]>
git.saurik.com Git - redis.git/log
Pieter Noordhuis [Fri, 14 May 2010 15:49:39 +0000 (17:49 +0200)]
check for memory leaks before killing a server
Pieter Noordhuis [Fri, 14 May 2010 15:45:27 +0000 (17:45 +0200)]
extract code to kill a server to a separate proc
Pieter Noordhuis [Fri, 14 May 2010 15:42:07 +0000 (17:42 +0200)]
start servers on different ports to prevent conflicts
Pieter Noordhuis [Fri, 14 May 2010 15:38:12 +0000 (17:38 +0200)]
use DEBUG DIGEST in new test suite
Pieter Noordhuis [Fri, 14 May 2010 15:31:11 +0000 (17:31 +0200)]
split test suite into multiple files; runs redis-server in isolation
antirez [Fri, 14 May 2010 14:41:24 +0000 (16:41 +0200)]
use DEBUG DIGEST in the test instead of a function that was doing a similar work, but in a much slower and buggy way
Bruno Deferrari [Fri, 14 May 2010 14:04:01 +0000 (11:04 -0300)]
If command is a shutdown, ignore errors on reply
antirez [Fri, 14 May 2010 11:41:57 +0000 (13:41 +0200)]
DEBUG DIGEST implemented, in order to improve the ability to test persistence and replication consistency
antirez [Thu, 13 May 2010 12:30:36 +0000 (14:30 +0200)]
makefile deps updated
antirez [Thu, 13 May 2010 12:24:41 +0000 (14:24 +0200)]
conflicts resolved
Pieter Noordhuis [Thu, 13 May 2010 12:15:06 +0000 (14:15 +0200)]
feed SETEX as SET and EXPIREAT to AOF
antirez [Thu, 13 May 2010 11:36:42 +0000 (13:36 +0200)]
very strong speedup in saving time performance when there are many integers in the dataset. Instead of decoding the object before to pass them to the rdbSaveObject layer we check asap if the object is integer encoded and can be written on disk as an integer.
antirez [Thu, 13 May 2010 09:53:56 +0000 (11:53 +0200)]
include limits.h otherwise no double precison macros
antirez [Thu, 13 May 2010 08:47:06 +0000 (10:47 +0200)]
explicitly checks with ifdefs if our floating point and long long assumptions are verified
antirez [Thu, 13 May 2010 08:28:00 +0000 (10:28 +0200)]
Yet another version of the double saving code, with comments explaining what's happening there
antirez [Wed, 12 May 2010 19:51:48 +0000 (21:51 +0200)]
added overflow check in the double -> long long conversion trick to avoid integer overflows. I think this was not needed in practical terms, but it is safer
antirez [Wed, 12 May 2010 15:22:39 +0000 (17:22 +0200)]
use withscores when performing the dataset digest
antirez [Wed, 12 May 2010 13:07:39 +0000 (15:07 +0200)]
If a float can be casted to a long long without rounding loss, we can use the integer conversion function to write the score on disk. This is a seriuous speedup
antirez [Wed, 12 May 2010 09:56:12 +0000 (11:56 +0200)]
fixed compilation warnings in the AOF sanity check tool
antirez [Wed, 12 May 2010 09:51:01 +0000 (11:51 +0200)]
Merge branch 'vm-speedup'
Pieter Noordhuis [Tue, 11 May 2010 20:46:24 +0000 (22:46 +0200)]
fix to return error when calling INCR on a non-string type
antirez [Tue, 11 May 2010 14:12:22 +0000 (16:12 +0200)]
load objects encoded from disk directly without useless conversion
antirez [Tue, 11 May 2010 12:30:56 +0000 (14:30 +0200)]
fixed a problem leading to crashes, as keys can't be currently specially encoded, so we can't encode integers at object loading time... For now this can be fixed passing a few flags, or later can be fixed allowing encoded keys as well
antirez [Tue, 11 May 2010 12:13:53 +0000 (14:13 +0200)]
long long to string conversion speedup applied in other places as well. Still the code has bugs, fixing right now...
antirez [Tue, 11 May 2010 10:13:37 +0000 (12:13 +0200)]
hand written code to turn a long long into a string -> very big speed win
antirez [Tue, 11 May 2010 09:15:36 +0000 (11:15 +0200)]
added specialized function to compare string objects for perfect match that is optimized for this task
antirez [Tue, 11 May 2010 09:02:24 +0000 (11:02 +0200)]
better use of encoding inforamtion in dictEncObjKeyCompare
antirez [Mon, 10 May 2010 18:56:54 +0000 (20:56 +0200)]
CONFIG now can change appendfsync policy at run time
antirez [Mon, 10 May 2010 16:50:55 +0000 (18:50 +0200)]
CONFIG command now supports hot modification of RDB saving parameters.
antirez [Mon, 10 May 2010 13:26:01 +0000 (15:26 +0200)]
while loading the rdb file don't add the key to the dictionary at all if it's already expired, instead of removing it just after the insertion.
antirez [Mon, 10 May 2010 13:09:25 +0000 (15:09 +0200)]
Merge branch 'check-aof' of git://github.com/pietern/redis
antirez [Sat, 8 May 2010 01:04:00 +0000 (03:04 +0200)]
minor changes to improve code readability
antirez [Sat, 8 May 2010 00:38:59 +0000 (02:38 +0200)]
swap objects out directly while loading an RDB file if we detect we can't stay in the vm max memory limits anyway
Pieter Noordhuis [Fri, 7 May 2010 14:48:43 +0000 (16:48 +0200)]
change command names no longer used to zunion/zinter
antirez [Fri, 7 May 2010 14:33:47 +0000 (16:33 +0200)]
DEBUG POPULATE command for fast creation of test databases
Pieter Noordhuis [Fri, 7 May 2010 12:36:59 +0000 (14:36 +0200)]
update TODO
Pieter Noordhuis [Fri, 7 May 2010 12:33:34 +0000 (14:33 +0200)]
swap arguments in blockClientOnSwappedKeys to be consistent
Pieter Noordhuis [Fri, 7 May 2010 12:17:10 +0000 (14:17 +0200)]
added function that preloads all keys needed to execute a MULTI/EXEC block
Pieter Noordhuis [Fri, 7 May 2010 10:59:34 +0000 (12:59 +0200)]
add sanity check to zunionInterBlockClientOnSwappedKeys, as the number of keys used is provided as argument to the function
Pieter Noordhuis [Fri, 7 May 2010 10:58:44 +0000 (12:58 +0200)]
make prototype of custom function to preload keys from the vm match the prototype of waitForMultipleSwappedKeys
Pieter Noordhuis [Fri, 7 May 2010 10:45:27 +0000 (12:45 +0200)]
extract preloading of multiple keys according to the command prototype to a separate function
Pieter Noordhuis [Fri, 7 May 2010 09:55:12 +0000 (11:55 +0200)]
make append only filename configurable
Pieter Noordhuis [Fri, 7 May 2010 09:40:26 +0000 (11:40 +0200)]
don't load value from VM for EXISTS
antirez [Fri, 7 May 2010 07:36:37 +0000 (09:36 +0200)]
swap file name pid expansion removed. Not suited for mission critical software...
antirez [Fri, 7 May 2010 07:32:26 +0000 (09:32 +0200)]
Swap file is now locked
antirez [Thu, 6 May 2010 21:19:46 +0000 (23:19 +0200)]
Merge branch 'master' into aof-speedup
Pieter Noordhuis [Thu, 6 May 2010 20:00:04 +0000 (22:00 +0200)]
log error and quit when the AOF contains an unfinished MULTI
Pieter Noordhuis [Thu, 6 May 2010 20:00:04 +0000 (22:00 +0200)]
log error and quit when the AOF contains an unfinished MULTI
Pieter Noordhuis [Thu, 6 May 2010 19:06:09 +0000 (21:06 +0200)]
Merge branch 'master' into check-aof
Pieter Noordhuis [Thu, 6 May 2010 18:35:00 +0000 (20:35 +0200)]
hincrby should report an error when called against a hash key that doesn't contain an integer
antirez [Thu, 6 May 2010 18:16:14 +0000 (20:16 +0200)]
AOF writes are now accumulated into a buffer and flushed into disk just before re-entering the event loop. A lot less writes but still this guarantees that AOF is written before the client gets a positive reply about a write operation, as no reply is trasnmitted before re-entering into the event loop.
antirez [Thu, 6 May 2010 16:04:51 +0000 (18:04 +0200)]
clarified a few messages in redis.conf
Pieter Noordhuis [Wed, 5 May 2010 14:38:50 +0000 (16:38 +0200)]
ask for confirmation before AOF is truncated
Pieter Noordhuis [Wed, 5 May 2010 12:06:55 +0000 (14:06 +0200)]
str can be free'd outside readString
Pieter Noordhuis [Wed, 5 May 2010 12:02:04 +0000 (14:02 +0200)]
moved argument parsing around
Pieter Noordhuis [Wed, 5 May 2010 11:47:17 +0000 (13:47 +0200)]
ignore redis-check-aof binary
Pieter Noordhuis [Wed, 5 May 2010 11:46:37 +0000 (13:46 +0200)]
allow AOF to be fixed by truncating to the portion of the file that is valid
Pieter Noordhuis [Wed, 5 May 2010 11:36:29 +0000 (13:36 +0200)]
tool to check if AOF is valid
antirez [Sun, 2 May 2010 13:11:05 +0000 (15:11 +0200)]
included fmacros.h in linenose.c to avoid compilation warnings on Linux
antirez [Sun, 2 May 2010 13:08:06 +0000 (15:08 +0200)]
compilation fix for mac os x
antirez [Sun, 2 May 2010 13:06:06 +0000 (15:06 +0200)]
Merge branch 'master' of git@github.com:antirez/redis
antirez [Sun, 2 May 2010 13:05:34 +0000 (15:05 +0200)]
On Linux now fdatasync() is used insetad of fsync() in order to flush the AOF file kernel buffers
antirez [Fri, 30 Apr 2010 09:52:36 +0000 (11:52 +0200)]
More tests for APPEND and tests for SUBSTR
antirez [Fri, 30 Apr 2010 07:21:48 +0000 (09:21 +0200)]
linenoise.c updated, now redis-cli can be used in a pipe
antirez [Thu, 29 Apr 2010 18:13:23 +0000 (20:13 +0200)]
redis-cli minor fix (less segfault is better)
antirez [Thu, 29 Apr 2010 16:43:51 +0000 (18:43 +0200)]
New MONITOR output format with timestamp, every command in a single line, string representations
antirez [Thu, 29 Apr 2010 16:07:35 +0000 (18:07 +0200)]
redis-cli INFO output format is now raw again
antirez [Thu, 29 Apr 2010 13:39:11 +0000 (15:39 +0200)]
Added more information about slave election in Redis Cluster alternative doc
antirez [Thu, 29 Apr 2010 09:00:19 +0000 (11:00 +0200)]
Redis cluster version 2
antirez [Tue, 27 Apr 2010 16:06:52 +0000 (18:06 +0200)]
Fixed a redis-cli bug, was using free instead of zfree call
antirez [Tue, 27 Apr 2010 14:58:08 +0000 (16:58 +0200)]
AOF is now rewritten on slave after SYNC with master. Thanks to @_km for finding this bug and any others'
antirez [Tue, 27 Apr 2010 14:21:15 +0000 (16:21 +0200)]
redis-cli is now using only the new protocol
antirez [Tue, 27 Apr 2010 14:07:31 +0000 (16:07 +0200)]
Minimal support for subscribe/psubscribe in redis-cli
antirez [Mon, 26 Apr 2010 17:56:16 +0000 (19:56 +0200)]
don't output the newline when stdout is not a tty
antirez [Mon, 26 Apr 2010 16:54:55 +0000 (18:54 +0200)]
redis-cli now is able to also output the string representation instead of the raw string. Much better for debugging
antirez [Mon, 26 Apr 2010 16:39:39 +0000 (18:39 +0200)]
Initial support for quoted strings in redis-cli
antirez [Fri, 23 Apr 2010 17:07:33 +0000 (19:07 +0200)]
SETEX implemented
antirez [Fri, 23 Apr 2010 13:36:00 +0000 (15:36 +0200)]
Pub/Sub API change: now messages received via pattern matching have a different message type and an additional field representing the original pattern the message matched
antirez [Thu, 22 Apr 2010 18:24:00 +0000 (20:24 +0200)]
typo fixed, reloaded
antirez [Thu, 22 Apr 2010 18:20:41 +0000 (20:20 +0200)]
typo fixed
antirez [Thu, 22 Apr 2010 17:46:39 +0000 (19:46 +0200)]
REDIS-CLUSTER doc updated
antirez [Thu, 22 Apr 2010 16:04:50 +0000 (18:04 +0200)]
Virtual memory design document removed, no longer needed as we have a full specification and implementation
antirez [Thu, 22 Apr 2010 13:09:07 +0000 (15:09 +0200)]
new units for bytes specification
antirez [Thu, 22 Apr 2010 10:05:51 +0000 (12:05 +0200)]
Now in redis.conf it is possible to specify units where appropriate instead of amounts of bytes, like 2Gi or 4M and so forth
antirez [Wed, 21 Apr 2010 09:31:29 +0000 (11:31 +0200)]
binary safe keys ready implementation of RANDOMKEYS
antirez [Wed, 21 Apr 2010 08:31:48 +0000 (10:31 +0200)]
Now that's the right 1.3.10
antirez [Wed, 21 Apr 2010 08:31:16 +0000 (10:31 +0200)]
Revert "fsync always now uses O_DIRECT on Linux"
This reverts commit
566c3c7a22c0ec217c40f7a95db2bb83c8cc16c7 .
antirez [Wed, 21 Apr 2010 08:31:08 +0000 (10:31 +0200)]
Revert "define __USE_GNU to get O_DIRECT"
This reverts commit
122c04983400a076423811740f2b87163ea95c00 .
antirez [Wed, 21 Apr 2010 08:30:36 +0000 (10:30 +0200)]
Merge branch 'master' of github.com:antirez/redis
antirez [Wed, 21 Apr 2010 08:29:24 +0000 (10:29 +0200)]
Revert "version 1.3.10"
This reverts commit
188714211af246b96f46a2e598c66fee95c52375 .
antirez [Wed, 21 Apr 2010 07:54:21 +0000 (09:54 +0200)]
version 1.3.10
antirez [Tue, 20 Apr 2010 16:29:06 +0000 (18:29 +0200)]
define __USE_GNU to get O_DIRECT
antirez [Tue, 20 Apr 2010 16:25:30 +0000 (18:25 +0200)]
fsync always now uses O_DIRECT on Linux
antirez [Tue, 20 Apr 2010 09:01:36 +0000 (11:01 +0200)]
More precise memory used guesswork in zmalloc.c
antirez [Mon, 19 Apr 2010 16:30:11 +0000 (18:30 +0200)]
Fix for MULTI/EXEC and Replication/AOF: now the block is correctly sent as MULTI/..writing operations../EXEC. Ok for slaves but more work needed for the AOF as it should be a write-all-or-nothing business
antirez [Mon, 19 Apr 2010 14:11:13 +0000 (16:11 +0200)]
running the test using tcl8.5 directly instead of tclsh that too often it's a symlink to 8.4
antirez [Mon, 19 Apr 2010 09:05:08 +0000 (11:05 +0200)]
Added package require Tcl 8.5 in redis.tcl so it will show a clear error when the test suit is attempted to run under 8.4
antirez [Sun, 18 Apr 2010 15:51:15 +0000 (17:51 +0200)]
Fix for a SORT bug introduced with commit
16fa22f1 , regression test added
antirez [Sat, 17 Apr 2010 22:03:49 +0000 (00:03 +0200)]
Guru mediation -> meditation
Pieter Noordhuis [Fri, 16 Apr 2010 12:42:14 +0000 (14:42 +0200)]
check eptr inline