]> git.saurik.com Git - redis.git/log
redis.git
12 years agoFix for slaves chains. Force resync of slaves (simply disconnecting them) when SLAVE...
antirez [Thu, 29 Mar 2012 07:24:02 +0000 (09:24 +0200)] 
Fix for slaves chains. Force resync of slaves (simply disconnecting them)  when SLAVEOF turns a master into a slave.

12 years agoTest for redis.sha1hex().
antirez [Wed, 28 Mar 2012 18:47:50 +0000 (20:47 +0200)] 
Test for redis.sha1hex().

12 years agoadded redis.sha1hex(string) as lua scripting function.
Nathan Fritz [Wed, 28 Mar 2012 18:10:24 +0000 (11:10 -0700)] 
added redis.sha1hex(string) as lua scripting function.

(The original implementation was modified by @antirez to conform Redis
coding standards.)

12 years agoFixes for redisLogFromHandler().
antirez [Wed, 28 Mar 2012 11:51:23 +0000 (13:51 +0200)] 
Fixes for redisLogFromHandler().

12 years agoLog from signal handlers is now safer.
antirez [Wed, 28 Mar 2012 11:45:39 +0000 (13:45 +0200)] 
Log from signal handlers is now safer.

12 years agoRedis test: regexp to check if valgrind reported errors modified. Now we no longer...
antirez [Wed, 28 Mar 2012 08:55:17 +0000 (10:55 +0200)] 
Redis test: regexp to check if valgrind reported errors modified. Now we no longer look at the total count because this includes "possibly lost" bytes that are not interesting for Redis (tons of false positives because of how sds.c works).

12 years agodefine zlibc_free() in a way that is not shadowed by jemalloc.
antirez [Tue, 27 Mar 2012 14:54:53 +0000 (16:54 +0200)] 
define zlibc_free() in a way that is not shadowed by jemalloc.

12 years agoProduce the watchlog warning log in a way that is safer from a signal handler. Fix...
antirez [Tue, 27 Mar 2012 13:24:33 +0000 (15:24 +0200)] 
Produce the watchlog warning log in a way that is safer from a signal handler. Fix a memory leak in the backtrace generation function.

12 years agoMask SIGALRM everything but in the main thread.
antirez [Tue, 27 Mar 2012 11:48:53 +0000 (13:48 +0200)] 
Mask SIGALRM everything but in the main thread.

This is required to ensure that the signal will be delivered to the main
thread when the watchdog timer expires.

12 years agoCorrectly set the SIGARLM timer for the software watchdog.
antirez [Tue, 27 Mar 2012 10:11:37 +0000 (12:11 +0200)] 
Correctly set the SIGARLM timer for the software watchdog.

12 years agoRedis software watchdog.
antirez [Tue, 27 Mar 2012 09:47:51 +0000 (11:47 +0200)] 
Redis software watchdog.

12 years agofix typo
huangz1990 [Thu, 15 Mar 2012 06:27:14 +0000 (14:27 +0800)] 
fix typo

12 years agofix time() instead of mstime() in expireIfNeeded
Premysl Hruby [Tue, 27 Mar 2012 15:19:54 +0000 (17:19 +0200)] 
fix time() instead of mstime() in expireIfNeeded

12 years agouse server.unixtime instead of time(NULL) where possible (cluster.c not checked though)
Premysl Hruby [Tue, 27 Mar 2012 15:39:58 +0000 (17:39 +0200)] 
use server.unixtime instead of time(NULL) where possible (cluster.c not checked though)

12 years agodeclare hashDictType as external too
Premysl Hruby [Tue, 27 Mar 2012 16:18:57 +0000 (18:18 +0200)] 
declare hashDictType as external too

12 years agoremove disk-store related comments
Premysl Hruby [Tue, 27 Mar 2012 16:46:51 +0000 (18:46 +0200)] 
remove disk-store related comments

12 years agoSIGSEGV handler refactored so that we can reuse stack trace and current client loggin...
antirez [Tue, 27 Mar 2012 08:33:45 +0000 (10:33 +0200)] 
SIGSEGV handler refactored so that we can reuse stack trace and current client logging functionalities in other contexts.

12 years agoCONFIG RESETSTAT resets two more fields.
antirez [Sun, 25 Mar 2012 09:43:19 +0000 (11:43 +0200)] 
CONFIG RESETSTAT resets two more fields.

12 years agoNew INFO field aof_delayed_fsync introduced.
antirez [Sun, 25 Mar 2012 09:27:35 +0000 (11:27 +0200)] 
New INFO field aof_delayed_fsync introduced.

This new field counts all the times Redis is configured with AOF enabled and
fsync policy 'everysec', but the previous fsync performed by the
background thread was not able to complete within two seconds, forcing
Redis to perform a write against the AOF file while the fsync is still
in progress (likely a blocking operation).

12 years agoVersion is now 2.5.3.
antirez [Sun, 25 Mar 2012 09:18:24 +0000 (11:18 +0200)] 
Version is now 2.5.3.

12 years agoconvert-zipmap-hash-on-load false positive fixed.
antirez [Sun, 25 Mar 2012 08:57:34 +0000 (10:57 +0200)] 
convert-zipmap-hash-on-load false positive fixed.

Apparently because the sample RDB file was not copied before every test
Redis had a chance to replace it with a newly written one, so that the
next test could fail.

12 years agoWhen running the test in valgrind mode, pass the right flags to show memory leaks...
antirez [Sat, 24 Mar 2012 11:06:56 +0000 (12:06 +0100)] 
When running the test in valgrind mode, pass the right flags to show memory leaks stack traces but only including the "definitely lost" items.

12 years agoAdd used allocator in redis-server -v output.
antirez [Sat, 24 Mar 2012 10:52:56 +0000 (11:52 +0100)] 
Add used allocator in redis-server -v output.

12 years agoconvert-zipmap-hash-on-load test enabled
antirez [Sat, 24 Mar 2012 10:42:20 +0000 (11:42 +0100)] 
convert-zipmap-hash-on-load test enabled

12 years agoContextualize comment.
antirez [Fri, 23 Mar 2012 19:20:43 +0000 (20:20 +0100)] 
Contextualize comment.

12 years agoMerge remote-tracking branch 'origin/2.6' into 2.6
antirez [Fri, 23 Mar 2012 19:20:13 +0000 (20:20 +0100)] 
Merge remote-tracking branch 'origin/2.6' into 2.6

12 years agoRDB load of different encodings test added.
antirez [Fri, 23 Mar 2012 14:22:25 +0000 (15:22 +0100)] 
RDB load of different encodings test added.

12 years agoFixed memory leak in hash loading.
antirez [Fri, 23 Mar 2012 19:21:19 +0000 (20:21 +0100)] 
Fixed memory leak in hash loading.

12 years agoRDB load of different encodings test added.
antirez [Fri, 23 Mar 2012 14:22:25 +0000 (15:22 +0100)] 
RDB load of different encodings test added.

12 years agoBig endian fix. The bug was introduced because of a typo.
antirez [Fri, 23 Mar 2012 11:42:20 +0000 (12:42 +0100)] 
Big endian fix. The bug was introduced because of a typo.

12 years agoReplicate HINCRBYFLOAT as HSET.
antirez [Fri, 23 Mar 2012 09:22:58 +0000 (10:22 +0100)] 
Replicate HINCRBYFLOAT as HSET.

12 years agoCode style hack.
antirez [Thu, 22 Mar 2012 17:16:41 +0000 (18:16 +0100)] 
Code style hack.

12 years agoResult of INCRBYFLOAT and HINCRBYFLOAT should never be in exponential form, and also...
antirez [Thu, 22 Mar 2012 17:14:32 +0000 (18:14 +0100)] 
Result of INCRBYFLOAT and HINCRBYFLOAT should never be in exponential form, and also should never contain trailing zeroes. This is not possible with vanilla printf() format specifiers, so we alter the output.

12 years agoComments about security of slave-read-only in redis.coinf.
antirez [Wed, 21 Mar 2012 11:26:05 +0000 (12:26 +0100)] 
Comments about security of slave-read-only in redis.coinf.

12 years agoCorrectly create shared.oomerr as an sds string.
antirez [Wed, 21 Mar 2012 11:11:07 +0000 (12:11 +0100)] 
Correctly create shared.oomerr as an sds string.

12 years agoDEBUG should not be flagged as w otherwise we can not call DEBUG DIGEST and other...
antirez [Tue, 20 Mar 2012 16:53:47 +0000 (17:53 +0100)] 
DEBUG should not be flagged as w otherwise we can not call DEBUG DIGEST and other commands against read only slaves.

12 years agoSupport for read-only slaves. Semantical fixes.
antirez [Tue, 20 Mar 2012 16:32:48 +0000 (17:32 +0100)] 
Support for read-only slaves. Semantical fixes.

This commit introduces support for read only slaves via redis.conf and CONFIG GET/SET commands. Also various semantical fixes are implemented here:

1) MULTI/EXEC with only read commands now work where the server is into a state where writes (or commands increasing memory usage) are not allowed. Before this patch everything inside a transaction would fail in this conditions.

2) Scripts just calling read-only commands will work against read only
slaves, when the server is out of memory, or when persistence is into an
error condition. Before the patch EVAL always failed in this condition.

12 years agoRedis cluster stuff removed from 2.6 redis.conf file.
antirez [Tue, 20 Mar 2012 12:07:26 +0000 (13:07 +0100)] 
Redis cluster stuff removed from 2.6 redis.conf file.

12 years agoredis_init_script template updated.
antirez [Tue, 20 Mar 2012 12:06:50 +0000 (13:06 +0100)] 
redis_init_script template updated.

12 years agoFixed typo in 2.6 release notes.
antirez [Mon, 19 Mar 2012 21:10:18 +0000 (22:10 +0100)] 
Fixed typo in 2.6 release notes.

12 years agoMerge remote-tracking branch 'origin/2.6' into 2.6
antirez [Mon, 19 Mar 2012 18:30:30 +0000 (19:30 +0100)] 
Merge remote-tracking branch 'origin/2.6' into 2.6

12 years agoSuppress warnings compiling redis-cli with certain gcc versions.
antirez [Mon, 19 Mar 2012 18:29:06 +0000 (19:29 +0100)] 
Suppress warnings compiling redis-cli with certain gcc versions.

12 years agoRead-only flag removed from PUBLISH command.
antirez [Mon, 19 Mar 2012 18:16:41 +0000 (19:16 +0100)] 
Read-only flag removed from PUBLISH command.

12 years agoMemory addressing test implemented.
antirez [Mon, 19 Mar 2012 13:02:34 +0000 (14:02 +0100)] 
Memory addressing test implemented.

12 years agoMore memory tests implemented. Default number of iterations lowered to a more accepta...
antirez [Sun, 18 Mar 2012 17:03:27 +0000 (18:03 +0100)] 
More memory tests implemented. Default number of iterations lowered to a more acceptable value of 50.

12 years agoFixed typo.
antirez [Sun, 18 Mar 2012 16:27:56 +0000 (17:27 +0100)] 
Fixed typo.

12 years agoNumber of iteration of --test-memory is now 300 (several minutes per gigabyte). Memte...
antirez [Sun, 18 Mar 2012 16:24:48 +0000 (17:24 +0100)] 
Number of iteration of --test-memory is now 300 (several minutes per gigabyte). Memtest86 and Memtester links are also displayed while running the test.

12 years agoOn crash suggest to give --test-memory a try.
antirez [Sun, 18 Mar 2012 10:35:35 +0000 (11:35 +0100)] 
On crash suggest to give --test-memory a try.

12 years agoMemory test function now less boring thanks to screen-wide progress bar.
antirez [Fri, 16 Mar 2012 20:19:53 +0000 (21:19 +0100)] 
Memory test function now less boring thanks to screen-wide progress bar.

12 years agoHem... actual memtest.c file added.
antirez [Fri, 16 Mar 2012 16:21:49 +0000 (17:21 +0100)] 
Hem... actual memtest.c file added.

12 years agoFirst implementation of --test-memory. Still a work in progress.
antirez [Fri, 16 Mar 2012 16:17:39 +0000 (17:17 +0100)] 
First implementation of --test-memory. Still a work in progress.

12 years agoFix for issue #391.
antirez [Thu, 15 Mar 2012 19:51:35 +0000 (20:51 +0100)] 
Fix for issue #391.

Use a simple protocol between clientsCron() and helper functions to
understand if the client is still valind and clientsCron() should
continue processing or if the client was freed and we should continue
with the next one.

12 years agoVersion bumped to 2.5.2
antirez [Wed, 14 Mar 2012 14:37:47 +0000 (15:37 +0100)] 
Version bumped to 2.5.2

12 years agoReclaim space from the client querybuf if needed.
antirez [Wed, 14 Mar 2012 14:32:30 +0000 (15:32 +0100)] 
Reclaim space from the client querybuf if needed.

12 years agosds.c: sdsAllocSize() function added.
antirez [Wed, 14 Mar 2012 13:58:26 +0000 (14:58 +0100)] 
sds.c: sdsAllocSize() function added.

12 years agosds.c new function sdsRemoveFreeSpace().
antirez [Wed, 14 Mar 2012 09:13:23 +0000 (10:13 +0100)] 
sds.c new function sdsRemoveFreeSpace().

The new function is used in order to resize the string allocation so
that only the minimal allocation possible is used, removing all the free
space at the end of the string normally used to improve efficiency of
concatenation operations.

12 years agoCall all the helper functions needed by clientsCron() as clientsCronSomething() for...
antirez [Wed, 14 Mar 2012 08:56:22 +0000 (09:56 +0100)] 
Call all the helper functions needed by clientsCron() as clientsCronSomething() for clarity.

12 years agoCLIENT LIST test modified to reflect the new output.
antirez [Tue, 13 Mar 2012 17:06:29 +0000 (18:06 +0100)] 
CLIENT LIST test modified to reflect the new output.

12 years agoProcess async client checks like client timeouts and BLPOP timeouts incrementally...
antirez [Tue, 13 Mar 2012 17:05:11 +0000 (18:05 +0100)] 
Process async client checks like client timeouts and BLPOP timeouts incrementally using a circular list.

12 years agoAdded a qbuf-free field to CLIENT LIST output.
antirez [Tue, 13 Mar 2012 12:26:33 +0000 (13:26 +0100)] 
Added a qbuf-free field to CLIENT LIST output.

12 years agoClient creation time in redisClient structure. New age field in CLIENT LIST output.
antirez [Tue, 13 Mar 2012 12:05:08 +0000 (13:05 +0100)] 
Client creation time in redisClient structure. New age field in CLIENT LIST output.

12 years agoc->bufpos initialization moved for aesthetics.
antirez [Tue, 13 Mar 2012 11:59:27 +0000 (12:59 +0100)] 
c->bufpos initialization moved for aesthetics.

12 years agoRDB hashes loading, fixed another bug in the loading of HT-encoded hashes: when the...
antirez [Tue, 13 Mar 2012 09:59:29 +0000 (10:59 +0100)] 
RDB hashes loading, fixed another bug in the loading of HT-encoded hashes: when the hash entry is too big for ziplist, add the field, then convert. The code used to break before the new entry was inserted, resulting into missing fields in the loaded Hash object.

12 years agoRDB hashes loading fixed removing the assertion that failed every time an HT-encoded...
antirez [Tue, 13 Mar 2012 08:49:11 +0000 (09:49 +0100)] 
RDB hashes loading fixed removing the assertion that failed every time an HT-encoded hash was loaded.

12 years agofix typo of redis.conf
quiver [Sat, 10 Mar 2012 12:06:08 +0000 (21:06 +0900)] 
fix typo of redis.conf

12 years agoRDB4 support in redis-check-dump.
antirez [Sat, 10 Mar 2012 11:38:42 +0000 (12:38 +0100)] 
RDB4 support in redis-check-dump.

12 years agoRDB version is no 4, because small hashes are now encoded as ziplists, so older versi...
antirez [Sat, 10 Mar 2012 11:35:15 +0000 (12:35 +0100)] 
RDB version is no 4, because small hashes are now encoded as ziplists, so older versions of Redis will not understand this format.

12 years agoVersion is now 2.5.1, first unstable release of Redis 2.6
antirez [Sat, 10 Mar 2012 11:29:47 +0000 (12:29 +0100)] 
Version is now 2.5.1, first unstable release of Redis 2.6

12 years agoBuild dependencies updated.
antirez [Sat, 10 Mar 2012 11:28:14 +0000 (12:28 +0100)] 
Build dependencies updated.

12 years agoRedis 2.6 branch obtained from unstable removing all the cluster related code.
antirez [Sat, 10 Mar 2012 11:26:37 +0000 (12:26 +0100)] 
Redis 2.6 branch obtained from unstable removing all the cluster related code.

12 years agohash-max-zipmap-... renamed hash-max-ziplist-... in defalt conf for tests.
antirez [Sat, 10 Mar 2012 11:14:17 +0000 (12:14 +0100)] 
hash-max-zipmap-... renamed hash-max-ziplist-... in defalt conf for tests.

12 years agoMore vertical space saved.
antirez [Sat, 10 Mar 2012 10:11:28 +0000 (11:11 +0100)] 
More vertical space saved.

12 years agoMinor code aesthetic change to use Redis code base style rule of saving vertical...
antirez [Sat, 10 Mar 2012 10:09:43 +0000 (11:09 +0100)] 
Minor code aesthetic change to use Redis code base style rule of saving vertical space when possible.

12 years agoRemoved handling of deprecated hash-max-zipmap-entries nad hash-map-zipmap-value...
antirez [Sat, 10 Mar 2012 09:38:53 +0000 (10:38 +0100)] 
Removed handling of deprecated hash-max-zipmap-entries nad hash-map-zipmap-value. Pieter is too good with users ;). Better to have them switch to a saner configuration ASAP after the 2.6 upgrade.

12 years agoAdded a top-function comment to rioWriteHashIteratorCursor() to better specify what...
antirez [Sat, 10 Mar 2012 09:36:51 +0000 (10:36 +0100)] 
Added a top-function comment to rioWriteHashIteratorCursor() to better specify what the function does. Not immediately clear from the name.

12 years agoMerge conflicts resolved.
antirez [Fri, 9 Mar 2012 21:07:45 +0000 (22:07 +0100)] 
Merge conflicts resolved.

12 years agoInstantaneous ops/sec figure in INFO output.
antirez [Thu, 8 Mar 2012 15:15:37 +0000 (16:15 +0100)] 
Instantaneous ops/sec figure in INFO output.

12 years agoSupport for all the redis.conf fields in CONFIG GET. config.c refactored a bit.
antirez [Thu, 8 Mar 2012 11:14:23 +0000 (12:14 +0100)] 
Support for all the redis.conf fields in CONFIG GET. config.c refactored a bit.

12 years agorun_id added to INFO output.
antirez [Thu, 8 Mar 2012 09:13:12 +0000 (10:13 +0100)] 
run_id added to INFO output.

The Run ID is a field that identifies a single execution of the Redis
server. It can be useful for many purposes as it makes easy to detect if
the instance we are talking about is the same, or if it is a different
one or was rebooted. An application of run_id will be in the partial
synchronization of replication, where a slave may request a partial sync
from a given offset only if it is talking with the same master. Another
application is in failover and monitoring scripts.

12 years agoclusterGetRandomName() generalized into getRandomHexChars() so that we can use it...
antirez [Thu, 8 Mar 2012 09:08:44 +0000 (10:08 +0100)] 
clusterGetRandomName() generalized into getRandomHexChars() so that we can use it for the run_id field as well.

12 years agoBy default Redis refuses writes with an error if the latest BGSAVE failed (and at...
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.

12 years agoRefuse writes if can't persist on disk.
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.

12 years agoBetter MONITOR output, now includes client ip:port or the lua string if the command...
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.

12 years agoanetPeerToString() automatically populates ip/port with something that may be provide...
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.

12 years agoTIME command.
antirez [Wed, 7 Mar 2012 09:38:01 +0000 (10:38 +0100)] 
TIME command.

12 years agoredis-cli: CSV output added, used for the --slave mode.
antirez [Wed, 29 Feb 2012 16:43:03 +0000 (17:43 +0100)] 
redis-cli: CSV output added, used for the --slave mode.

12 years agoInitial implementation of redis-cli --slave support.
antirez [Wed, 29 Feb 2012 16:10:21 +0000 (17:10 +0100)] 
Initial implementation of redis-cli --slave support.

12 years agoPing the slave using the standard protocol instead of the inline one.
antirez [Wed, 29 Feb 2012 15:33:54 +0000 (16:33 +0100)] 
Ping the slave using the standard protocol instead of the inline one.

12 years agoBetter implementation for BRPOP/BLPOP in the non blocking case.
antirez [Wed, 29 Feb 2012 13:41:57 +0000 (14:41 +0100)] 
Better implementation for BRPOP/BLPOP in the non blocking case.

12 years agolpush arguments vector rewrite modified for more speed and to memory leak removal.
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.

12 years agoVersion bumped to 2.9.5
antirez [Tue, 28 Feb 2012 23:54:52 +0000 (00:54 +0100)] 
Version bumped to 2.9.5

12 years agoBetter system for additional commands replication.
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.

12 years agoAdded a new API to replicate an additional command after the replication of the curre...
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.

12 years agopropagate() prototype added to redis.h
antirez [Tue, 28 Feb 2012 15:20:41 +0000 (16:20 +0100)] 
propagate() prototype added to redis.h

12 years agoVar renamed into pushGenericCommand() to better reflect what it means.
antirez [Tue, 28 Feb 2012 15:17:55 +0000 (16:17 +0100)] 
Var renamed into pushGenericCommand() to better reflect what it means.

12 years agoAdded command propagation API.
antirez [Tue, 28 Feb 2012 15:17:00 +0000 (16:17 +0100)] 
Added command propagation API.

12 years agoMerge pull request #362 from fawek/mac-build-fix
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

12 years agoFix 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

12 years agouse gcc sync builtins for memory counter when possible
Premysl Hruby [Sun, 26 Feb 2012 08:59:32 +0000 (09:59 +0100)] 
use gcc sync builtins for memory counter when possible

12 years agofix redis-benchmark memory leak
Premysl Hruby [Sun, 26 Feb 2012 09:01:27 +0000 (10:01 +0100)] 
fix redis-benchmark memory leak