antirez [Wed, 7 Mar 2012 17:02:26 +0000 (18:02 +0100)]
By default Redis refuses writes with an error if the latest BGSAVE failed (and at least one save point is configured). However people having good monitoring systems may prefer a server that continues to work, since they are notified that there are problems by their monitoring systems. This commit implements the ability to turn the feature on or off via redis.conf and CONFIG SET.
antirez [Wed, 7 Mar 2012 12:05:46 +0000 (13:05 +0100)]
Refuse writes if can't persist on disk.
Redis now refuses accepting write queries if RDB persistence is
configured, but RDB snapshots can't be generated for some reason.
The status of the latest background save operation is now exposed
in the INFO output as well. This fixes issue #90.
antirez [Wed, 7 Mar 2012 11:12:15 +0000 (12:12 +0100)]
Better MONITOR output, now includes client ip:port or the lua string if the command was executed by the scripting engine.
antirez [Wed, 7 Mar 2012 10:30:30 +0000 (11:30 +0100)]
anetPeerToString() automatically populates ip/port with something that may be provided to the user as output in case of errors.
antirez [Wed, 7 Mar 2012 09:38:01 +0000 (10:38 +0100)]
TIME command.
antirez [Wed, 29 Feb 2012 16:43:03 +0000 (17:43 +0100)]
redis-cli: CSV output added, used for the --slave mode.
antirez [Wed, 29 Feb 2012 16:10:21 +0000 (17:10 +0100)]
Initial implementation of redis-cli --slave support.
antirez [Wed, 29 Feb 2012 15:33:54 +0000 (16:33 +0100)]
Ping the slave using the standard protocol instead of the inline one.
antirez [Wed, 29 Feb 2012 13:41:57 +0000 (14:41 +0100)]
Better implementation for BRPOP/BLPOP in the non blocking case.
antirez [Wed, 29 Feb 2012 12:38:30 +0000 (13:38 +0100)]
lpush arguments vector rewrite modified for more speed and to memory leak removal.
antirez [Tue, 28 Feb 2012 23:54:52 +0000 (00:54 +0100)]
Version bumped to 2.9.5
antirez [Tue, 28 Feb 2012 23:46:50 +0000 (00:46 +0100)]
Better system for additional commands replication.
The new code uses a more generic data structure to describe redis operations.
The new design allows for multiple alsoPropagate() calls within the scope of a
single command, that is useful in different contexts. For instance there
when there are multiple clients doing BRPOPLPUSH against the same list,
and a variadic LPUSH is performed against this list, the blocked clients
will both be served, and we should correctly replicate multiple LPUSH
commands after the replication of the current command.
antirez [Tue, 28 Feb 2012 17:03:08 +0000 (18:03 +0100)]
Added a new API to replicate an additional command after the replication of the currently executed command, in order to propagte the LPUSH originating from RPOPLPUSH and indirectly by BRPOPLPUSH.
antirez [Tue, 28 Feb 2012 15:20:41 +0000 (16:20 +0100)]
propagate() prototype added to redis.h
antirez [Tue, 28 Feb 2012 15:17:55 +0000 (16:17 +0100)]
Var renamed into pushGenericCommand() to better reflect what it means.
antirez [Tue, 28 Feb 2012 15:17:00 +0000 (16:17 +0100)]
Added command propagation API.
Salvatore Sanfilippo [Sun, 26 Feb 2012 15:18:43 +0000 (07:18 -0800)]
Merge pull request #362 from fawek/mac-build-fix
Fix the build on Mac OS X, where features.h is non-existent
Jakub Wieczorek [Sun, 26 Feb 2012 14:10:42 +0000 (15:10 +0100)]
Fix the build on Mac OS X, where features.h is non-existent
Premysl Hruby [Sun, 26 Feb 2012 08:59:32 +0000 (09:59 +0100)]
use gcc sync builtins for memory counter when possible
Premysl Hruby [Sun, 26 Feb 2012 09:01:27 +0000 (10:01 +0100)]
fix redis-benchmark memory leak
antirez [Fri, 24 Feb 2012 14:46:25 +0000 (15:46 +0100)]
lua_cmsgpack.c added
antirez [Fri, 24 Feb 2012 14:45:16 +0000 (15:45 +0100)]
Lua_cmsgpack added to Redis scripting.
antirez [Thu, 23 Feb 2012 14:39:26 +0000 (15:39 +0100)]
redis-benchmark: a few fixes to pipelining implementation.
antirez [Thu, 23 Feb 2012 14:02:43 +0000 (15:02 +0100)]
redis-benchmark: Initial support for pipelining.
antirez [Wed, 22 Feb 2012 16:44:18 +0000 (17:44 +0100)]
unstable version called 2.9.4
antirez [Wed, 22 Feb 2012 16:22:40 +0000 (17:22 +0100)]
The universe is 80 columns wide, accept it.
Salvatore Sanfilippo [Wed, 22 Feb 2012 16:19:54 +0000 (08:19 -0800)]
Merge pull request #304 from bradvoth/unstable
Simple install script changes
antirez [Wed, 22 Feb 2012 15:07:06 +0000 (16:07 +0100)]
Fix for issue #306, thanks to tchajed (on github) for the pull request. The original patch was reworked a bit.
antirez [Wed, 22 Feb 2012 14:16:11 +0000 (15:16 +0100)]
Redis-cli should accept help command even if there is no connection to the server.
antirez [Wed, 22 Feb 2012 10:25:30 +0000 (11:25 +0100)]
even less false positive on obuf-limits test.
antirez [Wed, 22 Feb 2012 08:52:29 +0000 (09:52 +0100)]
Merge remote-tracking branch 'origin/unstable' into unstable
antirez [Wed, 22 Feb 2012 08:52:10 +0000 (09:52 +0100)]
zzlIsInRange() now is capable of handling empty sorted sets that may end inside the data set when loading very old RDB files produced by early-stage versions of Redis.
Salvatore Sanfilippo [Tue, 21 Feb 2012 18:29:55 +0000 (10:29 -0800)]
Merge pull request #347 from pietern/unstable-segv
Force SIGSEGV without HAVE_BACKTRACE (unstable)
Pieter Noordhuis [Tue, 21 Feb 2012 18:11:20 +0000 (10:11 -0800)]
Solaris has backtrace(3)
Pieter Noordhuis [Tue, 21 Feb 2012 18:05:46 +0000 (10:05 -0800)]
Also force SIGSEGV without HAVE_BACKTRACE
antirez [Tue, 21 Feb 2012 17:25:49 +0000 (18:25 +0100)]
Fixed undefined behavior in *INCR style functions overflow detection. Sorry clang!
antirez [Tue, 21 Feb 2012 09:06:04 +0000 (10:06 +0100)]
bzero -> memset
antirez [Sun, 19 Feb 2012 19:49:20 +0000 (20:49 +0100)]
Fix 32bit build of hiredis
antirez [Sun, 19 Feb 2012 19:33:59 +0000 (20:33 +0100)]
Applied a few modifications to hiredis to tune it for speed (redis-benchmark) and ability to read more deeply nested replies (redis-cli).
antirez [Sun, 19 Feb 2012 19:26:36 +0000 (20:26 +0100)]
The hiredis lib shipped with Redis was updated to latest version.
Pieter Noordhuis [Sat, 14 Jan 2012 01:49:16 +0000 (17:49 -0800)]
Don't expire keys when loading an RDB after a SYNC
The cron is responsible for expiring keys. When keys are expired at
load time, it is possible that the snapshot of a master node gets
modified. This can in turn lead to inconsistencies in the data set.
A more concrete example of this behavior follows. A user reported a
slave that would show an monotonically increase input buffer length,
shortly after completing a SYNC. Also, `INFO` output showed a single
blocked client, which could only be the master link. Investigation
showed that indeed the `BRPOP` command was fed by the master. This
command can only end up in the stream of write operations when it did
NOT block, and effectively executed `RPOP`. However, when the key
involved in the `BRPOP` is expired BEFORE the command is executed, the
client executing it will block. The client in this case, is the master
link.
antirez [Thu, 16 Feb 2012 15:13:40 +0000 (16:13 +0100)]
Issue #330 regression test.
antirez [Thu, 16 Feb 2012 15:09:08 +0000 (16:09 +0100)]
Now HINCRBY can detect overflows too. Fix for issue #330.
antirez [Thu, 16 Feb 2012 11:20:56 +0000 (12:20 +0100)]
Return ASAP from checkClientOutputBufferLimits() if c->reply_bytes is zero.
antirez [Thu, 16 Feb 2012 08:40:27 +0000 (09:40 +0100)]
Merge branch 'issue327' into unstable
antirez [Wed, 15 Feb 2012 11:21:04 +0000 (12:21 +0100)]
Add stdint.h in endianconv.h to fix issue #336.
antirez [Tue, 14 Feb 2012 18:57:31 +0000 (19:57 +0100)]
redis-check-aof is now large files safe also on 32 bit systems.
antirez [Tue, 14 Feb 2012 15:15:20 +0000 (16:15 +0100)]
add -f flag to cp when installing, otherwise stopping the server is
needed when installing a new Redis version. Thanks to Scott Kevill.
Fixes issue #335.
antirez [Tue, 14 Feb 2012 15:11:46 +0000 (16:11 +0100)]
endian.c/h -> endianconv.c/h to avoid issues with broken libraries search paths.
antirez [Tue, 14 Feb 2012 15:02:04 +0000 (16:02 +0100)]
Merge remote-tracking branch 'origin/unstable' into unstable
antirez [Tue, 14 Feb 2012 14:35:50 +0000 (15:35 +0100)]
intset.c endianess fixes.
Salvatore Sanfilippo [Mon, 13 Feb 2012 23:05:59 +0000 (15:05 -0800)]
Merge pull request #334 from lsbardel/quantredis
added lua struct c extension
lsbardel [Mon, 13 Feb 2012 21:05:21 +0000 (21:05 +0000)]
added lua struct c extension
antirez [Thu, 9 Feb 2012 16:09:01 +0000 (17:09 +0100)]
ziplist.c endianess fixes, chapter 5.
antirez [Thu, 9 Feb 2012 15:36:25 +0000 (16:36 +0100)]
ziplist.c endianess fixes, chapter 4.
antirez [Thu, 9 Feb 2012 15:28:35 +0000 (16:28 +0100)]
ziplist.c endianess fixes, chapter 3.
antirez [Wed, 8 Feb 2012 22:20:39 +0000 (23:20 +0100)]
more ziplist.c endianess fixes
antirez [Wed, 8 Feb 2012 21:59:35 +0000 (22:59 +0100)]
ziplist.c fixes for bigendian
antirez [Wed, 8 Feb 2012 21:24:59 +0000 (22:24 +0100)]
A few small BSD related fixes.
antirez [Tue, 7 Feb 2012 23:20:46 +0000 (00:20 +0100)]
more practical maxmemory+slaves hint in redis.conf
antirez [Tue, 7 Feb 2012 23:17:27 +0000 (00:17 +0100)]
redis.conf updated with new maxmemory semantics
antirez [Tue, 7 Feb 2012 23:10:20 +0000 (00:10 +0100)]
debugging messages removed from freeMemoryIfNeeded()
antirez [Tue, 7 Feb 2012 16:41:31 +0000 (17:41 +0100)]
Fixes to c->reply_bytes computation, and debug messages to closely study the behavior of memory pressure + slaves + maxmemory + blocked slaves.
antirez [Tue, 7 Feb 2012 14:08:38 +0000 (15:08 +0100)]
Fixes to 2.6 release notes file
antirez [Tue, 7 Feb 2012 12:05:36 +0000 (13:05 +0100)]
Precision of getClientOutputBufferMemoryUsage() greatily improved, see issue #327 for more information.
antirez [Mon, 6 Feb 2012 15:56:42 +0000 (16:56 +0100)]
freeMemoryIfNeeded() minor refactoring
antirez [Mon, 6 Feb 2012 15:35:43 +0000 (16:35 +0100)]
Also remove size of AOF buffers from used memory when doing the math for freeMemoryIfNeeded()
antirez [Sun, 5 Feb 2012 10:08:01 +0000 (11:08 +0100)]
A first (work in progress) release notes for 2.6
antirez [Sun, 5 Feb 2012 08:38:41 +0000 (09:38 +0100)]
INSTALL now redirects the user to README
antirez [Sun, 5 Feb 2012 08:37:08 +0000 (09:37 +0100)]
Redis Manifesto moved from src to root dir
antirez [Sat, 4 Feb 2012 13:05:54 +0000 (14:05 +0100)]
This fixes issue #327, is a very complex fix (unfortunately), details:
1) sendReplyToClient() now no longer stops transferring data to a single
client in the case we are out of memory (maxmemory-wise).
2) in processCommand() the idea of we being out of memory is no longer
the naive zmalloc_used_memory() > server.maxmemory. To say if we can
accept or not write queries is up to the return value of
freeMemoryIfNeeded(), that has full control about that.
3) freeMemoryIfNeeded() now does its math without considering output
buffers size. But at the same time it can't let the output buffers to
put us too much outside the max memory limit, so at the same time it
makes sure there is enough effort into delivering the output buffers to
the slaves, calling the write handler directly.
This three changes are the result of many tests, I found (partially
empirically) that is the best way to address the problem, but maybe
we'll find better solutions in the future.
antirez [Sat, 4 Feb 2012 07:58:37 +0000 (08:58 +0100)]
Use less memory when emitting the protocol, by using more shared objects for commonly emitted parts of the protocol.
antirez [Thu, 2 Feb 2012 15:30:52 +0000 (16:30 +0100)]
Now Lua scripts dispatch Redis commands properly calling the call() function. In order to make this possible call() was improved with a new flags argument that controls how the Redis command is executed.
antirez [Thu, 2 Feb 2012 09:17:16 +0000 (10:17 +0100)]
Set a 3.5 GB maxmemory limit with noeviction policy if a 32 bit instance without user-provided memory limits is detected.
antirez [Thu, 2 Feb 2012 09:02:40 +0000 (10:02 +0100)]
Added a server.arch_bits field instead of computing it at runtime for INFO.
antirez [Wed, 1 Feb 2012 20:47:41 +0000 (21:47 +0100)]
Only incremnet stats for key miss/hit when the key is semantically accessed in read-only.
antirez [Wed, 1 Feb 2012 16:49:03 +0000 (17:49 +0100)]
Added tests checking ability of the scripting engine to reorder the output of commands with a random output regarding signle elements position in the multi bulk reply.
antirez [Wed, 1 Feb 2012 16:37:48 +0000 (17:37 +0100)]
A few SORT tests made more resistant to false negatives resulitng from poor randomization of Redis hash function with one byte inputs.
antirez [Wed, 1 Feb 2012 16:17:52 +0000 (17:17 +0100)]
New SORT tests checking the new more deterministic behavior of SORT sorting algorithm.
antirez [Wed, 1 Feb 2012 16:05:45 +0000 (17:05 +0100)]
Make SORT BY <constant> STORE ... to always produce the same output by force sorting, so that we have deterministic replication of this command.
antirez [Wed, 1 Feb 2012 14:22:28 +0000 (15:22 +0100)]
SORT is now more deterministic: does not accept to compare by score items that have scores not representing a valid double. Also items with the same score are compared lexycographically. At the same time the scripting side introduced the ability to sort the output of SORT when sort uses the BY <constant> optimization, resulting in no specific ordering. Since in this case the user may use GET, and the result of GET can be null, converted into false as Lua data type, this commit also introduces the ability to sort Lua tables containining false, only if the first (faster) attempt at using just table.sort with a single argument fails.
antirez [Tue, 31 Jan 2012 15:09:21 +0000 (16:09 +0100)]
Order output of commands returning random arrays using table.sort when called from Lua, partially fixing issue #165. The issue is yet not completely fixed since we can't add the REDIS_CMD_SORT_FOR_SCRIPT flag in SORT currently, both because it may contain NULLs and because it is not cool to re-sort everything at every call when instead this should be sorted only if BY <constant> is used.
antirez [Tue, 31 Jan 2012 10:43:32 +0000 (11:43 +0100)]
Fixed redis-benchmark --help output typo
antirez [Tue, 31 Jan 2012 09:35:52 +0000 (10:35 +0100)]
64 bit instances are no longer limited to have at max 2^32-1 elements in lists.
antirez [Mon, 30 Jan 2012 20:08:10 +0000 (21:08 +0100)]
minimal change to obuf-limits.tcl test to make sure there are no false positives with 32bit instances as well.
antirez [Mon, 30 Jan 2012 09:40:28 +0000 (10:40 +0100)]
Merge remote-tracking branch 'origin/unstable' into unstable
Salvatore Sanfilippo [Mon, 30 Jan 2012 09:40:17 +0000 (01:40 -0800)]
Merge pull request #319 from fawek/lua-error-location
Lua reports line numbers off by one in error messages
antirez [Mon, 30 Jan 2012 09:27:50 +0000 (10:27 +0100)]
setKey(): call the higher level wrapper setModifiedKey() instead of touchWatchedKey() even if currently they are exactly the same.
Salvatore Sanfilippo [Mon, 30 Jan 2012 09:25:34 +0000 (01:25 -0800)]
Merge pull request #321 from mkwiatkowski/ticket227
SORT with STORE removes key if result is empty. This fixes issue #227.
Michal Kwiatkowski [Mon, 30 Jan 2012 06:36:49 +0000 (07:36 +0100)]
SORT with STORE removes key if result is empty. This fixes issue #227.
Jakub Wieczorek [Sun, 29 Jan 2012 13:53:49 +0000 (14:53 +0100)]
Lua reports line numbers off by one in error messages
antirez [Thu, 26 Jan 2012 15:45:08 +0000 (16:45 +0100)]
false positive in expire tests mitigated with a sleep, but other solutions exist if needed later.
antirez [Thu, 26 Jan 2012 15:08:24 +0000 (16:08 +0100)]
Less false positives for obuf-limits.tcl tests
antirez [Wed, 25 Jan 2012 17:34:56 +0000 (18:34 +0100)]
Added test for client output buffer limit (soft limit).
antirez [Wed, 25 Jan 2012 17:11:04 +0000 (18:11 +0100)]
Added test for client output buffer limit (hard limit).
antirez [Wed, 25 Jan 2012 17:07:56 +0000 (18:07 +0100)]
Fixed typo in getClientLimitClassByName()
antirez [Wed, 25 Jan 2012 15:59:58 +0000 (16:59 +0100)]
Merge branch 'unstable' into limits
antirez [Wed, 25 Jan 2012 15:59:32 +0000 (16:59 +0100)]
Fixed another possible bug in cluster.c found by clang --analyze.
antirez [Wed, 25 Jan 2012 15:59:32 +0000 (16:59 +0100)]
Fixed another possible bug in cluster.c found by clang --analyze.
antirez [Wed, 25 Jan 2012 15:46:53 +0000 (16:46 +0100)]
Merge branch 'unstable' into limits