file unreadable
+file attributes [file join $server_path dump.rdb] -permissions 0222
+
+# Detect root account (it is able to read the file even with 002 perm)
+set isroot 0
+catch {
+    open [file join $server_path dump.rdb]
+    set isroot 1
+}
+
+# Now make sure the server aborted with an error
+if {!$isroot} {
+    start_server_and_kill_it [list "dir" $server_path] {
+        test {Server should not start if RDB file can't be open} {
+            wait_for_condition 50 100 {
+                [string match {*Fatal error loading*} \
+                    [exec tail -n1 < [dict get $srv stdout]]]
+            } else {
+                fail "Server started even if RDB was unreadable!"
+            }
+        }
+    }
+}
+
+# Fix permissions of the RDB file.
+file attributes [file join $server_path dump.rdb] -permissions 0666
+
+# Corrupt its CRC64 checksum.
+set filesize [file size [file join $server_path dump.rdb]]
+set fd [open [file join $server_path dump.rdb] r+]
+fconfigure $fd -translation binary
+seek $fd -8 end
+puts -nonewline $fd "foobar00"; # Corrupt the checksum
+close $fd
+
+# Now make sure the server aborted with an error
+start_server_and_kill_it [list "dir" $server_path] {
+    test {Server should not start if RDB is corrupted} {
+        wait_for_condition 50 100 {
+            [string match {*RDB checksum*} \
+                [exec tail -n1 < [dict get $srv stdout]]]
+        } else {
+            fail "Server started even if RDB was corrupted!"
+        }
+    }
+}
@@ -10,21 +10,22 @@ proc stop_bg_complex_data {handle} {
 start_server {tags {"repl"}} {
    start_server {} {

-        set master [srv 0 client]
-        set master_host [srv 0 host]
-        set master_port [srv 0 port]
+        set master [srv -1 client]
+        set master_host [srv -1 host]
+        set master_port [srv -1 port]
+        set slave [srv 0 client]
+
        set load_handle0 [start_bg_complex_data $master_host $master_port 9 100000]
        set load_handle1 [start_bg_complex_data $master_host $master_port 11 100000]
        set load_handle2 [start_bg_complex_data $master_host $master_port 12 100000]

        test {First server should have role slave after SLAVEOF} {
-            r -1 slaveof [srv 0 host] [srv 0 port]
+            $slave slaveof $master_host $master_port
            after 1000
-            s -1 role
+            s 0 role
        } {slave}

        test {Test replication with parallel clients writing in differnet DBs} {
-            lappend slave [srv 0 client]
            after 5000
            stop_bg_complex_data $load_handle0
            stop_bg_complex_data $load_handle1
@@ -37,7 +38,7 @@ start_server {tags {"repl"}} {
            }
            assert {[$master dbsize] > 0}

-            if {[r debug digest] ne [r -1 debug digest]} {
+            if {[$master debug digest] ne [$slave debug digest]} {
                set csv1 [csvdump r]
                set csv2 [csvdump {r -1}]
                set fd [open /tmp/repldump1.txt w]
@@ -408,6 +408,65 @@ for {set j 0} {$j < [llength $argv]} {incr j} {
    }
 }

+proc attach_to_replication_stream {} {
+    set s [socket [srv 0 "host"] [srv 0 "port"]]
+    fconfigure $s -translation binary
+    puts -nonewline $s "SYNC\r\n"
+    flush $s
+
+    # Get the count
+    set count [gets $s]
+    set prefix [string range $count 0 0]
+    if {$prefix ne {$}} {
+        error "attach_to_replication_stream error. Received '$count' as count."
+    }
+    set count [string range $count 1 end]
+
+    # Consume the bulk payload
+    while {$count} {
+        set buf [read $s $count]
+        set count [expr {$count-[string length $buf]}]
+    }
+    return $s
+}
+
+proc read_from_replication_stream {s} {
+    fconfigure $s -blocking 0
+    set attempt 0
+    while {[gets $s count] == -1} {
+        if {[incr attempt] == 10} return ""
+        after 100
+    }
+    fconfigure $s -blocking 1
+
+    # Workaround for Redis 2.6, not always using the new protocol in the
+    # replication channel (this was fixed in >= 2.8).
+    if {[string tolower [lindex [split $count] 0]] eq {select}} {
+        return $count
+    }
+
+    # Return a list of arguments for the command.
+    set count [string range $count 1 end]
+    set res {}
+    for {set j 0} {$j < $count} {incr j} {
+        read $s 1
+        set arg [::redis::redis_bulk_read $s]
+        if {$j == 0} {set arg [string tolower $arg]}
+        lappend res $arg
+    }
+    return $res
+}
+
+proc assert_replication_stream {s patterns} {
+    for {set j 0} {$j < [llength $patterns]} {incr j} {
+        assert_match [lindex $patterns $j] [read_from_replication_stream $s]
+    }
+}
+
+proc close_replication_stream {s} {
+    close $s
+}
+
 # With the parallel test running multiple Redis instances at the same time
 # we need a fast enough computer, otherwise a lot of tests may generate
 # false positives.
@@ -712,4 +712,46 @@ start_server {tags {"basic"}} {
            assert_equal [string range $bin $_start $_end] [r getrange bin $start $end]
        }
    }
+
+    test {Extended SET can detect syntax errors} {
+        set e {}
+        catch {r set foo bar non-existing-option} e
+        set e
+    } {*syntax*}
+
+    test {Extended SET NX option} {
+        r del foo
+        set v1 [r set foo 1 nx]
+        set v2 [r set foo 2 nx]
+        list $v1 $v2 [r get foo]
+    } {OK {} 1}
+
+    test {Extended SET XX option} {
+        r del foo
+        set v1 [r set foo 1 xx]
+        r set foo bar
+        set v2 [r set foo 2 xx]
+        list $v1 $v2 [r get foo]
+    } {{} OK 2}
+
+    test {Extended SET EX option} {
+        r del foo
+        r set foo bar ex 10
+        set ttl [r ttl foo]
+        assert {$ttl <= 10 && $ttl > 5}
+    }
+
+    test {Extended SET PX option} {
+        r del foo
+        r set foo bar px 10000
+        set ttl [r ttl foo]
+        assert {$ttl <= 10 && $ttl > 5}
+    }
+
+    test {Extended SET using multiple options at once} {
+        r set foo val
+        assert {[r set foo bar xx px 10000] eq {OK}}
+        set ttl [r ttl foo]
+        assert {$ttl <= 10 && $ttl > 5}
+    }
 }
@@ -142,6 +142,34 @@ start_server {tags {"expire"}} {
        list $size1 $size2
    } {3 0}

+    test {Redis should lazy expire keys} {
+        r flushdb
+        r debug set-active-expire 0
+        r psetex key1 500 a
+        r psetex key2 500 a
+        r psetex key3 500 a
+        set size1 [r dbsize]
+        # Redis expires random keys ten times every second so we are
+        # fairly sure that all the three keys should be evicted after
+        # one second.
+        after 1000
+        set size2 [r dbsize]
+        r mget key1 key2 key3
+        set size3 [r dbsize]
+        r debug set-active-expire 1
+        list $size1 $size2 $size3
+    } {3 3 0}
+
+    test {EXPIRE should not resurrect keys (issue #1026)} {
+        r debug set-active-expire 0
+        r set foo bar
+        r pexpire foo 500
+        after 1000
+        r expire foo 10
+        r debug set-active-expire 1
+        r exists foo
+    } {0}
+
    test {5 keys in, 5 keys out} {
        r flushdb
        r set a c
@@ -205,7 +205,10 @@ start_server {tags {"multi"}} {
        r select 5
        r multi
        r ping
-        r exec
+        set res [r exec]
+        # Restore original DB
+        r select 9
+        set res
    } {PONG}

    test {WATCH will consider touched keys target of EXPIRE} {
@@ -249,4 +252,58 @@ start_server {tags {"multi"}} {
        r incr x
        r exec
    } {11}
+
+    test {MULTI / EXEC is propagated correctly (single write command)} {
+        set repl [attach_to_replication_stream]
+        r multi
+        r set foo bar
+        r exec
+        assert_replication_stream $repl {
+            {select *}
+            {multi}
+            {set foo bar}
+            {exec}
+        }
+        close_replication_stream $repl
+    }
+
+    test {MULTI / EXEC is propagated correctly (empty transaction)} {
+        set repl [attach_to_replication_stream]
+        r multi
+        r exec
+        r set foo bar
+        assert_replication_stream $repl {
+            {select *}
+            {set foo bar}
+        }
+        close_replication_stream $repl
+    }
+
+    test {MULTI / EXEC is propagated correctly (read-only commands)} {
+        r set foo value1
+        set repl [attach_to_replication_stream]
+        r multi
+        r get foo
+        r exec
+        r set foo value2
+        assert_replication_stream $repl {
+            {select *}
+            {set foo value2}
+        }
+        close_replication_stream $repl
+    }
+
+    test {MULTI / EXEC is propagated correctly (write command, no effect)} {
+        r del bar foo bar
+        set repl [attach_to_replication_stream]
+        r multi
+        r del foo
+        r exec
+        assert_replication_stream $repl {
+            {select *}
+            {multi}
+            {exec}
+        }
+        close_replication_stream $repl
+    }
 }
@@ -15,7 +15,7 @@ start_server {tags {"obuf-limits"}} {
            if {![regexp {omem=([0-9]+)} $c - omem]} break
            if {$omem > 200000} break
        }
-        assert {$omem >= 99000 && $omem < 200000}
+        assert {$omem >= 90000 && $omem < 200000}
        $rd1 close
    }

@@ -281,6 +281,23 @@ start_server {tags {"scripting"}} {
        assert_equal $rand1 $rand2
        assert {$rand2 ne $rand3}
    }
+
+    test {EVAL processes writes from AOF in read-only slaves} {
+        r flushall
+        r config set appendonly yes
+        r eval {redis.call("set","foo","100")} 0
+        r eval {redis.call("incr","foo")} 0
+        r eval {redis.call("incr","foo")} 0
+        wait_for_condition 50 100 {
+            [s aof_rewrite_in_progress] == 0
+        } else {
+            fail "AOF rewrite can't complete after CONFIG SET appendonly yes."
+        }
+        r config set slave-read-only yes
+        r slaveof 127.0.0.1 0
+        r debug loadaof
+        r get foo
+    } {102}
 }

 # Start a new server since the last test in this stanza will kill the
Author	SHA1	Message	Date
antirez	f2f2b4eb9d	Redis 2.6.14	2013-06-20 10:36:40 +02:00
antirez	252f0b6556	Sentinel: parse new INFO replication output correctly. Sentinel was not able to detect slaves when connected to a very recent version of Redis master since a previos non-backward compatible change to INFO broken the parsing of the slaves ip:port INFO output. This fixes issue #1164	2013-06-20 10:24:43 +02:00
antirez	ee92c043fc	Test: regression test for #1163 .	2013-06-19 18:53:17 +02:00
antirez	8e63fd0734	Allow writes from scripts called by AOF loading in read-only slaves. This fixes issue #1163	2013-06-19 18:29:10 +02:00
antirez	a73da27853	Binary safe dump of object content in redisLogObjectDebugInfo().	2013-06-04 15:56:26 +02:00
antirez	feffd5635b	redis-cli: help.h updated.	2013-05-14 11:22:50 +02:00
antirez	2c93d34bde	Test: various issues with the replication-4.tcl test fixed. The test actually worked, but vars for master and slave were inverted and sometimes used incorrectly.	2013-05-08 11:59:47 +02:00
Jiahao Huangandantirez	6251c3e66e	in 32bit machine, popcount don't work with a input string length up to 512 MB, bitcount commant may return negtive integer with string length more than 256 MB	2013-05-08 09:59:35 +02:00
antirez	71b442276f	Added more info about 32 bit builds into README.	2013-05-08 09:48:49 +02:00
antirez	6a947b1d54	CONFIG SET server.masterauth aesthetic change. This is just to make the code exactly like the above instance used for requirepass. No actual change nor the original code violated the Redis coding style.	2013-05-02 17:22:42 +02:00
Michel Martensandantirez	736220b86f	Reset masterauth if an empty string is configured.	2013-05-02 17:22:38 +02:00
charsyamandantirez	b37512a28f	Fix AOF bug: expire could be removed from key on AOF rewrite. There was a race condition in the AOF rewrite code that, with bad enough timing, could cause a volatile key just about to expire to be turned into a non-volatile key. The bug was never reported to cause actualy issues, but was found analytically by an user in the Redis mailing list: https://groups.google.com/forum/?fromgroups=#!topic/redis-db/Kvh2FAGK4Uk This commit fixes issue #1079.	2013-05-02 15:46:59 +02:00
antirez	d96497961c	Redis 2.6.13	2013-04-30 15:15:56 +02:00
antirez	4f38d03254	Sentinel: changes to tilt mode. Tilt mode was too aggressive (not processing INFO output), this resulted in a few problems: 1) Redirections were not followed when in tilt mode. This opened a window to misinform clients about the current master when a Sentinel was in tilt mode and a fail over happened during the time it was not able to update the state. 2) It was possible for a Sentinel exiting tilt mode to detect a false fail over start, if a slave rebooted with a wrong configuration about at the same time. This used to happen since in tilt mode we lose the information that the runid changed (reboot). Now instead the Sentinel in tilt mode will still remove the instance from the list of slaves if it changes state AND runid at the same time. Both are edge conditions but the changes should overall improve the reliability of Sentinel.	2013-04-30 15:09:34 +02:00
antirez	59ff2fe9a4	Sentinel: more sensible delay in master demote after tilt.	2013-04-30 15:09:31 +02:00
antirez	fb0d08e35f	Sentinel: only demote old master into slave under certain conditions. We used to always turn a master into a slave if the DEMOTE flag was set, as this was a resurrecting master instance. However the following race condition is possible for a Sentinel that got partitioned or internal issues (tilt mode), and was not able to refresh the state in the meantime: 1) Sentinel X is running, master is instance "A". 3) "A" fails, sentinels will promote slave "B" as master. 2) Sentinel X goes down because of a network partition. 4) "A" returns available, Sentinels will demote it as a slave. 5) "B" fails, other Sentinels will promote slave "A" as master. 6) At this point Sentinel X comes back. When "X" comes back he thinks that: "B" is the master. "A" is the slave to demote. We want to avoid that Sentinel "X" will demote "A" into a slave. We also want that Sentinel "X" will detect that the conditions changed and will reconfigure itself to monitor the right master. There are two main ways for the Sentinel to reconfigure itself after this event: 1) If "B" is reachable and already configured as a slave by other sentinels, "X" will perform a redirection to "A". 2) If there are not the conditions to demote "A", the fact that "A" reports to be a master will trigger a failover detection in "X", that will end into a reconfiguraiton to monitor "A". However if the Sentinel was not reachable, its state may not be updated, so in case it titled, or was partiitoned from the master instance of the slave to demote, the new implementation waits some time (enough to guarantee we can detect the new INFO, and new DOWN conditions). If after some time still there are not the right condiitons to demote the instance, the DEMOTE flag is cleared.	2013-04-30 15:09:27 +02:00
antirez	395e6a5872	Lua updated to version 5.1.5.	2013-04-29 12:02:04 +02:00
antirez	34a57a5b52	Sentinel: always redirect on master->slave transition. Sentinel redirected to the master if the instance changed runid or it was the first time we got INFO, and a role change was detected from master to slave. While this is a good idea in case of slave->master, since otherwise we could detect a failover without good reasons just after a reboot with a slave with a wrong configuration, in the case of master->slave transition is much better to always perform the redirection for the following reasons: 1) A Sentinel may go down for some time. When it is back online there is no other way to understand there was a failover. 2) Pointing clients to a slave seems to be always the wrong thing to do. 3) There is no good rationale about handling things differently once an instance is rebooted (runid change) in that case.	2013-04-24 11:34:05 +02:00
antirez	967ae8ca67	Config option to turn AOF rewrite incremental fsync on/off.	2013-04-24 10:57:38 +02:00
antirez	c735116c87	AOF: sync data on disk every 32MB when rewriting. This prevents the kernel from putting too much stuff in the output buffers, doing too heavy I/O all at once. So the goal of this commit is to split the disk pressure due to the AOF rewrite process into smaller spikes. Please see issue #1019 for more information.	2013-04-24 10:28:43 +02:00
antirez	8b41e19e47	rio.c: added ability to fdatasync() from time to time while writing.	2013-04-24 10:27:55 +02:00
antirez	cf7b2c329b	Test: fix RDB test checking file permissions. When the test is executed using the root account, setting the permission to 222 does not work as expected, as root can read files with 222 permission. Now we skip the test if root is detected. This fixes issue #1034 and the duplicated #1040 issue. Thanks to Jan-Erik Rediger (@badboy on Github) for finding a way to reproduce the issue.	2013-04-23 14:16:58 +02:00
antirez	bac29c8b7c	Sentinel: turn old master into a slave when it comes back.	2013-04-22 11:26:33 +02:00
antirez	4848af1b9e	Test: split conceptually unrelated comments in RDB test.	2013-04-22 11:25:59 +02:00
antirez	172eac504a	More explicit panic message on out of memory.	2013-04-19 15:11:59 +02:00
antirez	5bf9b53a83	Redis/Jemalloc Gitignore were too aggressive. Redis gitignore was too aggressive since simply broken. Jemalloc gitignore was too agressive because it is conceived to just keep the files that allow to generate all the rest in development environments (so for instance the "configure" file is excluded).	2013-04-18 16:21:32 +02:00
antirez	40861516ac	redis-cli: raise error on bad command line switch. Previously redis-cli never tried to raise an error when an unrecognized switch was encountered, as everything after the initial options is to be transmitted to the server. However this is too liberal, as there are no commands starting with "-". So the new behavior is to produce an error if there is an unrecognized switch starting with "-". This should not break past redis-cli usages but should prevent broken options to be silently discarded. As far the first token not starting with "-" is encountered, all the rest is considered to be part of the command, so you cna still use strings starting with "-" as values, like in: redis-cli --port 6380 set foo --my-value	2013-04-11 13:19:39 +02:00
antirez	92a7b0102c	redis-cli: --latency-history mode implemented.	2013-04-11 13:19:33 +02:00
antirez	9f2e6ac0ed	Test: remove useless statements and comments from test default config.	2013-04-04 14:32:17 +02:00
antirez	63978d338d	Make rio.c comment 80-columns friendly.	2013-04-03 12:41:31 +02:00
antirez	ed2d988192	Throttle BGSAVE attempt on saving error. When a BGSAVE fails, Redis used to flood itself trying to BGSAVE at every next cron call, that is either 10 or 100 times per second depending on configuration and server version. This commit does not allow a new automatic BGSAVE attempt to be performed before a few seconds delay (currently 5). This avoids both the auto-flood problem and filling the disk with logs at a serious rate. The five seconds limit, considering a log entry of 200 bytes, will use less than 4 MB of disk space per day that is reasonable, the sysadmin should notice before of catastrofic events especially since by default Redis will stop serving write queries after the first failed BGSAVE. This fixes issue #849	2013-04-02 14:13:03 +02:00
antirez	7da5980848	Redis 2.6.12	2013-03-29 17:42:39 +01:00
charsyamandantirez	062f60a36e	Support for case unsensitive SET options.	2013-03-29 10:37:33 +01:00
antirez	4243483d3e	Test: Extended SET tests.	2013-03-28 16:45:45 +01:00
antirez	d785413d86	Extended SET command implemented (issue #931 ).	2013-03-28 16:45:37 +01:00
antirez	e8a1a169dd	EXPIRE should not resurrect keys. Issue #1026 .	2013-03-28 12:49:03 +01:00
antirez	dc0f13774b	Test: regression test for issue #1026 .	2013-03-28 12:48:59 +01:00
antirez	47911b8630	Better DEBUG error message when num of arguments is wrong.	2013-03-28 12:48:55 +01:00
antirez	7ff729e192	Test: verify that lazy-expire works.	2013-03-28 12:48:51 +01:00
antirez	02b9a72bae	DEBUG set-active-expire added. We need the ability to disable the activeExpireCycle() (active expired key collection) call for testing purposes.	2013-03-28 12:48:47 +01:00
antirez	1862239136	Test: read_from_replication_stream, workaround for 2.6	2013-03-27 12:03:46 +01:00
antirez	f8791a158b	Test: test replication of MULTI/EXEC.	2013-03-27 11:57:32 +01:00
antirez	26319d08b5	Test: Restore DB back to 9 after testing MULTI/EXEC with DB 5.	2013-03-27 11:57:28 +01:00
antirez	fc78d978de	Test: new functions to capture and analyze the replication stream.	2013-03-27 11:57:23 +01:00
antirez	6cb7860658	Flag PUBLISH as read-only in the command table.	2013-03-27 09:07:43 +01:00
antirez	611dcb56ee	Transactions: propagate MULTI/EXEC only when needed. MULTI/EXEC is now propagated to the AOF / Slaves only once we encounter the first command that is not a read-only one inside the transaction. The old behavior was to always propagate an empty MULTI/EXEC block when the transaction was composed just of read only commands, or even completely empty. This created two problems: 1) It's a bandwidth waste in the replication link and a space waste inside the AOF file. 2) We used to always increment server.dirty to force the propagation of the EXEC command, resulting into triggering RDB saves more often than needed. Note: even read-only commands may also trigger writes that will be propagated, when we access a key that is found expired and Redis will synthesize a DEL operation. However there is no need for this to stay inside the transaction itself, but only to be ordered. So for instance something like: MULTI GET foo SET key zap EXEC May be propagated into: DEL foo MULTI SET key zap EXEC While the DEL is outside the transaction, the commands are delivered in the right order and it is not possible for other commands to be inserted between DEL and MULTI.	2013-03-27 09:07:23 +01:00
antirez	889b017403	Transactions: use discardTransaction() in EXEC implementation.	2013-03-27 09:07:18 +01:00
antirez	d1369c3d9f	Transactions: use the propagate() API to propagate MULTI. The behavior is the same, but the code is now cleaner and uses the proper interface instead of dealing directly with AOF/replication functions.	2013-03-27 09:06:59 +01:00
antirez	6818905406	Allow SELECT while loading the DB. Fixes issue #1024.	2013-03-26 13:59:00 +01:00
NanXiaoandantirez	ef1cf15c14	Update config.c Fix bug in configGetCommand function: get correct masterauth value.	2013-03-25 19:32:24 +01:00
antirez	dee8d84270	Test: obuf-limits test false positive removed. Fixes #621.	2013-03-25 11:56:08 +01:00
antirez	5576a28977	redis-cli --stat, stolen from redis-tools. Redis-tools is a connection of tools no longer mantained that was intented as a way to economically make sense of Redis in the pre-vmware sponsorship era. However there was a nice redis-stat utility, this commit imports one of the functionalities of this tool here in redis-cli as it seems to be pretty useful. Usage: redis-cli --stat The output is similar to vmstat in the format, but with Redis specific stuff of course. From the point of view of the monitored instance, only INFO is used in order to grab data.	2013-03-25 11:52:32 +01:00
antirez	4f8b18f3dd	Replication: master_link_down_since_seconds initial value should be huge. server.repl_down_since used to be initialized to the current time at startup. This is wrong since the replication never started. Clients testing this filed to check if data is uptodate should never believe data is recent if we never ever connected to our master.	2013-03-13 12:55:06 +01:00
antirez	c64255990e	Test: make sure broken RDB checksum is detected.	2013-03-13 11:15:33 +01:00
antirez	ba3c5494d8	Test: more RDB loading checks. A test for issue #1001 is included.	2013-03-13 10:09:21 +01:00
antirez	684b61505a	Test: check that Redis starts empty without an RDB file.	2013-03-13 10:09:16 +01:00
antirez	189e865c51	rdbLoad(): rework code to save vertical space.	2013-03-13 10:09:09 +01:00
Damian Janowskiandantirez	b9f8c2a5b0	Abort when opening the RDB file results in an error other than ENOENT. This fixes cases where the RDB file does exist but can't be accessed for any reason. For instance, when the Redis process doesn't have enough permissions on the file.	2013-03-13 10:08:47 +01:00
antirez	18d16f8592	Set default for stop_writes_on_bgsave_err in initServerConfig(). It was placed for error in initServer() that's called after the configuation is already loaded, causing issue #1000.	2013-03-12 18:36:07 +01:00
antirez	3a00520e5f	Add a missing bugfix entry in 2.6.11 release notes.	2013-03-12 17:55:23 +01:00
antirez	f93d9929d8	redis-cli --bigkeys: don't crash with empty DBs.	2013-03-12 09:57:49 +01:00