antirez
7b2b598ff9
not used parameter for redisLog() removed. Error message grammar fixed.
...
Thanks to @ardsrk.
2011-10-08 10:59:04 +02:00
antirez
773749c096
Redis Cluster: process node to node CLUSTERMSG_TYPE_PUBLISH messages and send it to the local clients.
2011-10-07 16:34:16 +02:00
antirez
4e4da410fd
propagate PUBLISH messages using the redis cluster nodes bus. Still need to process the incoming packets of that type. Work in progress.
2011-10-07 15:37:34 +02:00
antirez
5d78296230
Totally hackish and dirty, but working, support for Redis Cluster in redis-cli
2011-10-05 19:55:33 +02:00
antirez
96c208fe73
A node can populate a slot if a message from a trusted slot is received claiming ownership of this slot, that is currently empty or served by a node in FAIL state. However this feature was broken since calling clusterAddSlot() was not enough as the slot bit is already set in the node that pong us. We need to directly alter the table. This commit fixes the issue.
2011-10-05 17:40:02 +02:00
antirez
d7265db874
CONFIG SET/GET support for loglevel
2011-10-05 16:08:35 +02:00
antirez
cdd2206102
just minor aesthetic changes to cluster.c
2011-10-05 16:02:45 +02:00
antirez
357f49db2f
replaced redisAssert() with redisAssertWithInfo() in a shitload of places.
2011-10-04 18:43:03 +02:00
antirez
3e622d2c1d
redisAssertWithClientInfo() is now redisAssertWithInfo() that is also able to report an optional object. The client is also optional. Specifying NULL will prevent dumping the not available information (either client or object).
2011-10-04 18:05:26 +02:00
antirez
9f58a0880d
Introduced a redisAssert() variant that is able to show information about the client in the context where the failed assertion was detected.
2011-10-04 17:22:29 +02:00
antirez
fb589e2110
redis-trib: create subcommand fixed calling node info loading method. Was needed after recent refactoring.
2011-10-04 15:53:07 +02:00
antirez
154fe41066
Revert "Use the new install script as make install target. Message about install script requiring root changed a bit to make it more evident."
...
After talking with Pieter he changed my mind about this, it is better to
have a simpler install script that works everywhere, and the complex one
can be always executed if needed by hand. We'll make possibly a new
target for the full featured installation script, and even suggest it
after a Make install.
This reverts commit 51154dc772039fe633427e61d17b828d648329ff.
2011-10-03 16:04:44 +02:00
antirez
be453487d8
Rewrite MIGRATE as DEL for AOF/replication. Also increment the dirty counter to both force replication and persistence.
2011-10-03 15:51:10 +02:00
antirez
213df92591
increment server.dirty in restore command to make sure it is replicated and increments the dirty count for persistence concerns.
2011-10-03 15:45:14 +02:00
Salvatore Sanfilippo
e8c691b617
Merge pull request #114 from pietern/unstable-zcount
...
Use rank information to compute ZCOUNT
2011-10-03 06:20:23 -07:00
Pieter Noordhuis
bfad95e4cb
Remove ZCOUNT branches from generic RANGEBYSCORE code
2011-10-03 14:23:31 +02:00
Pieter Noordhuis
68360a1d39
Use element rank instead of iterating in ZCOUNT
2011-10-03 14:14:43 +02:00
antirez
51154dc772
Use the new install script as make install target. Message about install script requiring root changed a bit to make it more evident.
2011-10-03 13:28:54 +02:00
antirez
f564dd941c
Exact variant of CRC16 specified into crc16.c
2011-10-02 14:05:29 +02:00
antirez
223ec0fad1
when processing gossip packets make sure to also update the node bitmap when associating slots to nodes. Fixed simply using the appropriate helper function to add a slot.
2011-10-01 15:11:07 +02:00
antirez
90adbdc496
fixed a few problems with redis-trib resharding.
2011-10-01 01:31:46 +02:00
antirez
7561bd826a
redis-trib: initial working implementation of cluster rehashing, more work needed to make it more correct, more reliable.
2011-09-30 22:18:07 +02:00
antirez
f962242490
Fixed two bugs in the CLUSTER SETSLOT ... NODE subcommand
2011-09-30 22:17:24 +02:00
antirez
63b7e1d9b2
redis-trib: actual slot migration work in progress, more work needed.
2011-09-30 19:21:36 +02:00
antirez
ce26413ea5
Clear the importing status from the slot if a CLUSTER SETSLOT command permanently assigns the slot to the importing node.
2011-09-30 19:20:56 +02:00
antirez
0101386afe
redis-trib: better slots allocation strategy for resharding
2011-09-30 18:41:25 +02:00
antirez
76fb2ba03d
redis-trib: initial slots allocation strategy, still to improve for better balancing when there are just a few slots to move.
2011-09-30 18:28:51 +02:00
antirez
da1044c890
redis-trib: reshard command work in progress
2011-09-29 16:06:49 +02:00
antirez
7b7ca05bbd
redis-trib: minor refactoring creating a method to load the cluster information starting from a single node
2011-09-29 15:45:10 +02:00
antirez
f8a5bb158f
redis-trib: don't crash loading node info for nodes without assigned slots
2011-09-29 15:31:02 +02:00
antirez
1bec2c3634
added comment about SETSLOT NODE form.
2011-09-29 15:26:01 +02:00
antirez
fd9f590e2f
Fixed command table for RESTORE to make it accepting keys only for the right slots when cluster is enabled.
2011-09-29 15:14:06 +02:00
antirez
0c464a8087
redis-trib: added cluster state to node info string
2011-09-29 15:04:23 +02:00
antirez
d783732be5
redis-trib: all node information into a single hash
2011-09-29 14:44:27 +02:00
antirez
6ad5367a23
redis-trib: when loading node info also fetch the output of CLUSTER INFO.
2011-09-29 13:18:09 +02:00
antirez
1fbfc721f0
redis-trib cluster check command: check that all the 4096 slots are covered
2011-09-29 12:50:46 +02:00
antirez
9641edf1b4
Compilation fixed on OpenBSD making sure that _XOPEN_SOURCE is set to 700 in fmacros.h
2011-09-29 10:20:03 +02:00
antirez
58f345e77d
initial node loading info code for redis-trib
2011-09-28 19:08:35 +02:00
antirez
0ff23b8ab3
show node with myself flag always as connected in CLUSTER NODES output.
2011-09-28 18:28:36 +02:00
antirez
8f4045ff11
Scripting engine now only loads selected libraries, using code originally contributed by @loopole.
2011-09-27 18:46:23 +02:00
antirez
dec78a35a5
Return errors if a write command is called inside a Lua script after a random command was called. See https://github.com/antirez/redis/issues/95 for more information.
2011-09-27 15:30:31 +02:00
antirez
cd80380f20
Deny commands flagged as REDIS_CMD_NOSCRIPT from Lua scripts
2011-09-27 13:57:10 +02:00
antirez
2c4d7c157d
added the NOSCRIPT and RANDOM command flags
2011-09-27 13:45:46 +02:00
antirez
b84f94b1eb
command table refactoring to make it simpler adding new flags
2011-09-26 15:40:39 +02:00
antirez
66b7c21fc7
Lua math.random and math.randomseed replaced with our own version based on redisSrand48(). Seed initialized to the same value at every EVAL/EVALSHA call.
2011-09-23 15:40:58 +02:00
antirez
3ccc3570c2
Defined macro with bigger number that redisLrand48() can output.
2011-09-23 15:06:07 +02:00
antirez
148e45be5b
libc neutral random function derived from a drand48() implementation added. Will be used to replace Lua's math.random implementation.
2011-09-23 14:51:48 +02:00
antirez
3ad12623d4
rioInitWithFile nad rioInitWithBuffer functions now take a rio structure pointer to avoid copying a structure to return value to the caller.
2011-09-22 16:00:40 +02:00
antirez
596ac8715e
make sure to return just 1 for rio.c write when the target is a buffer, as we do when the target is a file.
2011-09-22 15:54:08 +02:00
antirez
06c356d912
comment on top of the _rio structure modified for correctness as actually fwrite/fread semantics is different in general, but was 0/1 in our old usage before rio.c as we always used 1 as number items, and the actual number of bytes to read as item length.
2011-09-22 15:47:48 +02:00
