]> git.saurik.com Git - redis.git/blame - redis.conf
Fixed some spelling errors in the comments
[redis.git] / redis.conf
CommitLineData
ed9b544e 1# Redis configuration file example
2
57c0cf8b 3# Note on units: when memory size is needed, it is possible to specify
72324005 4# it in the usual form of 1k 5GB 4M and so forth:
5#
6# 1k => 1000 bytes
7# 1kb => 1024 bytes
8# 1m => 1000000 bytes
9# 1mb => 1024*1024 bytes
10# 1g => 1000000000 bytes
11# 1gb => 1024*1024*1024 bytes
12#
13# units are case insensitive so 1GB 1Gb 1gB are all the same.
14
ed9b544e 15# By default Redis does not run as a daemon. Use 'yes' if you need it.
16# Note that Redis will write a pid file in /var/run/redis.pid when daemonized.
17daemonize no
18
029245fe 19# When running daemonized, Redis writes a pid file in /var/run/redis.pid by
20# default. You can specify a custom pid file location here.
ed329fcf
LH
21pidfile /var/run/redis.pid
22
a5639e7d 23# Accept connections on the specified port, default is 6379.
68d6345d 24# If port 0 is specified Redis will not listen on a TCP socket.
ed9b544e 25port 6379
26
27# If you want you can bind a single interface, if the bind option is not
029245fe 28# specified all the interfaces will listen for incoming connections.
ed9b544e 29#
30# bind 127.0.0.1
31
5d10923f
PN
32# Specify the path for the unix socket that will be used to listen for
33# incoming connections. There is no default, so Redis will not listen
34# on a unix socket when not specified.
a5639e7d 35#
5d10923f 36# unixsocket /tmp/redis.sock
85238765 37# unixsocketperm 755
a5639e7d 38
0150db36 39# Close the connection after a client is idle for N seconds (0 to disable)
aba4adb7 40timeout 0
ed9b544e 41
121f70cf 42# Set server verbosity to 'debug'
43# it can be one of:
44# debug (a lot of information, useful for development/testing)
38aba9a1 45# verbose (many rarely useful info, but not a mess like the debug level)
121f70cf 46# notice (moderately verbose, what you want in production probably)
47# warning (only very important / critical messages are logged)
c6f9ee88 48loglevel notice
121f70cf 49
50# Specify the log file name. Also 'stdout' can be used to force
029245fe 51# Redis to log on the standard output. Note that if you use standard
121f70cf 52# output for logging but daemonize, logs will be sent to /dev/null
53logfile stdout
54
e1a586ee
JH
55# To enable logging to the system logger, just set 'syslog-enabled' to yes,
56# and optionally update the other syslog parameters to suit your needs.
57# syslog-enabled no
58
59# Specify the syslog identity.
60# syslog-ident redis
61
62# Specify the syslog facility. Must be USER or between LOCAL0-LOCAL7.
63# syslog-facility local0
64
121f70cf 65# Set the number of databases. The default database is DB 0, you can select
66# a different one on a per-connection basis using SELECT <dbid> where
67# dbid is a number between 0 and 'databases'-1
68databases 16
69
70################################ SNAPSHOTTING #################################
71#
ed9b544e 72# Save the DB on disk:
73#
74# save <seconds> <changes>
75#
76# Will save the DB if both the given number of seconds and the given
77# number of write operations against the DB occurred.
78#
79# In the example below the behaviour will be to save:
80# after 900 sec (15 min) if at least 1 key changed
81# after 300 sec (5 min) if at least 10 keys changed
82# after 60 sec if at least 10000 keys changed
e7546c63 83#
84# Note: you can disable saving at all commenting all the "save" lines.
4aac3ff2 85#
86# It is also possible to remove all the previously configured save
87# points by adding a save directive with a single empty string argument
88# like in the following example:
89#
90# save ""
e7546c63 91
38aba9a1 92save 900 1
93save 300 10
94save 60 10000
ed9b544e 95
4d3bbf35 96# By default Redis will stop accepting writes if RDB snapshots are enabled
97# (at least one save point) and the latest background save failed.
98# This will make the user aware (in an hard way) that data is not persisting
99# on disk properly, otherwise chances are that no one will notice and some
100# distater will happen.
101#
102# If the background saving process will start working again Redis will
103# automatically allow writes again.
104#
105# However if you have setup your proper monitoring of the Redis server
106# and persistence, you may want to disable this feature so that Redis will
107# continue to work as usually even if there are problems with disk,
108# permissions, and so forth.
109stop-writes-on-bgsave-error yes
110
121f70cf 111# Compress string objects using LZF when dump .rdb databases?
b0553789 112# For default that's set to 'yes' as it's almost always a win.
113# If you want to save some CPU in the saving child set it to 'no' but
114# the dataset will likely be bigger if you have compressible values or keys.
115rdbcompression yes
121f70cf 116
39d1e350 117# Since verison 5 of RDB a CRC64 checksum is placed at the end of the file.
118# This makes the format more resistant to corruption but there is a performance
119# hit to pay (around 10%) when saving and loading RDB files, so you can disable it
120# for maximum performances.
121#
122# RDB files created with checksum disabled have a checksum of zero that will
123# tell the loading code to skip the check.
124rdbchecksum yes
125
b8b553c8 126# The filename where to dump the DB
127dbfilename dump.rdb
128
029245fe 129# The working directory.
130#
131# The DB will be written inside this directory, with the filename specified
132# above using the 'dbfilename' configuration directive.
133#
134# Also the Append Only File will be created inside this directory.
135#
136# Note that you must specify a directory here, not a file name.
ed9b544e 137dir ./
138
ed9b544e 139################################# REPLICATION #################################
140
141# Master-Slave replication. Use slaveof to make a Redis instance a copy of
142# another Redis server. Note that the configuration is local to the slave
143# so for example it is possible to configure the slave to save the DB with a
144# different interval, or to listen to another port, and so on.
3f477979 145#
ed9b544e 146# slaveof <masterip> <masterport>
147
3f477979 148# If the master is password protected (using the "requirepass" configuration
149# directive below) it is possible to tell the slave to authenticate before
150# starting the replication synchronization process, otherwise the master will
151# refuse the slave request.
152#
153# masterauth <master-password>
154
4ebfc455 155# When a slave lost the connection with the master, or when the replication
156# is still in progress, the slave can act in two different ways:
157#
158# 1) if slave-serve-stale-data is set to 'yes' (the default) the slave will
92a157ea 159# still reply to client requests, possibly with out of date data, or the
4ebfc455 160# data set may just be empty if this is the first synchronization.
161#
162# 2) if slave-serve-stale data is set to 'no' the slave will reply with
163# an error "SYNC with master in progress" to all the kind of commands
164# but to INFO and SLAVEOF.
165#
166slave-serve-stale-data yes
167
05406168 168# You can configure a slave instance to accept writes or not. Writing against
169# a slave instance may be useful to store some ephemeral data (because data
170# written on a slave will be easily deleted after resync with the master) but
9aba884b 171# may also cause problems if clients are writing to it because of a
172# misconfiguration.
05406168 173#
174# Since Redis 2.6 by default slaves are read-only.
9aba884b 175#
176# Note: read only slaves are not designed to be exposed to untrusted clients
177# on the internet. It's just a protection layer against misuse of the instance.
178# Still a read only slave exports by default all the administrative commands
179# such as CONFIG, DEBUG, and so forth. To a limited extend you can improve
180# security of read only slaves using 'rename-command' to shadow all the
181# administrative / dangerous commands.
05406168 182slave-read-only yes
183
8996bf77 184# Slaves send PINGs to server in a predefined interval. It's possible to change
185# this interval with the repl_ping_slave_period option. The default value is 10
186# seconds.
187#
f15e33a8 188# repl-ping-slave-period 10
8996bf77 189
190# The following option sets a timeout for both Bulk transfer I/O timeout and
191# master data or ping response timeout. The default value is 60 seconds.
192#
85ccd576 193# It is important to make sure that this value is greater than the value
194# specified for repl-ping-slave-period otherwise a timeout will be detected
195# every time there is low traffic between the master and the slave.
196#
f15e33a8 197# repl-timeout 60
8996bf77 198
851ac9d0 199# The slave priority is an integer number published by Redis in the INFO output.
200# It is used by Redis Sentinel in order to select a slave to promote into a
201# master if the master is no longer working correctly.
202#
203# A slave with a low priority number is considered better for promotion, so
204# for instance if there are three slaves with priority 10, 100, 25 Sentinel will
205# pick the one wtih priority 10, that is the lowest.
206#
207# However a special priority of 0 marks the slave as not able to perform the
208# role of master, so a slave with priority of 0 will never be selected by
209# Redis Sentinel for promotion.
210#
211# By default the priority is 100.
212slave-priority 100
213
f2aa84bd 214################################## SECURITY ###################################
215
216# Require clients to issue AUTH <PASSWORD> before processing any other
217# commands. This might be useful in environments in which you do not trust
218# others with access to the host running redis-server.
219#
220# This should stay commented out for backward compatibility and because most
221# people do not need auth (e.g. they run their own servers).
1b677732 222#
223# Warning: since Redis is pretty fast an outside user can try up to
224# 150k passwords per second against a good box. This means that you should
225# use a very strong password otherwise it will be very easy to break.
3f477979 226#
290deb8b 227# requirepass foobared
f2aa84bd 228
8d3e063a 229# Command renaming.
230#
57c0cf8b 231# It is possible to change the name of dangerous commands in a shared
8d3e063a 232# environment. For instance the CONFIG command may be renamed into something
233# of hard to guess so that it will be still available for internal-use
234# tools but not available for general clients.
235#
236# Example:
237#
238# rename-command CONFIG b840fc02d524045429941cc15f59e41cb7be6c52
239#
57c0cf8b 240# It is also possible to completely kill a command renaming it into
8d3e063a 241# an empty string:
242#
243# rename-command CONFIG ""
244
285add55 245################################### LIMITS ####################################
246
58732c23 247# Set the max number of connected clients at the same time. By default
248# this limit is set to 10000 clients, however if the Redis server is not
249# able ot configure the process file limit to allow for the specified limit
250# the max number of allowed clients is set to the current file limit
251# minus 32 (as Redis reserves a few file descriptors for internal uses).
252#
285add55 253# Once the limit is reached Redis will close all the new connections sending
254# an error 'max number of clients reached'.
3f477979 255#
58732c23 256# maxclients 10000
285add55 257
3fd78bcd 258# Don't use more memory than the specified amount of bytes.
cebb7b92 259# When the memory limit is reached Redis will try to remove keys
260# accordingly to the eviction policy selected (see maxmemmory-policy).
261#
262# If Redis can't remove keys according to the policy, or if the policy is
263# set to 'noeviction', Redis will start to reply with errors to commands
264# that would use more memory, like SET, LPUSH, and so on, and will continue
265# to reply to read-only commands like GET.
266#
267# This option is usually useful when using Redis as an LRU cache, or to set
268# an hard memory limit for an instance (using the 'noeviction' policy).
269#
270# WARNING: If you have slaves attached to an instance with maxmemory on,
271# the size of the output buffers needed to feed the slaves are subtracted
272# from the used memory count, so that network problems / resyncs will
273# not trigger a loop where keys are evicted, and in turn the output
274# buffer of slaves is full with DELs of keys evicted triggering the deletion
275# of more keys, and so forth until the database is completely emptied.
3f477979 276#
f9ef912c 277# In short... if you have slaves attached it is suggested that you set a lower
278# limit for maxmemory so that there is some free RAM on the system for slave
279# output buffers (but this is not needed if the policy is 'noeviction').
280#
3fd78bcd 281# maxmemory <bytes>
282
165346ca 283# MAXMEMORY POLICY: how Redis will select what to remove when maxmemory
284# is reached? You can select among five behavior:
285#
286# volatile-lru -> remove the key with an expire set using an LRU algorithm
287# allkeys-lru -> remove any key accordingly to the LRU algorithm
288# volatile-random -> remove a random key with an expire set
dfc25454 289# allkeys-random -> remove a random key, any key
165346ca 290# volatile-ttl -> remove the key with the nearest expire time (minor TTL)
5402c426 291# noeviction -> don't expire at all, just return an error on write operations
292#
293# Note: with all the kind of policies, Redis will return an error on write
294# operations, when there are not suitable keys for eviction.
295#
296# At the date of writing this commands are: set setnx setex append
297# incr decr rpush lpush rpushx lpushx linsert lset rpoplpush sadd
298# sinter sinterstore sunion sunionstore sdiff sdiffstore zadd zincrby
299# zunionstore zinterstore hset hsetnx hmset hincrby incrby decrby
300# getset mset msetnx exec sort
301#
302# The default is:
165346ca 303#
304# maxmemory-policy volatile-lru
305
306# LRU and minimal TTL algorithms are not precise algorithms but approximated
307# algorithms (in order to save memory), so you can select as well the sample
308# size to check. For instance for default Redis will check three keys and
309# pick the one that was used less recently, you can change the sample size
310# using the following configuration directive.
311#
a3687929 312# maxmemory-samples 3
165346ca 313
44b38ef4 314############################## APPEND ONLY MODE ###############################
315
c11a01a0 316# By default Redis asynchronously dumps the dataset on disk. This mode is
317# good enough in many applications, but an issue with the Redis process or
318# a power outage may result into a few minutes of writes lost (depending on
319# the configured save points).
320#
321# The Append Only File is an alternative persistence mode that provides
322# much better durability. For instance using the default data fsync policy
323# (see later in the config file) Redis can lose just one second of writes in a
324# dramatic event like a server power outage, or a single write if something
325# wrong with the Redis process itself happens, but the operating system is
326# still running correctly.
327#
328# AOF and RDB persistence can be enabled at the same time without problems.
329# If the AOF is enabled on startup Redis will load the AOF, that is the file
330# with the better durability guarantees.
331#
332# Please check http://redis.io/topics/persistence for more information.
44b38ef4 333
4e141d5a 334appendonly no
44b38ef4 335
f3b52411
PN
336# The name of the append only file (default: "appendonly.aof")
337# appendfilename appendonly.aof
338
4e141d5a 339# The fsync() call tells the Operating System to actually write data on disk
48f0308a 340# instead to wait for more data in the output buffer. Some OS will really flush
341# data on disk, some other OS will just try to do it ASAP.
342#
343# Redis supports three different modes:
344#
345# no: don't fsync, just let the OS flush the data when it wants. Faster.
346# always: fsync after every write to the append only log . Slow, Safest.
c11a01a0 347# everysec: fsync only one time every second. Compromise.
48f0308a 348#
6766f45e 349# The default is "everysec" that's usually the right compromise between
350# speed and data safety. It's up to you to understand if you can relax this to
ce6628da 351# "no" that will let the operating system flush the output buffer when
6766f45e 352# it wants, for better performances (but if you can live with the idea of
353# some data loss consider the default persistence mode that's snapshotting),
354# or on the contrary, use "always" that's very slow but a bit safer than
355# everysec.
356#
c11a01a0 357# More details please check the following article:
358# http://antirez.com/post/redis-persistence-demystified.html
359#
6766f45e 360# If unsure, use "everysec".
361
362# appendfsync always
363appendfsync everysec
4e141d5a 364# appendfsync no
48f0308a 365
d5d23dab 366# When the AOF fsync policy is set to always or everysec, and a background
367# saving process (a background save or AOF log background rewriting) is
368# performing a lot of I/O against the disk, in some Linux configurations
369# Redis may block too long on the fsync() call. Note that there is no fix for
370# this currently, as even performing fsync in a different thread will block
371# our synchronous write(2) call.
372#
373# In order to mitigate this problem it's possible to use the following option
374# that will prevent fsync() from being called in the main process while a
375# BGSAVE or BGREWRITEAOF is in progress.
376#
377# This means that while another child is saving the durability of Redis is
57c0cf8b 378# the same as "appendfsync none", that in practical terms means that it is
d5d23dab 379# possible to lost up to 30 seconds of log in the worst scenario (with the
380# default Linux settings).
381#
382# If you have latency problems turn this to "yes". Otherwise leave it as
383# "no" that is the safest pick from the point of view of durability.
384no-appendfsync-on-rewrite no
385
b333e239 386# Automatic rewrite of the append only file.
387# Redis is able to automatically rewrite the log file implicitly calling
388# BGREWRITEAOF when the AOF log size will growth by the specified percentage.
389#
390# This is how it works: Redis remembers the size of the AOF file after the
391# latest rewrite (or if no rewrite happened since the restart, the size of
392# the AOF at startup is used).
393#
394# This base size is compared to the current size. If the current size is
395# bigger than the specified percentage, the rewrite is triggered. Also
396# you need to specify a minimal size for the AOF file to be rewritten, this
397# is useful to avoid rewriting the AOF file even if the percentage increase
398# is reached but it is still pretty small.
399#
57c0cf8b 400# Specify a percentage of zero in order to disable the automatic AOF
b333e239 401# rewrite feature.
402
403auto-aof-rewrite-percentage 100
404auto-aof-rewrite-min-size 64mb
405
eeffcf38 406################################ LUA SCRIPTING ###############################
407
408# Max execution time of a Lua script in milliseconds.
115e3ff3 409#
410# If the maximum execution time is reached Redis will log that a script is
57c0cf8b 411# still in execution after the maximum allowed time and will start to
115e3ff3 412# reply to queries with an error.
413#
57c0cf8b 414# When a long running script exceed the maximum execution time only the
0b14e441 415# SCRIPT KILL and SHUTDOWN NOSAVE commands are available. The first can be
416# used to stop a script that did not yet called write commands. The second
417# is the only way to shut down the server in the case a write commands was
418# already issue by the script but the user don't want to wait for the natural
419# termination of the script.
115e3ff3 420#
421# Set it to 0 or a negative value for unlimited execution without warnings.
422lua-time-limit 5000
eeffcf38 423
35a60441 424################################## SLOW LOG ###################################
425
426# The Redis Slow Log is a system to log queries that exceeded a specified
427# execution time. The execution time does not include the I/O operations
428# like talking with the client, sending the reply and so forth,
429# but just the time needed to actually execute the command (this is the only
430# stage of command execution where the thread is blocked and can not serve
431# other requests in the meantime).
432#
433# You can configure the slow log with two parameters: one tells Redis
434# what is the execution time, in microseconds, to exceed in order for the
435# command to get logged, and the other parameter is the length of the
436# slow log. When a new command is logged the oldest one is removed from the
437# queue of logged commands.
438
de32c37c 439# The following time is expressed in microseconds, so 1000000 is equivalent
440# to one second. Note that a negative number disables the slow log, while
441# a value of zero forces the logging of every command.
35a60441 442slowlog-log-slower-than 10000
de32c37c 443
444# There is no limit to this length. Just be aware that it will consume memory.
445# You can reclaim memory used by the slow log with SLOWLOG RESET.
590d55a2 446slowlog-max-len 128
35a60441 447
ed9b544e 448############################### ADVANCED CONFIG ###############################
449
d3ea4c86
PN
450# Hashes are encoded using a memory efficient data structure when they have a
451# small number of entries, and the biggest entry does not exceed a given
452# threshold. These thresholds can be configured using the following directives.
453hash-max-ziplist-entries 512
454hash-max-ziplist-value 64
b3f83f12 455
6a246b1e 456# Similarly to hashes, small lists are also encoded in a special way in order
457# to save a lot of space. The special representation is only used when
458# you are under the following limits:
459list-max-ziplist-entries 512
460list-max-ziplist-value 64
461
462# Sets have a special encoding in just one case: when a set is composed
463# of just strings that happens to be integers in radix 10 in the range
464# of 64 bit signed integers.
465# The following configuration setting sets the limit in the size of the
466# set in order to use this special memory saving encoding.
467set-max-intset-entries 512
468
3ea204e1
PN
469# Similarly to hashes and lists, sorted sets are also specially encoded in
470# order to save a lot of space. This encoding is only used when the length and
471# elements of a sorted set are below the following limits:
472zset-max-ziplist-entries 128
473zset-max-ziplist-value 64
474
8ca3e9d1 475# Active rehashing uses 1 millisecond every 100 milliseconds of CPU time in
476# order to help rehashing the main Redis hash table (the one mapping top-level
57c0cf8b 477# keys to values). The hash table implementation Redis uses (see dict.c)
8ca3e9d1 478# performs a lazy rehashing: the more operation you run into an hash table
57c0cf8b 479# that is rehashing, the more rehashing "steps" are performed, so if the
8ca3e9d1 480# server is idle the rehashing is never complete and some more memory is used
481# by the hash table.
482#
483# The default is to use this millisecond 10 times every second in order to
484# active rehashing the main dictionaries, freeing memory when possible.
485#
486# If unsure:
487# use "activerehashing no" if you have hard latency requirements and it is
488# not a good thing in your environment that Redis can reply form time to time
489# to queries with 2 milliseconds delay.
490#
491# use "activerehashing yes" if you don't have such hard requirements but
492# want to free memory asap when possible.
493activerehashing yes
494
c8a607f2 495# The client output buffer limits can be used to force disconnection of clients
496# that are not reading data from the server fast enough for some reason (a
497# common reason is that a Pub/Sub client can't consume messages as fast as the
498# publisher can produce them).
499#
500# The limit can be set differently for the three different classes of clients:
501#
502# normal -> normal clients
503# slave -> slave clients and MONITOR clients
504# pubsub -> clients subcribed to at least one pubsub channel or pattern
505#
506# The syntax of every client-output-buffer-limit directive is the following:
507#
3cbce4f4 508# client-output-buffer-limit <class> <hard limit> <soft limit> <soft seconds>
c8a607f2 509#
510# A client is immediately disconnected once the hard limit is reached, or if
511# the soft limit is reached and remains reached for the specified number of
512# seconds (continuously).
513# So for instance if the hard limit is 32 megabytes and the soft limit is
514# 16 megabytes / 10 seconds, the client will get disconnected immediately
515# if the size of the output buffers reach 32 megabytes, but will also get
516# disconnected if the client reaches 16 megabytes and continuously overcomes
517# the limit for 10 seconds.
518#
519# By default normal clients are not limited because they don't receive data
520# without asking (in a push way), but just after a request, so only
521# asynchronous clients may create a scenario where data is requested faster
522# than it can read.
523#
524# Instead there is a default limit for pubsub and slave clients, since
525# subscribers and slaves receive data in a push fashion.
526#
527# Both the hard or the soft limit can be disabled just setting it to zero.
528client-output-buffer-limit normal 0 0 0
529client-output-buffer-limit slave 256mb 64mb 60
530client-output-buffer-limit pubsub 32mb 8mb 60
531
b3f83f12
JZ
532################################## INCLUDES ###################################
533
534# Include one or more other config files here. This is useful if you
57c0cf8b 535# have a standard template that goes to all Redis server but also need
b3f83f12
JZ
536# to customize a few per-server settings. Include files can include
537# other files, so use this wisely.
538#
539# include /path/to/local.conf
540# include /path/to/other.conf