Skip to content
This repository
Newer
Older
100644 266 lines (228 sloc) 11.563 kb
ed9b544e »
2009-03-22 first commit
1 # Redis configuration file example
2
3 # By default Redis does not run as a daemon. Use 'yes' if you need it.
4 # Note that Redis will write a pid file in /var/run/redis.pid when daemonized.
5 daemonize no
6
ed329fcf »
2009-03-23 Allow to specify the pid file from the config file.
7 # When run as a daemon, Redis write a pid file in /var/run/redis.pid by default.
8 # You can specify a custom pid file location here.
9 pidfile /var/run/redis.pid
10
ed9b544e »
2009-03-22 first commit
11 # Accept connections on the specified port, default is 6379
12 port 6379
13
14 # If you want you can bind a single interface, if the bind option is not
15 # specified all the interfaces will listen for connections.
16 #
17 # bind 127.0.0.1
18
0150db36 »
2009-05-19 Allow timeout=0 config to disable client timeouts
19 # Close the connection after a client is idle for N seconds (0 to disable)
ed9b544e »
2009-03-22 first commit
20 timeout 300
21
121f70cf »
2009-12-16 lzf compression switched off by default now, with config file option …
22 # Set server verbosity to 'debug'
23 # it can be one of:
24 # debug (a lot of information, useful for development/testing)
38aba9a1 »
2010-01-06 a new default redis.conf
25 # verbose (many rarely useful info, but not a mess like the debug level)
121f70cf »
2009-12-16 lzf compression switched off by default now, with config file option …
26 # notice (moderately verbose, what you want in production probably)
27 # warning (only very important / critical messages are logged)
38aba9a1 »
2010-01-06 a new default redis.conf
28 loglevel verbose
121f70cf »
2009-12-16 lzf compression switched off by default now, with config file option …
29
30 # Specify the log file name. Also 'stdout' can be used to force
31 # the demon to log on the standard output. Note that if you use standard
32 # output for logging but daemonize, logs will be sent to /dev/null
33 logfile stdout
34
35 # Set the number of databases. The default database is DB 0, you can select
36 # a different one on a per-connection basis using SELECT <dbid> where
37 # dbid is a number between 0 and 'databases'-1
38 databases 16
39
40 ################################ SNAPSHOTTING #################################
41 #
ed9b544e »
2009-03-22 first commit
42 # Save the DB on disk:
43 #
44 # save <seconds> <changes>
45 #
46 # Will save the DB if both the given number of seconds and the given
47 # number of write operations against the DB occurred.
48 #
49 # In the example below the behaviour will be to save:
50 # after 900 sec (15 min) if at least 1 key changed
51 # after 300 sec (5 min) if at least 10 keys changed
52 # after 60 sec if at least 10000 keys changed
e7546c63 »
2010-01-05 BGREWRITEAOF now works with swapping on
53 #
54 # Note: you can disable saving at all commenting all the "save" lines.
55
38aba9a1 »
2010-01-06 a new default redis.conf
56 save 900 1
57 save 300 10
58 save 60 10000
ed9b544e »
2009-03-22 first commit
59
121f70cf »
2009-12-16 lzf compression switched off by default now, with config file option …
60 # Compress string objects using LZF when dump .rdb databases?
b0553789 »
2009-12-16 LZF compression re-enabled by default, but with INIT_HTAB set to 0 to…
61 # For default that's set to 'yes' as it's almost always a win.
62 # If you want to save some CPU in the saving child set it to 'no' but
63 # the dataset will likely be bigger if you have compressible values or keys.
64 rdbcompression yes
121f70cf »
2009-12-16 lzf compression switched off by default now, with config file option …
65
b8b553c8 »
2009-04-27 ability to specify a different file name for the DB
66 # The filename where to dump the DB
67 dbfilename dump.rdb
68
ed9b544e »
2009-03-22 first commit
69 # For default save/load DB in/from the working directory
70 # Note that you must specify a directory not a file name.
71 dir ./
72
73 ################################# REPLICATION #################################
74
75 # Master-Slave replication. Use slaveof to make a Redis instance a copy of
76 # another Redis server. Note that the configuration is local to the slave
77 # so for example it is possible to configure the slave to save the DB with a
78 # different interval, or to listen to another port, and so on.
3f477979 »
2009-12-12 TODO list modified. What's planned for 1.4 is now written in the ston…
79 #
ed9b544e »
2009-03-22 first commit
80 # slaveof <masterip> <masterport>
81
3f477979 »
2009-12-12 TODO list modified. What's planned for 1.4 is now written in the ston…
82 # If the master is password protected (using the "requirepass" configuration
83 # directive below) it is possible to tell the slave to authenticate before
84 # starting the replication synchronization process, otherwise the master will
85 # refuse the slave request.
86 #
87 # masterauth <master-password>
88
f2aa84bd »
2009-03-26 Lua client added thanks to Daniele Alessandri
89 ################################## SECURITY ###################################
90
91 # Require clients to issue AUTH <PASSWORD> before processing any other
92 # commands. This might be useful in environments in which you do not trust
93 # others with access to the host running redis-server.
94 #
95 # This should stay commented out for backward compatibility and because most
96 # people do not need auth (e.g. they run their own servers).
3f477979 »
2009-12-12 TODO list modified. What's planned for 1.4 is now written in the ston…
97 #
290deb8b »
2009-04-08 TODO changes, minor change to default redis.conf
98 # requirepass foobared
f2aa84bd »
2009-03-26 Lua client added thanks to Daniele Alessandri
99
285add55 »
2009-05-25 maxclients implemented, see redis.conf for details
100 ################################### LIMITS ####################################
101
102 # Set the max number of connected clients at the same time. By default there
103 # is no limit, and it's up to the number of file descriptors the Redis process
92f8e882 »
2010-01-09 non-blocking VM data structures, just a start
104 # is able to open. The special value '0' means no limits.
285add55 »
2009-05-25 maxclients implemented, see redis.conf for details
105 # Once the limit is reached Redis will close all the new connections sending
106 # an error 'max number of clients reached'.
3f477979 »
2009-12-12 TODO list modified. What's planned for 1.4 is now written in the ston…
107 #
285add55 »
2009-05-25 maxclients implemented, see redis.conf for details
108 # maxclients 128
109
3fd78bcd »
2009-05-27 maxmemory implemented
110 # Don't use more memory than the specified amount of bytes.
111 # When the memory limit is reached Redis will try to remove keys with an
112 # EXPIRE set. It will try to start freeing keys that are going to expire
113 # in little time and preserve keys with a longer time to live.
114 # Redis will also try to remove objects from free lists if possible.
115 #
116 # If all this fails, Redis will start to reply with errors to commands
117 # that will use more memory, like SET, LPUSH, and so on, and will continue
118 # to reply to most read-only commands like GET.
144d479b »
2009-05-28 minor fix for Solaris boxes
119 #
120 # WARNING: maxmemory can be a good idea mainly if you want to use Redis as a
121 # 'state' server or cache, not as a real DB. When Redis is used as a real
122 # database the memory usage will grow over the weeks, it will be obvious if
123 # it is going to use too much memory in the long run, and you'll have the time
124 # to upgrade. With maxmemory after the limit is reached you'll start to get
125 # errors for write operations, and this may even lead to DB inconsistency.
3f477979 »
2009-12-12 TODO list modified. What's planned for 1.4 is now written in the ston…
126 #
3fd78bcd »
2009-05-27 maxmemory implemented
127 # maxmemory <bytes>
128
44b38ef4 »
2009-10-30 Initial implementation of append-only mode. Loading still not impleme…
129 ############################## APPEND ONLY MODE ###############################
130
131 # By default Redis asynchronously dumps the dataset on disk. If you can live
132 # with the idea that the latest records will be lost if something like a crash
133 # happens this is the preferred way to run Redis. If instead you care a lot
134 # about your data and don't want to that a single record can get lost you should
135 # enable the append only mode: when this mode is enabled Redis will append
136 # every write operation received in the file appendonly.log. This file will
137 # be read on startup in order to rebuild the full dataset in memory.
138 #
139 # Note that you can have both the async dumps and the append only file if you
140 # like (you have to comment the "save" statements above to disable the dumps).
141 # Still if append only mode is enabled Redis will load the data from the
142 # log file at startup ignoring the dump.rdb file.
0154acdc »
2009-10-30 appendfsync is now set to NO by default
143 #
144 # The name of the append only file is "appendonly.log"
49b99ab4 »
2009-12-13 Changed the reply of BGSAVE and BGREWRITEAOF from +OK to a more meani…
145 #
146 # IMPORTANT: Check the BGREWRITEAOF to check how to rewrite the append
147 # log file in background when it gets too big.
44b38ef4 »
2009-10-30 Initial implementation of append-only mode. Loading still not impleme…
148
4e141d5a »
2009-10-30 append only mode is now able to translate EXPIRE into EXPIREAT transp…
149 appendonly no
44b38ef4 »
2009-10-30 Initial implementation of append-only mode. Loading still not impleme…
150
4e141d5a »
2009-10-30 append only mode is now able to translate EXPIRE into EXPIREAT transp…
151 # The fsync() call tells the Operating System to actually write data on disk
48f0308a »
2009-10-30 support for appendonly mode no, always, everysec
152 # instead to wait for more data in the output buffer. Some OS will really flush
153 # data on disk, some other OS will just try to do it ASAP.
154 #
155 # Redis supports three different modes:
156 #
157 # no: don't fsync, just let the OS flush the data when it wants. Faster.
158 # always: fsync after every write to the append only log . Slow, Safest.
159 # everysec: fsync only if one second passed since the last fsync. Compromise.
160 #
4e141d5a »
2009-10-30 append only mode is now able to translate EXPIRE into EXPIREAT transp…
161 # The default is "always" that's the safer of the options. It's up to you to
162 # understand if you can relax this to "everysec" that will fsync every second
163 # or to "no" that will let the operating system flush the output buffer when
164 # it want, for better performances (but if you can live with the idea of
165 # some data loss consider the default persistence mode that's snapshotting).
48f0308a »
2009-10-30 support for appendonly mode no, always, everysec
166
4e141d5a »
2009-10-30 append only mode is now able to translate EXPIRE into EXPIREAT transp…
167 appendfsync always
48f0308a »
2009-10-30 support for appendonly mode no, always, everysec
168 # appendfsync everysec
4e141d5a »
2009-10-30 append only mode is now able to translate EXPIRE into EXPIREAT transp…
169 # appendfsync no
48f0308a »
2009-10-30 support for appendonly mode no, always, everysec
170
a35ddf12 »
2010-01-05 more object-level VM primitives
171 ################################ VIRTUAL MEMORY ###############################
172
4ef8de8a »
2010-01-05 basic VM mostly working!
173 # Virtual Memory allows Redis to work with datasets bigger than the actual
174 # amount of RAM needed to hold the whole dataset in memory.
175 # In order to do so very used keys are taken in memory while the other keys
176 # are swapped into a swap file, similarly to what operating systems do
177 # with memory pages.
178 #
179 # To enable VM just set 'vm-enabled' to yes, and set the following three
180 # VM parameters accordingly to your needs.
c9e5c23d »
2010-01-06 vm-enabled set to no by default in redis.conf
181
182 vm-enabled no
183 # vm-enabled yes
4ef8de8a »
2010-01-05 basic VM mostly working!
184
054e426d »
2010-01-15 support for named VM swap file. Fixed a few important interaction iss…
185 # This is the path of the Redis swap file. As you can guess, swap files
186 # can't be shared by different Redis instances, so make sure to use a swap
187 # file for every redis process you are running.
188 #
189 # The swap file name may contain "%p" that is substituted with the PID of
190 # the Redis process, so the default name /tmp/redis-%p.vm will work even
191 # with multiple instances as Redis will use, for example, redis-811.vm
192 # for one instance and redis-593.vm for another one.
193 #
194 # Useless to say, the best kind of disk for a Redis swap file (that's accessed
195 # at random) is a Solid State Disk (SSD).
196 vm-swap-file /tmp/redis-%p.vm
197
4ef8de8a »
2010-01-05 basic VM mostly working!
198 # vm-max-memory configures the VM to use at max the specified amount of
199 # RAM. Everything that deos not fit will be swapped on disk *if* possible, that
200 # is, if there is still enough contiguous space in the swap file.
38aba9a1 »
2010-01-06 a new default redis.conf
201 #
202 # With vm-max-memory 10000000 the system will swap almost everything it
203 # can. Not a good default, just specify the max amount of RAM you can
204 # in bytes, but it's better to leave some margin. For instance specify
205 # an amount of RAM that's more or less 80% of your free RAM.
4ef8de8a »
2010-01-05 basic VM mostly working!
206 vm-max-memory 10000000
207
208 # Redis swap files is split into pages. An object can be saved using multiple
209 # contiguous pages, but pages can't be shared between different objects.
210 # So if your page is too big, small objects swapped out on disk will waste
211 # a lot of space. If you page is too small, there is less space in the swap
212 # file (assuming you configured the same number of total swap file pages).
213 #
214 # If you use a lot of small objects, use a page size of 64 or 32 bytes.
215 # If you use a lot of big objects, use a bigger page size.
92f8e882 »
2010-01-09 non-blocking VM data structures, just a start
216 # If unsure, use the default :)
4ef8de8a »
2010-01-05 basic VM mostly working!
217 vm-page-size 256
218
219 # Number of total memory pages in the swap file.
220 # Given that the page table (a bitmap of free/used pages) is taken in memory,
221 # every 8 pages on disk will consume 1 byte of RAM.
222 #
223 # The total swap size is vm-page-size * vm-pages
224 #
225 # With the default of 256-bytes memory pages and 104857600 pages Redis will
92f8e882 »
2010-01-09 non-blocking VM data structures, just a start
226 # use a 25 GB swap file, that will use roughly 13 MB of RAM for the page table.
38aba9a1 »
2010-01-06 a new default redis.conf
227 #
228 # It's better to use the smallest acceptable value for your application,
229 # but the default is large in order to work in most conditions.
230 vm-pages 104857600
a35ddf12 »
2010-01-05 more object-level VM primitives
231
92f8e882 »
2010-01-09 non-blocking VM data structures, just a start
232 # Max number of VM I/O threads running at the same time.
233 # This threads are used to read/write data from/to swap file, since they
234 # also encode and decode objects from disk to memory or the reverse, a bigger
235 # number of threads can help with big objects even if they can't help with
236 # I/O itself as the physical device may not be able to couple with many
237 # reads/writes operations at the same time.
72e9fd40 »
2010-01-12 support for blocking VM in config file
238 #
239 # The special value of 0 turn off threaded I/O and enables the blocking
240 # Virtual Memory implementation.
92f8e882 »
2010-01-09 non-blocking VM data structures, just a start
241 vm-max-threads 4
242
ed9b544e »
2009-03-22 first commit
243 ############################### ADVANCED CONFIG ###############################
244
245 # Glue small output buffers together in order to send small replies in a
246 # single TCP packet. Uses a bit more CPU but most of the times it is a win
247 # in terms of number of queries per second. Use 'yes' if unsure.
248 glueoutputbuf yes
10c43610 »
2009-03-25 Nasty bug of the new DB format fixed, objects sharing implemented
249
250 # Use object sharing. Can save a lot of memory if you have many common
251 # string in your dataset, but performs lookups against the shared objects
252 # pool so it uses more CPU and can be a bit slower. Usually it's a good
253 # idea.
e52c65b9 »
2009-06-03 shareobjectspoolsize implemented in reds.conf, in order to control th…
254 #
255 # When object sharing is enabled (shareobjects yes) you can use
256 # shareobjectspoolsize to control the size of the pool used in order to try
257 # object sharing. A bigger pool size will lead to better sharing capabilities.
258 # In general you want this value to be at least the double of the number of
259 # very common strings you have in your dataset.
260 #
261 # WARNING: object sharing is experimental, don't enable this feature
262 # in production before of Redis 1.0-stable. Still please try this feature in
263 # your development environment so that we can test it better.
10c43610 »
2009-03-25 Nasty bug of the new DB format fixed, objects sharing implemented
264 shareobjects no
e52c65b9 »
2009-06-03 shareobjectspoolsize implemented in reds.conf, in order to control th…
265 shareobjectspoolsize 1024
Something went wrong with that request. Please try again.