Skip to content
Permalink
1d13ff0b39
Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
Go to file
 
 
Cannot retrieve contributors at this time
5079 lines (4447 sloc) 183 KB
/*
* Copyright (c) 2009-2016, Salvatore Sanfilippo <antirez at gmail dot com>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include "server.h"
#include "cluster.h"
#include "slowlog.h"
#include "bio.h"
#include "latency.h"
#include "atomicvar.h"
#include <time.h>
#include <signal.h>
#include <sys/wait.h>
#include <errno.h>
#include <assert.h>
#include <ctype.h>
#include <stdarg.h>
#include <arpa/inet.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <sys/time.h>
#include <sys/resource.h>
#include <sys/uio.h>
#include <sys/un.h>
#include <limits.h>
#include <float.h>
#include <math.h>
#include <sys/resource.h>
#include <sys/utsname.h>
#include <locale.h>
#include <sys/socket.h>
/* Our shared "common" objects */
struct sharedObjectsStruct shared;
/* Global vars that are actually used as constants. The following double
* values are used for double on-disk serialization, and are initialized
* at runtime to avoid strange compiler optimizations. */
double R_Zero, R_PosInf, R_NegInf, R_Nan;
/*================================= Globals ================================= */
/* Global vars */
struct redisServer server; /* Server global state */
volatile unsigned long lru_clock; /* Server global current LRU time. */
/* Our command table.
*
* Every entry is composed of the following fields:
*
* name: A string representing the command name.
*
* function: Pointer to the C function implementing the command.
*
* arity: Number of arguments, it is possible to use -N to say >= N
*
* sflags: Command flags as string. See below for a table of flags.
*
* flags: Flags as bitmask. Computed by Redis using the 'sflags' field.
*
* get_keys_proc: An optional function to get key arguments from a command.
* This is only used when the following three fields are not
* enough to specify what arguments are keys.
*
* first_key_index: First argument that is a key
*
* last_key_index: Last argument that is a key
*
* key_step: Step to get all the keys from first to last argument.
* For instance in MSET the step is two since arguments
* are key,val,key,val,...
*
* microseconds: Microseconds of total execution time for this command.
*
* calls: Total number of calls of this command.
*
* id: Command bit identifier for ACLs or other goals.
*
* The flags, microseconds and calls fields are computed by Redis and should
* always be set to zero.
*
* Command flags are expressed using space separated strings, that are turned
* into actual flags by the populateCommandTable() function.
*
* This is the meaning of the flags:
*
* write: Write command (may modify the key space).
*
* read-only: All the non special commands just reading from keys without
* changing the content, or returning other informations like
* the TIME command. Special commands such administrative commands
* or transaction related commands (multi, exec, discard, ...)
* are not flagged as read-only commands, since they affect the
* server or the connection in other ways.
*
* use-memory: May increase memory usage once called. Don't allow if out
* of memory.
*
* admin: Administrative command, like SAVE or SHUTDOWN.
*
* pub-sub: Pub/Sub related command.
*
* no-script: Command not allowed in scripts.
*
* random: Random command. Command is not deterministic, that is, the same
* command with the same arguments, with the same key space, may
* have different results. For instance SPOP and RANDOMKEY are
* two random commands.
*
* to-sort: Sort command output array if called from script, so that the
* output is deterministic. When this flag is used (not always
* possible), then the "random" flag is not needed.
*
* ok-loading: Allow the command while loading the database.
*
* ok-stale: Allow the command while a slave has stale data but is not
* allowed to serve this data. Normally no command is accepted
* in this condition but just a few.
*
* no-monitor: Do not automatically propagate the command on MONITOR.
*
* no-slowlog: Do not automatically propagate the command to the slowlog.
*
* cluster-asking: Perform an implicit ASKING for this command, so the
* command will be accepted in cluster mode if the slot is marked
* as 'importing'.
*
* fast: Fast command: O(1) or O(log(N)) command that should never
* delay its execution as long as the kernel scheduler is giving
* us time. Note that commands that may trigger a DEL as a side
* effect (like SET) are not fast commands.
*
* The following additional flags are only used in order to put commands
* in a specific ACL category. Commands can have multiple ACL categories.
*
* @keyspace, @read, @write, @set, @sortedset, @list, @hash, @string, @bitmap,
* @hyperloglog, @stream, @admin, @fast, @slow, @pubsub, @blocking, @dangerous,
* @connection, @transaction, @scripting, @geo.
*
* Note that:
*
* 1) The read-only flag implies the @read ACL category.
* 2) The write flag implies the @write ACL category.
* 3) The fast flag implies the @fast ACL category.
* 4) The admin flag implies the @admin and @dangerous ACL category.
* 5) The pub-sub flag implies the @pubsub ACL category.
* 6) The lack of fast flag implies the @slow ACL category.
* 7) The non obvious "keyspace" category includes the commands
* that interact with keys without having anything to do with
* specific data structures, such as: DEL, RENAME, MOVE, SELECT,
* TYPE, EXPIRE*, PEXPIRE*, TTL, PTTL, ...
*/
struct redisCommand redisCommandTable[] = {
{"module",moduleCommand,-2,
"admin no-script",
0,NULL,0,0,0,0,0,0},
{"get",getCommand,2,
"read-only fast @string",
0,NULL,1,1,1,0,0,0},
/* Note that we can't flag set as fast, since it may perform an
* implicit DEL of a large key. */
{"set",setCommand,-3,
"write use-memory @string",
0,NULL,1,1,1,0,0,0},
{"setnx",setnxCommand,3,
"write use-memory fast @string",
0,NULL,1,1,1,0,0,0},
{"setex",setexCommand,4,
"write use-memory @string",
0,NULL,1,1,1,0,0,0},
{"psetex",psetexCommand,4,
"write use-memory @string",
0,NULL,1,1,1,0,0,0},
{"append",appendCommand,3,
"write use-memory fast @string",
0,NULL,1,1,1,0,0,0},
{"strlen",strlenCommand,2,
"read-only fast @string",
0,NULL,1,1,1,0,0,0},
{"del",delCommand,-2,
"write @keyspace",
0,NULL,1,-1,1,0,0,0},
{"unlink",unlinkCommand,-2,
"write fast @keyspace",
0,NULL,1,-1,1,0,0,0},
{"exists",existsCommand,-2,
"read-only fast @keyspace",
0,NULL,1,-1,1,0,0,0},
{"setbit",setbitCommand,4,
"write use-memory @bitmap",
0,NULL,1,1,1,0,0,0},
{"getbit",getbitCommand,3,
"read-only fast @bitmap",
0,NULL,1,1,1,0,0,0},
{"bitfield",bitfieldCommand,-2,
"write use-memory @bitmap",
0,NULL,1,1,1,0,0,0},
{"bitfield_ro",bitfieldroCommand,-2,
"read-only fast @bitmap",
0,NULL,1,1,1,0,0,0},
{"setrange",setrangeCommand,4,
"write use-memory @string",
0,NULL,1,1,1,0,0,0},
{"getrange",getrangeCommand,4,
"read-only @string",
0,NULL,1,1,1,0,0,0},
{"substr",getrangeCommand,4,
"read-only @string",
0,NULL,1,1,1,0,0,0},
{"incr",incrCommand,2,
"write use-memory fast @string",
0,NULL,1,1,1,0,0,0},
{"decr",decrCommand,2,
"write use-memory fast @string",
0,NULL,1,1,1,0,0,0},
{"mget",mgetCommand,-2,
"read-only fast @string",
0,NULL,1,-1,1,0,0,0},
{"rpush",rpushCommand,-3,
"write use-memory fast @list",
0,NULL,1,1,1,0,0,0},
{"lpush",lpushCommand,-3,
"write use-memory fast @list",
0,NULL,1,1,1,0,0,0},
{"rpushx",rpushxCommand,-3,
"write use-memory fast @list",
0,NULL,1,1,1,0,0,0},
{"lpushx",lpushxCommand,-3,
"write use-memory fast @list",
0,NULL,1,1,1,0,0,0},
{"linsert",linsertCommand,5,
"write use-memory @list",
0,NULL,1,1,1,0,0,0},
{"rpop",rpopCommand,2,
"write fast @list",
0,NULL,1,1,1,0,0,0},
{"lpop",lpopCommand,2,
"write fast @list",
0,NULL,1,1,1,0,0,0},
{"brpop",brpopCommand,-3,
"write no-script @list @blocking",
0,NULL,1,-2,1,0,0,0},
{"brpoplpush",brpoplpushCommand,4,
"write use-memory no-script @list @blocking",
0,NULL,1,2,1,0,0,0},
{"blpop",blpopCommand,-3,
"write no-script @list @blocking",
0,NULL,1,-2,1,0,0,0},
{"llen",llenCommand,2,
"read-only fast @list",
0,NULL,1,1,1,0,0,0},
{"lindex",lindexCommand,3,
"read-only @list",
0,NULL,1,1,1,0,0,0},
{"lset",lsetCommand,4,
"write use-memory @list",
0,NULL,1,1,1,0,0,0},
{"lrange",lrangeCommand,4,
"read-only @list",
0,NULL,1,1,1,0,0,0},
{"ltrim",ltrimCommand,4,
"write @list",
0,NULL,1,1,1,0,0,0},
{"lrem",lremCommand,4,
"write @list",
0,NULL,1,1,1,0,0,0},
{"rpoplpush",rpoplpushCommand,3,
"write use-memory @list",
0,NULL,1,2,1,0,0,0},
{"sadd",saddCommand,-3,
"write use-memory fast @set",
0,NULL,1,1,1,0,0,0},
{"srem",sremCommand,-3,
"write fast @set",
0,NULL,1,1,1,0,0,0},
{"smove",smoveCommand,4,
"write fast @set",
0,NULL,1,2,1,0,0,0},
{"sismember",sismemberCommand,3,
"read-only fast @set",
0,NULL,1,1,1,0,0,0},
{"scard",scardCommand,2,
"read-only fast @set",
0,NULL,1,1,1,0,0,0},
{"spop",spopCommand,-2,
"write random fast @set",
0,NULL,1,1,1,0,0,0},
{"srandmember",srandmemberCommand,-2,
"read-only random @set",
0,NULL,1,1,1,0,0,0},
{"sinter",sinterCommand,-2,
"read-only to-sort @set",
0,NULL,1,-1,1,0,0,0},
{"sinterstore",sinterstoreCommand,-3,
"write use-memory @set",
0,NULL,1,-1,1,0,0,0},
{"sunion",sunionCommand,-2,
"read-only to-sort @set",
0,NULL,1,-1,1,0,0,0},
{"sunionstore",sunionstoreCommand,-3,
"write use-memory @set",
0,NULL,1,-1,1,0,0,0},
{"sdiff",sdiffCommand,-2,
"read-only to-sort @set",
0,NULL,1,-1,1,0,0,0},
{"sdiffstore",sdiffstoreCommand,-3,
"write use-memory @set",
0,NULL,1,-1,1,0,0,0},
{"smembers",sinterCommand,2,
"read-only to-sort @set",
0,NULL,1,1,1,0,0,0},
{"sscan",sscanCommand,-3,
"read-only random @set",
0,NULL,1,1,1,0,0,0},
{"zadd",zaddCommand,-4,
"write use-memory fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zincrby",zincrbyCommand,4,
"write use-memory fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zrem",zremCommand,-3,
"write fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zremrangebyscore",zremrangebyscoreCommand,4,
"write @sortedset",
0,NULL,1,1,1,0,0,0},
{"zremrangebyrank",zremrangebyrankCommand,4,
"write @sortedset",
0,NULL,1,1,1,0,0,0},
{"zremrangebylex",zremrangebylexCommand,4,
"write @sortedset",
0,NULL,1,1,1,0,0,0},
{"zunionstore",zunionstoreCommand,-4,
"write use-memory @sortedset",
0,zunionInterGetKeys,0,0,0,0,0,0},
{"zinterstore",zinterstoreCommand,-4,
"write use-memory @sortedset",
0,zunionInterGetKeys,0,0,0,0,0,0},
{"zrange",zrangeCommand,-4,
"read-only @sortedset",
0,NULL,1,1,1,0,0,0},
{"zrangebyscore",zrangebyscoreCommand,-4,
"read-only @sortedset",
0,NULL,1,1,1,0,0,0},
{"zrevrangebyscore",zrevrangebyscoreCommand,-4,
"read-only @sortedset",
0,NULL,1,1,1,0,0,0},
{"zrangebylex",zrangebylexCommand,-4,
"read-only @sortedset",
0,NULL,1,1,1,0,0,0},
{"zrevrangebylex",zrevrangebylexCommand,-4,
"read-only @sortedset",
0,NULL,1,1,1,0,0,0},
{"zcount",zcountCommand,4,
"read-only fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zlexcount",zlexcountCommand,4,
"read-only fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zrevrange",zrevrangeCommand,-4,
"read-only @sortedset",
0,NULL,1,1,1,0,0,0},
{"zcard",zcardCommand,2,
"read-only fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zscore",zscoreCommand,3,
"read-only fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zrank",zrankCommand,3,
"read-only fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zrevrank",zrevrankCommand,3,
"read-only fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zscan",zscanCommand,-3,
"read-only random @sortedset",
0,NULL,1,1,1,0,0,0},
{"zpopmin",zpopminCommand,-2,
"write fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"zpopmax",zpopmaxCommand,-2,
"write fast @sortedset",
0,NULL,1,1,1,0,0,0},
{"bzpopmin",bzpopminCommand,-3,
"write no-script fast @sortedset @blocking",
0,NULL,1,-2,1,0,0,0},
{"bzpopmax",bzpopmaxCommand,-3,
"write no-script fast @sortedset @blocking",
0,NULL,1,-2,1,0,0,0},
{"hset",hsetCommand,-4,
"write use-memory fast @hash",
0,NULL,1,1,1,0,0,0},
{"hsetnx",hsetnxCommand,4,
"write use-memory fast @hash",
0,NULL,1,1,1,0,0,0},
{"hget",hgetCommand,3,
"read-only fast @hash",
0,NULL,1,1,1,0,0,0},
{"hmset",hsetCommand,-4,
"write use-memory fast @hash",
0,NULL,1,1,1,0,0,0},
{"hmget",hmgetCommand,-3,
"read-only fast @hash",
0,NULL,1,1,1,0,0,0},
{"hincrby",hincrbyCommand,4,
"write use-memory fast @hash",
0,NULL,1,1,1,0,0,0},
{"hincrbyfloat",hincrbyfloatCommand,4,
"write use-memory fast @hash",
0,NULL,1,1,1,0,0,0},
{"hdel",hdelCommand,-3,
"write fast @hash",
0,NULL,1,1,1,0,0,0},
{"hlen",hlenCommand,2,
"read-only fast @hash",
0,NULL,1,1,1,0,0,0},
{"hstrlen",hstrlenCommand,3,
"read-only fast @hash",
0,NULL,1,1,1,0,0,0},
{"hkeys",hkeysCommand,2,
"read-only to-sort @hash",
0,NULL,1,1,1,0,0,0},
{"hvals",hvalsCommand,2,
"read-only to-sort @hash",
0,NULL,1,1,1,0,0,0},
{"hgetall",hgetallCommand,2,
"read-only random @hash",
0,NULL,1,1,1,0,0,0},
{"hexists",hexistsCommand,3,
"read-only fast @hash",
0,NULL,1,1,1,0,0,0},
{"hscan",hscanCommand,-3,
"read-only random @hash",
0,NULL,1,1,1,0,0,0},
{"incrby",incrbyCommand,3,
"write use-memory fast @string",
0,NULL,1,1,1,0,0,0},
{"decrby",decrbyCommand,3,
"write use-memory fast @string",
0,NULL,1,1,1,0,0,0},
{"incrbyfloat",incrbyfloatCommand,3,
"write use-memory fast @string",
0,NULL,1,1,1,0,0,0},
{"getset",getsetCommand,3,
"write use-memory fast @string",
0,NULL,1,1,1,0,0,0},
{"mset",msetCommand,-3,
"write use-memory @string",
0,NULL,1,-1,2,0,0,0},
{"msetnx",msetnxCommand,-3,
"write use-memory @string",
0,NULL,1,-1,2,0,0,0},
{"randomkey",randomkeyCommand,1,
"read-only random @keyspace",
0,NULL,0,0,0,0,0,0},
{"select",selectCommand,2,
"ok-loading fast ok-stale @keyspace",
0,NULL,0,0,0,0,0,0},
{"swapdb",swapdbCommand,3,
"write fast @keyspace @dangerous",
0,NULL,0,0,0,0,0,0},
{"move",moveCommand,3,
"write fast @keyspace",
0,NULL,1,1,1,0,0,0},
/* Like for SET, we can't mark rename as a fast command because
* overwriting the target key may result in an implicit slow DEL. */
{"rename",renameCommand,3,
"write @keyspace",
0,NULL,1,2,1,0,0,0},
{"renamenx",renamenxCommand,3,
"write fast @keyspace",
0,NULL,1,2,1,0,0,0},
{"expire",expireCommand,3,
"write fast @keyspace",
0,NULL,1,1,1,0,0,0},
{"expireat",expireatCommand,3,
"write fast @keyspace",
0,NULL,1,1,1,0,0,0},
{"pexpire",pexpireCommand,3,
"write fast @keyspace",
0,NULL,1,1,1,0,0,0},
{"pexpireat",pexpireatCommand,3,
"write fast @keyspace",
0,NULL,1,1,1,0,0,0},
{"keys",keysCommand,2,
"read-only to-sort @keyspace @dangerous",
0,NULL,0,0,0,0,0,0},
{"scan",scanCommand,-2,
"read-only random @keyspace",
0,NULL,0,0,0,0,0,0},
{"dbsize",dbsizeCommand,1,
"read-only fast @keyspace",
0,NULL,0,0,0,0,0,0},
{"auth",authCommand,-2,
"no-auth no-script ok-loading ok-stale fast no-monitor no-slowlog @connection",
0,NULL,0,0,0,0,0,0},
/* We don't allow PING during loading since in Redis PING is used as
* failure detection, and a loading server is considered to be
* not available. */
{"ping",pingCommand,-1,
"ok-stale fast @connection",
0,NULL,0,0,0,0,0,0},
{"echo",echoCommand,2,
"read-only fast @connection",
0,NULL,0,0,0,0,0,0},
{"save",saveCommand,1,
"admin no-script",
0,NULL,0,0,0,0,0,0},
{"bgsave",bgsaveCommand,-1,
"admin no-script",
0,NULL,0,0,0,0,0,0},
{"bgrewriteaof",bgrewriteaofCommand,1,
"admin no-script",
0,NULL,0,0,0,0,0,0},
{"shutdown",shutdownCommand,-1,
"admin no-script ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"lastsave",lastsaveCommand,1,
"read-only random fast ok-loading ok-stale @admin @dangerous",
0,NULL,0,0,0,0,0,0},
{"type",typeCommand,2,
"read-only fast @keyspace",
0,NULL,1,1,1,0,0,0},
{"multi",multiCommand,1,
"no-script fast @transaction",
0,NULL,0,0,0,0,0,0},
{"exec",execCommand,1,
"no-script no-monitor no-slowlog @transaction",
0,NULL,0,0,0,0,0,0},
{"discard",discardCommand,1,
"no-script fast @transaction",
0,NULL,0,0,0,0,0,0},
{"sync",syncCommand,1,
"admin no-script",
0,NULL,0,0,0,0,0,0},
{"psync",syncCommand,3,
"admin no-script",
0,NULL,0,0,0,0,0,0},
{"replconf",replconfCommand,-1,
"admin no-script ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"flushdb",flushdbCommand,-1,
"write @keyspace @dangerous",
0,NULL,0,0,0,0,0,0},
{"flushall",flushallCommand,-1,
"write @keyspace @dangerous",
0,NULL,0,0,0,0,0,0},
{"sort",sortCommand,-2,
"write use-memory @list @set @sortedset @dangerous",
0,sortGetKeys,1,1,1,0,0,0},
{"info",infoCommand,-1,
"ok-loading ok-stale random @dangerous",
0,NULL,0,0,0,0,0,0},
{"monitor",monitorCommand,1,
"admin no-script ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"ttl",ttlCommand,2,
"read-only fast random @keyspace",
0,NULL,1,1,1,0,0,0},
{"touch",touchCommand,-2,
"read-only fast @keyspace",
0,NULL,1,-1,1,0,0,0},
{"pttl",pttlCommand,2,
"read-only fast random @keyspace",
0,NULL,1,1,1,0,0,0},
{"persist",persistCommand,2,
"write fast @keyspace",
0,NULL,1,1,1,0,0,0},
{"slaveof",replicaofCommand,3,
"admin no-script ok-stale",
0,NULL,0,0,0,0,0,0},
{"replicaof",replicaofCommand,3,
"admin no-script ok-stale",
0,NULL,0,0,0,0,0,0},
{"role",roleCommand,1,
"ok-loading ok-stale no-script fast read-only @dangerous",
0,NULL,0,0,0,0,0,0},
{"debug",debugCommand,-2,
"admin no-script ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"config",configCommand,-2,
"admin ok-loading ok-stale no-script",
0,NULL,0,0,0,0,0,0},
{"subscribe",subscribeCommand,-2,
"pub-sub no-script ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"unsubscribe",unsubscribeCommand,-1,
"pub-sub no-script ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"psubscribe",psubscribeCommand,-2,
"pub-sub no-script ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"punsubscribe",punsubscribeCommand,-1,
"pub-sub no-script ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"publish",publishCommand,3,
"pub-sub ok-loading ok-stale fast",
0,NULL,0,0,0,0,0,0},
{"pubsub",pubsubCommand,-2,
"pub-sub ok-loading ok-stale random",
0,NULL,0,0,0,0,0,0},
{"watch",watchCommand,-2,
"no-script fast @transaction",
0,NULL,1,-1,1,0,0,0},
{"unwatch",unwatchCommand,1,
"no-script fast @transaction",
0,NULL,0,0,0,0,0,0},
{"cluster",clusterCommand,-2,
"admin ok-stale random",
0,NULL,0,0,0,0,0,0},
{"restore",restoreCommand,-4,
"write use-memory @keyspace @dangerous",
0,NULL,1,1,1,0,0,0},
{"restore-asking",restoreCommand,-4,
"write use-memory cluster-asking @keyspace @dangerous",
0,NULL,1,1,1,0,0,0},
{"migrate",migrateCommand,-6,
"write random @keyspace @dangerous",
0,migrateGetKeys,0,0,0,0,0,0},
{"asking",askingCommand,1,
"fast @keyspace",
0,NULL,0,0,0,0,0,0},
{"readonly",readonlyCommand,1,
"fast @keyspace",
0,NULL,0,0,0,0,0,0},
{"readwrite",readwriteCommand,1,
"fast @keyspace",
0,NULL,0,0,0,0,0,0},
{"dump",dumpCommand,2,
"read-only random @keyspace",
0,NULL,1,1,1,0,0,0},
{"object",objectCommand,-2,
"read-only random @keyspace",
0,NULL,2,2,1,0,0,0},
{"memory",memoryCommand,-2,
"random read-only",
0,memoryGetKeys,0,0,0,0,0,0},
{"client",clientCommand,-2,
"admin no-script random ok-loading ok-stale @connection",
0,NULL,0,0,0,0,0,0},
{"hello",helloCommand,-2,
"no-auth no-script fast no-monitor ok-loading ok-stale no-slowlog @connection",
0,NULL,0,0,0,0,0,0},
/* EVAL can modify the dataset, however it is not flagged as a write
* command since we do the check while running commands from Lua. */
{"eval",evalCommand,-3,
"no-script @scripting",
0,evalGetKeys,0,0,0,0,0,0},
{"evalsha",evalShaCommand,-3,
"no-script @scripting",
0,evalGetKeys,0,0,0,0,0,0},
{"slowlog",slowlogCommand,-2,
"admin random ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"script",scriptCommand,-2,
"no-script @scripting",
0,NULL,0,0,0,0,0,0},
{"time",timeCommand,1,
"read-only random fast ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"bitop",bitopCommand,-4,
"write use-memory @bitmap",
0,NULL,2,-1,1,0,0,0},
{"bitcount",bitcountCommand,-2,
"read-only @bitmap",
0,NULL,1,1,1,0,0,0},
{"bitpos",bitposCommand,-3,
"read-only @bitmap",
0,NULL,1,1,1,0,0,0},
{"wait",waitCommand,3,
"no-script @keyspace",
0,NULL,0,0,0,0,0,0},
{"command",commandCommand,-1,
"ok-loading ok-stale random @connection",
0,NULL,0,0,0,0,0,0},
{"geoadd",geoaddCommand,-5,
"write use-memory @geo",
0,NULL,1,1,1,0,0,0},
/* GEORADIUS has store options that may write. */
{"georadius",georadiusCommand,-6,
"write @geo",
0,georadiusGetKeys,1,1,1,0,0,0},
{"georadius_ro",georadiusroCommand,-6,
"read-only @geo",
0,georadiusGetKeys,1,1,1,0,0,0},
{"georadiusbymember",georadiusbymemberCommand,-5,
"write @geo",
0,georadiusGetKeys,1,1,1,0,0,0},
{"georadiusbymember_ro",georadiusbymemberroCommand,-5,
"read-only @geo",
0,georadiusGetKeys,1,1,1,0,0,0},
{"geohash",geohashCommand,-2,
"read-only @geo",
0,NULL,1,1,1,0,0,0},
{"geopos",geoposCommand,-2,
"read-only @geo",
0,NULL,1,1,1,0,0,0},
{"geodist",geodistCommand,-4,
"read-only @geo",
0,NULL,1,1,1,0,0,0},
{"pfselftest",pfselftestCommand,1,
"admin @hyperloglog",
0,NULL,0,0,0,0,0,0},
{"pfadd",pfaddCommand,-2,
"write use-memory fast @hyperloglog",
0,NULL,1,1,1,0,0,0},
/* Technically speaking PFCOUNT may change the key since it changes the
* final bytes in the HyperLogLog representation. However in this case
* we claim that the representation, even if accessible, is an internal
* affair, and the command is semantically read only. */
{"pfcount",pfcountCommand,-2,
"read-only @hyperloglog",
0,NULL,1,-1,1,0,0,0},
{"pfmerge",pfmergeCommand,-2,
"write use-memory @hyperloglog",
0,NULL,1,-1,1,0,0,0},
{"pfdebug",pfdebugCommand,-3,
"admin write",
0,NULL,0,0,0,0,0,0},
{"xadd",xaddCommand,-5,
"write use-memory fast random @stream",
0,NULL,1,1,1,0,0,0},
{"xrange",xrangeCommand,-4,
"read-only @stream",
0,NULL,1,1,1,0,0,0},
{"xrevrange",xrevrangeCommand,-4,
"read-only @stream",
0,NULL,1,1,1,0,0,0},
{"xlen",xlenCommand,2,
"read-only fast @stream",
0,NULL,1,1,1,0,0,0},
{"xread",xreadCommand,-4,
"read-only no-script @stream @blocking",
0,xreadGetKeys,1,1,1,0,0,0},
{"xreadgroup",xreadCommand,-7,
"write no-script @stream @blocking",
0,xreadGetKeys,1,1,1,0,0,0},
{"xgroup",xgroupCommand,-2,
"write use-memory @stream",
0,NULL,2,2,1,0,0,0},
{"xsetid",xsetidCommand,3,
"write use-memory fast @stream",
0,NULL,1,1,1,0,0,0},
{"xack",xackCommand,-4,
"write fast random @stream",
0,NULL,1,1,1,0,0,0},
{"xpending",xpendingCommand,-3,
"read-only random @stream",
0,NULL,1,1,1,0,0,0},
{"xclaim",xclaimCommand,-6,
"write random fast @stream",
0,NULL,1,1,1,0,0,0},
{"xinfo",xinfoCommand,-2,
"read-only random @stream",
0,NULL,2,2,1,0,0,0},
{"xdel",xdelCommand,-3,
"write fast @stream",
0,NULL,1,1,1,0,0,0},
{"xtrim",xtrimCommand,-2,
"write random @stream",
0,NULL,1,1,1,0,0,0},
{"post",securityWarningCommand,-1,
"ok-loading ok-stale read-only",
0,NULL,0,0,0,0,0,0},
{"host:",securityWarningCommand,-1,
"ok-loading ok-stale read-only",
0,NULL,0,0,0,0,0,0},
{"latency",latencyCommand,-2,
"admin no-script ok-loading ok-stale",
0,NULL,0,0,0,0,0,0},
{"lolwut",lolwutCommand,-1,
"read-only fast",
0,NULL,0,0,0,0,0,0},
{"acl",aclCommand,-2,
"admin no-script no-slowlog ok-loading ok-stale",
0,NULL,0,0,0,0,0,0}
};
/*============================ Utility functions ============================ */
/* We use a private localtime implementation which is fork-safe. The logging
* function of Redis may be called from other threads. */
void nolocks_localtime(struct tm *tmp, time_t t, time_t tz, int dst);
/* Low level logging. To use only for very big messages, otherwise
* serverLog() is to prefer. */
void serverLogRaw(int level, const char *msg) {
const int syslogLevelMap[] = { LOG_DEBUG, LOG_INFO, LOG_NOTICE, LOG_WARNING };
const char *c = ".-*#";
FILE *fp;
char buf[64];
int rawmode = (level & LL_RAW);
int log_to_stdout = server.logfile[0] == '\0';
level &= 0xff; /* clear flags */
if (level < server.verbosity) return;
fp = log_to_stdout ? stdout : fopen(server.logfile,"a");
if (!fp) return;
if (rawmode) {
fprintf(fp,"%s",msg);
} else {
int off;
struct timeval tv;
int role_char;
pid_t pid = getpid();
gettimeofday(&tv,NULL);
struct tm tm;
nolocks_localtime(&tm,tv.tv_sec,server.timezone,server.daylight_active);
off = strftime(buf,sizeof(buf),"%d %b %Y %H:%M:%S.",&tm);
snprintf(buf+off,sizeof(buf)-off,"%03d",(int)tv.tv_usec/1000);
if (server.sentinel_mode) {
role_char = 'X'; /* Sentinel. */
} else if (pid != server.pid) {
role_char = 'C'; /* RDB / AOF writing child. */
} else {
role_char = (server.masterhost ? 'S':'M'); /* Slave or Master. */
}
fprintf(fp,"%d:%c %s %c %s\n",
(int)getpid(),role_char, buf,c[level],msg);
}
fflush(fp);
if (!log_to_stdout) fclose(fp);
if (server.syslog_enabled) syslog(syslogLevelMap[level], "%s", msg);
}
/* Like serverLogRaw() but with printf-alike support. This is the function that
* is used across the code. The raw version is only used in order to dump
* the INFO output on crash. */
void serverLog(int level, const char *fmt, ...) {
va_list ap;
char msg[LOG_MAX_LEN];
if ((level&0xff) < server.verbosity) return;
va_start(ap, fmt);
vsnprintf(msg, sizeof(msg), fmt, ap);
va_end(ap);
serverLogRaw(level,msg);
}
/* Log a fixed message without printf-alike capabilities, in a way that is
* safe to call from a signal handler.
*
* We actually use this only for signals that are not fatal from the point
* of view of Redis. Signals that are going to kill the server anyway and
* where we need printf-alike features are served by serverLog(). */
void serverLogFromHandler(int level, const char *msg) {
int fd;
int log_to_stdout = server.logfile[0] == '\0';
char buf[64];
if ((level&0xff) < server.verbosity || (log_to_stdout && server.daemonize))
return;
fd = log_to_stdout ? STDOUT_FILENO :
open(server.logfile, O_APPEND|O_CREAT|O_WRONLY, 0644);
if (fd == -1) return;
ll2string(buf,sizeof(buf),getpid());
if (write(fd,buf,strlen(buf)) == -1) goto err;
if (write(fd,":signal-handler (",17) == -1) goto err;
ll2string(buf,sizeof(buf),time(NULL));
if (write(fd,buf,strlen(buf)) == -1) goto err;
if (write(fd,") ",2) == -1) goto err;
if (write(fd,msg,strlen(msg)) == -1) goto err;
if (write(fd,"\n",1) == -1) goto err;
err:
if (!log_to_stdout) close(fd);
}
/* Return the UNIX time in microseconds */
long long ustime(void) {
struct timeval tv;
long long ust;
gettimeofday(&tv, NULL);
ust = ((long long)tv.tv_sec)*1000000;
ust += tv.tv_usec;
return ust;
}
/* Return the UNIX time in milliseconds */
mstime_t mstime(void) {
return ustime()/1000;
}
/* After an RDB dump or AOF rewrite we exit from children using _exit() instead of
* exit(), because the latter may interact with the same file objects used by
* the parent process. However if we are testing the coverage normal exit() is
* used in order to obtain the right coverage information. */
void exitFromChild(int retcode) {
#ifdef COVERAGE_TEST
exit(retcode);
#else
_exit(retcode);
#endif
}
/*====================== Hash table type implementation ==================== */
/* This is a hash table type that uses the SDS dynamic strings library as
* keys and redis objects as values (objects can hold SDS strings,
* lists, sets). */
void dictVanillaFree(void *privdata, void *val)
{
DICT_NOTUSED(privdata);
zfree(val);
}
void dictListDestructor(void *privdata, void *val)
{
DICT_NOTUSED(privdata);
listRelease((list*)val);
}
int dictSdsKeyCompare(void *privdata, const void *key1,
const void *key2)
{
int l1,l2;
DICT_NOTUSED(privdata);
l1 = sdslen((sds)key1);
l2 = sdslen((sds)key2);
if (l1 != l2) return 0;
return memcmp(key1, key2, l1) == 0;
}
/* A case insensitive version used for the command lookup table and other
* places where case insensitive non binary-safe comparison is needed. */
int dictSdsKeyCaseCompare(void *privdata, const void *key1,
const void *key2)
{
DICT_NOTUSED(privdata);
return strcasecmp(key1, key2) == 0;
}
void dictObjectDestructor(void *privdata, void *val)
{
DICT_NOTUSED(privdata);
if (val == NULL) return; /* Lazy freeing will set value to NULL. */
decrRefCount(val);
}
void dictSdsDestructor(void *privdata, void *val)
{
DICT_NOTUSED(privdata);
sdsfree(val);
}
int dictObjKeyCompare(void *privdata, const void *key1,
const void *key2)
{
const robj *o1 = key1, *o2 = key2;
return dictSdsKeyCompare(privdata,o1->ptr,o2->ptr);
}
uint64_t dictObjHash(const void *key) {
const robj *o = key;
return dictGenHashFunction(o->ptr, sdslen((sds)o->ptr));
}
uint64_t dictSdsHash(const void *key) {
return dictGenHashFunction((unsigned char*)key, sdslen((char*)key));
}
uint64_t dictSdsCaseHash(const void *key) {
return dictGenCaseHashFunction((unsigned char*)key, sdslen((char*)key));
}
int dictEncObjKeyCompare(void *privdata, const void *key1,
const void *key2)
{
robj *o1 = (robj*) key1, *o2 = (robj*) key2;
int cmp;
if (o1->encoding == OBJ_ENCODING_INT &&
o2->encoding == OBJ_ENCODING_INT)
return o1->ptr == o2->ptr;
o1 = getDecodedObject(o1);
o2 = getDecodedObject(o2);
cmp = dictSdsKeyCompare(privdata,o1->ptr,o2->ptr);
decrRefCount(o1);
decrRefCount(o2);
return cmp;
}
uint64_t dictEncObjHash(const void *key) {
robj *o = (robj*) key;
if (sdsEncodedObject(o)) {
return dictGenHashFunction(o->ptr, sdslen((sds)o->ptr));
} else {
if (o->encoding == OBJ_ENCODING_INT) {
char buf[32];
int len;
len = ll2string(buf,32,(long)o->ptr);
return dictGenHashFunction((unsigned char*)buf, len);
} else {
uint64_t hash;
o = getDecodedObject(o);
hash = dictGenHashFunction(o->ptr, sdslen((sds)o->ptr));
decrRefCount(o);
return hash;
}
}
}
/* Generic hash table type where keys are Redis Objects, Values
* dummy pointers. */
dictType objectKeyPointerValueDictType = {
dictEncObjHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictEncObjKeyCompare, /* key compare */
dictObjectDestructor, /* key destructor */
NULL /* val destructor */
};
/* Like objectKeyPointerValueDictType(), but values can be destroyed, if
* not NULL, calling zfree(). */
dictType objectKeyHeapPointerValueDictType = {
dictEncObjHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictEncObjKeyCompare, /* key compare */
dictObjectDestructor, /* key destructor */
dictVanillaFree /* val destructor */
};
/* Set dictionary type. Keys are SDS strings, values are ot used. */
dictType setDictType = {
dictSdsHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCompare, /* key compare */
dictSdsDestructor, /* key destructor */
NULL /* val destructor */
};
/* Sorted sets hash (note: a skiplist is used in addition to the hash table) */
dictType zsetDictType = {
dictSdsHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCompare, /* key compare */
NULL, /* Note: SDS string shared & freed by skiplist */
NULL /* val destructor */
};
/* Db->dict, keys are sds strings, vals are Redis objects. */
dictType dbDictType = {
dictSdsHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCompare, /* key compare */
dictSdsDestructor, /* key destructor */
dictObjectDestructor /* val destructor */
};
/* server.lua_scripts sha (as sds string) -> scripts (as robj) cache. */
dictType shaScriptObjectDictType = {
dictSdsCaseHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCaseCompare, /* key compare */
dictSdsDestructor, /* key destructor */
dictObjectDestructor /* val destructor */
};
/* Db->expires */
dictType keyptrDictType = {
dictSdsHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCompare, /* key compare */
NULL, /* key destructor */
NULL /* val destructor */
};
/* Command table. sds string -> command struct pointer. */
dictType commandTableDictType = {
dictSdsCaseHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCaseCompare, /* key compare */
dictSdsDestructor, /* key destructor */
NULL /* val destructor */
};
/* Hash type hash table (note that small hashes are represented with ziplists) */
dictType hashDictType = {
dictSdsHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCompare, /* key compare */
dictSdsDestructor, /* key destructor */
dictSdsDestructor /* val destructor */
};
/* Keylist hash table type has unencoded redis objects as keys and
* lists as values. It's used for blocking operations (BLPOP) and to
* map swapped keys to a list of clients waiting for this keys to be loaded. */
dictType keylistDictType = {
dictObjHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictObjKeyCompare, /* key compare */
dictObjectDestructor, /* key destructor */
dictListDestructor /* val destructor */
};
/* Cluster nodes hash table, mapping nodes addresses 1.2.3.4:6379 to
* clusterNode structures. */
dictType clusterNodesDictType = {
dictSdsHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCompare, /* key compare */
dictSdsDestructor, /* key destructor */
NULL /* val destructor */
};
/* Cluster re-addition blacklist. This maps node IDs to the time
* we can re-add this node. The goal is to avoid readding a removed
* node for some time. */
dictType clusterNodesBlackListDictType = {
dictSdsCaseHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCaseCompare, /* key compare */
dictSdsDestructor, /* key destructor */
NULL /* val destructor */
};
/* Cluster re-addition blacklist. This maps node IDs to the time
* we can re-add this node. The goal is to avoid readding a removed
* node for some time. */
dictType modulesDictType = {
dictSdsCaseHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCaseCompare, /* key compare */
dictSdsDestructor, /* key destructor */
NULL /* val destructor */
};
/* Migrate cache dict type. */
dictType migrateCacheDictType = {
dictSdsHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCompare, /* key compare */
dictSdsDestructor, /* key destructor */
NULL /* val destructor */
};
/* Replication cached script dict (server.repl_scriptcache_dict).
* Keys are sds SHA1 strings, while values are not used at all in the current
* implementation. */
dictType replScriptCacheDictType = {
dictSdsCaseHash, /* hash function */
NULL, /* key dup */
NULL, /* val dup */
dictSdsKeyCaseCompare, /* key compare */
dictSdsDestructor, /* key destructor */
NULL /* val destructor */
};
int htNeedsResize(dict *dict) {
long long size, used;
size = dictSlots(dict);
used = dictSize(dict);
return (size > DICT_HT_INITIAL_SIZE &&
(used*100/size < HASHTABLE_MIN_FILL));
}
/* If the percentage of used slots in the HT reaches HASHTABLE_MIN_FILL
* we resize the hash table to save memory */
void tryResizeHashTables(int dbid) {
if (htNeedsResize(server.db[dbid].dict))
dictResize(server.db[dbid].dict);
if (htNeedsResize(server.db[dbid].expires))
dictResize(server.db[dbid].expires);
}
/* Our hash table implementation performs rehashing incrementally while
* we write/read from the hash table. Still if the server is idle, the hash
* table will use two tables for a long time. So we try to use 1 millisecond
* of CPU time at every call of this function to perform some rehahsing.
*
* The function returns 1 if some rehashing was performed, otherwise 0
* is returned. */
int incrementallyRehash(int dbid) {
/* Keys dictionary */
if (dictIsRehashing(server.db[dbid].dict)) {
dictRehashMilliseconds(server.db[dbid].dict,1);
return 1; /* already used our millisecond for this loop... */
}
/* Expires */
if (dictIsRehashing(server.db[dbid].expires)) {
dictRehashMilliseconds(server.db[dbid].expires,1);
return 1; /* already used our millisecond for this loop... */
}
return 0;
}
/* This function is called once a background process of some kind terminates,
* as we want to avoid resizing the hash tables when there is a child in order
* to play well with copy-on-write (otherwise when a resize happens lots of
* memory pages are copied). The goal of this function is to update the ability
* for dict.c to resize the hash tables accordingly to the fact we have o not
* running childs. */
void updateDictResizePolicy(void) {
if (!hasActiveChildProcess())
dictEnableResize();
else
dictDisableResize();
}
/* Return true if there are no active children processes doing RDB saving,
* AOF rewriting, or some side process spawned by a loaded module. */
int hasActiveChildProcess() {
return server.rdb_child_pid != -1 ||
server.aof_child_pid != -1 ||
server.module_child_pid != -1;
}
/* Return true if this instance has persistence completely turned off:
* both RDB and AOF are disabled. */
int allPersistenceDisabled(void) {
return server.saveparamslen == 0 && server.aof_state == AOF_OFF;
}
/* ======================= Cron: called every 100 ms ======================== */
/* Add a sample to the operations per second array of samples. */
void trackInstantaneousMetric(int metric, long long current_reading) {
long long t = mstime() - server.inst_metric[metric].last_sample_time;
long long ops = current_reading -
server.inst_metric[metric].last_sample_count;
long long ops_sec;
ops_sec = t > 0 ? (ops*1000/t) : 0;
server.inst_metric[metric].samples[server.inst_metric[metric].idx] =
ops_sec;
server.inst_metric[metric].idx++;
server.inst_metric[metric].idx %= STATS_METRIC_SAMPLES;
server.inst_metric[metric].last_sample_time = mstime();
server.inst_metric[metric].last_sample_count = current_reading;
}
/* Return the mean of all the samples. */
long long getInstantaneousMetric(int metric) {
int j;
long long sum = 0;
for (j = 0; j < STATS_METRIC_SAMPLES; j++)
sum += server.inst_metric[metric].samples[j];
return sum / STATS_METRIC_SAMPLES;
}
/* The client query buffer is an sds.c string that can end with a lot of
* free space not used, this function reclaims space if needed.
*
* The function always returns 0 as it never terminates the client. */
int clientsCronResizeQueryBuffer(client *c) {
size_t querybuf_size = sdsAllocSize(c->querybuf);
time_t idletime = server.unixtime - c->lastinteraction;
/* There are two conditions to resize the query buffer:
* 1) Query buffer is > BIG_ARG and too big for latest peak.
* 2) Query buffer is > BIG_ARG and client is idle. */
if (querybuf_size > PROTO_MBULK_BIG_ARG &&
((querybuf_size/(c->querybuf_peak+1)) > 2 ||
idletime > 2))
{
/* Only resize the query buffer if it is actually wasting
* at least a few kbytes. */
if (sdsavail(c->querybuf) > 1024*4) {
c->querybuf = sdsRemoveFreeSpace(c->querybuf);
}
}
/* Reset the peak again to capture the peak memory usage in the next
* cycle. */
c->querybuf_peak = 0;
/* Clients representing masters also use a "pending query buffer" that
* is the yet not applied part of the stream we are reading. Such buffer
* also needs resizing from time to time, otherwise after a very large
* transfer (a huge value or a big MIGRATE operation) it will keep using
* a lot of memory. */
if (c->flags & CLIENT_MASTER) {
/* There are two conditions to resize the pending query buffer:
* 1) Pending Query buffer is > LIMIT_PENDING_QUERYBUF.
* 2) Used length is smaller than pending_querybuf_size/2 */
size_t pending_querybuf_size = sdsAllocSize(c->pending_querybuf);
if(pending_querybuf_size > LIMIT_PENDING_QUERYBUF &&
sdslen(c->pending_querybuf) < (pending_querybuf_size/2))
{
c->pending_querybuf = sdsRemoveFreeSpace(c->pending_querybuf);
}
}
return 0;
}
/* This function is used in order to track clients using the biggest amount
* of memory in the latest few seconds. This way we can provide such information
* in the INFO output (clients section), without having to do an O(N) scan for
* all the clients.
*
* This is how it works. We have an array of CLIENTS_PEAK_MEM_USAGE_SLOTS slots
* where we track, for each, the biggest client output and input buffers we
* saw in that slot. Every slot correspond to one of the latest seconds, since
* the array is indexed by doing UNIXTIME % CLIENTS_PEAK_MEM_USAGE_SLOTS.
*
* When we want to know what was recently the peak memory usage, we just scan
* such few slots searching for the maximum value. */
#define CLIENTS_PEAK_MEM_USAGE_SLOTS 8
size_t ClientsPeakMemInput[CLIENTS_PEAK_MEM_USAGE_SLOTS];
size_t ClientsPeakMemOutput[CLIENTS_PEAK_MEM_USAGE_SLOTS];
int clientsCronTrackExpansiveClients(client *c) {
size_t in_usage = sdsAllocSize(c->querybuf);
size_t out_usage = getClientOutputBufferMemoryUsage(c);
int i = server.unixtime % CLIENTS_PEAK_MEM_USAGE_SLOTS;
int zeroidx = (i+1) % CLIENTS_PEAK_MEM_USAGE_SLOTS;
/* Always zero the next sample, so that when we switch to that second, we'll
* only register samples that are greater in that second without considering
* the history of such slot.
*
* Note: our index may jump to any random position if serverCron() is not
* called for some reason with the normal frequency, for instance because
* some slow command is called taking multiple seconds to execute. In that
* case our array may end containing data which is potentially older
* than CLIENTS_PEAK_MEM_USAGE_SLOTS seconds: however this is not a problem
* since here we want just to track if "recently" there were very expansive
* clients from the POV of memory usage. */
ClientsPeakMemInput[zeroidx] = 0;
ClientsPeakMemOutput[zeroidx] = 0;
/* Track the biggest values observed so far in this slot. */
if (in_usage > ClientsPeakMemInput[i]) ClientsPeakMemInput[i] = in_usage;
if (out_usage > ClientsPeakMemOutput[i]) ClientsPeakMemOutput[i] = out_usage;
return 0; /* This function never terminates the client. */
}
/* Return the max samples in the memory usage of clients tracked by
* the function clientsCronTrackExpansiveClients(). */
void getExpansiveClientsInfo(size_t *in_usage, size_t *out_usage) {
size_t i = 0, o = 0;
for (int j = 0; j < CLIENTS_PEAK_MEM_USAGE_SLOTS; j++) {
if (ClientsPeakMemInput[j] > i) i = ClientsPeakMemInput[j];
if (ClientsPeakMemOutput[j] > o) o = ClientsPeakMemOutput[j];
}
*in_usage = i;
*out_usage = o;
}
/* This function is called by serverCron() and is used in order to perform
* operations on clients that are important to perform constantly. For instance
* we use this function in order to disconnect clients after a timeout, including
* clients blocked in some blocking command with a non-zero timeout.
*
* The function makes some effort to process all the clients every second, even
* if this cannot be strictly guaranteed, since serverCron() may be called with
* an actual frequency lower than server.hz in case of latency events like slow
* commands.
*
* It is very important for this function, and the functions it calls, to be
* very fast: sometimes Redis has tens of hundreds of connected clients, and the
* default server.hz value is 10, so sometimes here we need to process thousands
* of clients per second, turning this function into a source of latency.
*/
#define CLIENTS_CRON_MIN_ITERATIONS 5
void clientsCron(void) {
/* Try to process at least numclients/server.hz of clients
* per call. Since normally (if there are no big latency events) this
* function is called server.hz times per second, in the average case we
* process all the clients in 1 second. */
int numclients = listLength(server.clients);
int iterations = numclients/server.hz;
mstime_t now = mstime();
/* Process at least a few clients while we are at it, even if we need
* to process less than CLIENTS_CRON_MIN_ITERATIONS to meet our contract
* of processing each client once per second. */
if (iterations < CLIENTS_CRON_MIN_ITERATIONS)
iterations = (numclients < CLIENTS_CRON_MIN_ITERATIONS) ?
numclients : CLIENTS_CRON_MIN_ITERATIONS;
while(listLength(server.clients) && iterations--) {
client *c;
listNode *head;
/* Rotate the list, take the current head, process.
* This way if the client must be removed from the list it's the
* first element and we don't incur into O(N) computation. */
listRotate(server.clients);
head = listFirst(server.clients);
c = listNodeValue(head);
/* The following functions do different service checks on the client.
* The protocol is that they return non-zero if the client was
* terminated. */
if (clientsCronHandleTimeout(c,now)) continue;
if (clientsCronResizeQueryBuffer(c)) continue;
if (clientsCronTrackExpansiveClients(c)) continue;
}
}
/* This function handles 'background' operations we are required to do
* incrementally in Redis databases, such as active key expiring, resizing,
* rehashing. */
void databasesCron(void) {
/* Expire keys by random sampling. Not required for slaves
* as master will synthesize DELs for us. */
if (server.active_expire_enabled) {
if (iAmMaster()) {
activeExpireCycle(ACTIVE_EXPIRE_CYCLE_SLOW);
} else {
expireSlaveKeys();
}
}
/* Defrag keys gradually. */
activeDefragCycle();
/* Perform hash tables rehashing if needed, but only if there are no
* other processes saving the DB on disk. Otherwise rehashing is bad
* as will cause a lot of copy-on-write of memory pages. */
if (!hasActiveChildProcess()) {
/* We use global counters so if we stop the computation at a given
* DB we'll be able to start from the successive in the next
* cron loop iteration. */
static unsigned int resize_db = 0;
static unsigned int rehash_db = 0;
int dbs_per_call = CRON_DBS_PER_CALL;
int j;
/* Don't test more DBs than we have. */
if (dbs_per_call > server.dbnum) dbs_per_call = server.dbnum;
/* Resize */
for (j = 0; j < dbs_per_call; j++) {
tryResizeHashTables(resize_db % server.dbnum);
resize_db++;
}
/* Rehash */
if (server.activerehashing) {
for (j = 0; j < dbs_per_call; j++) {
int work_done = incrementallyRehash(rehash_db);
if (work_done) {
/* If the function did some work, stop here, we'll do
* more at the next cron loop. */
break;
} else {
/* If this db didn't need rehash, we'll try the next one. */
rehash_db++;
rehash_db %= server.dbnum;
}
}
}
}
}
/* We take a cached value of the unix time in the global state because with
* virtual memory and aging there is to store the current time in objects at
* every object access, and accuracy is not needed. To access a global var is
* a lot faster than calling time(NULL).
*
* This function should be fast because it is called at every command execution
* in call(), so it is possible to decide if to update the daylight saving
* info or not using the 'update_daylight_info' argument. Normally we update
* such info only when calling this function from serverCron() but not when
* calling it from call(). */
void updateCachedTime(int update_daylight_info) {
server.ustime = ustime();
server.mstime = server.ustime / 1000;
server.unixtime = server.mstime / 1000;
/* To get information about daylight saving time, we need to call
* localtime_r and cache the result. However calling localtime_r in this
* context is safe since we will never fork() while here, in the main
* thread. The logging function will call a thread safe version of
* localtime that has no locks. */
if (update_daylight_info) {
struct tm tm;
time_t ut = server.unixtime;
localtime_r(&ut,&tm);
server.daylight_active = tm.tm_isdst;
}
}
void checkChildrenDone(void) {
int statloc;
pid_t pid;
/* If we have a diskless rdb child (note that we support only one concurrent
* child), we want to avoid collecting it's exit status and acting on it
* as long as we didn't finish to drain the pipe, since then we're at risk
* of starting a new fork and a new pipe before we're done with the previous
* one. */
if (server.rdb_child_pid != -1 && server.rdb_pipe_conns)
return;
if ((pid = wait3(&statloc,WNOHANG,NULL)) != 0) {
int exitcode = WEXITSTATUS(statloc);
int bysignal = 0;
if (WIFSIGNALED(statloc)) bysignal = WTERMSIG(statloc);
/* sigKillChildHandler catches the signal and calls exit(), but we
* must make sure not to flag lastbgsave_status, etc incorrectly.
* We could directly terminate the child process via SIGUSR1
* without handling it, but in this case Valgrind will log an
* annoying error. */
if (exitcode == SERVER_CHILD_NOERROR_RETVAL) {
bysignal = SIGUSR1;
exitcode = 1;
}
if (pid == -1) {
serverLog(LL_WARNING,"wait3() returned an error: %s. "
"rdb_child_pid = %d, aof_child_pid = %d, module_child_pid = %d",
strerror(errno),
(int) server.rdb_child_pid,
(int) server.aof_child_pid,
(int) server.module_child_pid);
} else if (pid == server.rdb_child_pid) {
backgroundSaveDoneHandler(exitcode,bysignal);
if (!bysignal && exitcode == 0) receiveChildInfo();
} else if (pid == server.aof_child_pid) {
backgroundRewriteDoneHandler(exitcode,bysignal);
if (!bysignal && exitcode == 0) receiveChildInfo();
} else if (pid == server.module_child_pid) {
ModuleForkDoneHandler(exitcode,bysignal);
if (!bysignal && exitcode == 0) receiveChildInfo();
} else {
if (!ldbRemoveChild(pid)) {
serverLog(LL_WARNING,
"Warning, detected child with unmatched pid: %ld",
(long)pid);
}
}
updateDictResizePolicy();
closeChildInfoPipe();
}
}
/* This is our timer interrupt, called server.hz times per second.
* Here is where we do a number of things that need to be done asynchronously.
* For instance:
*
* - Active expired keys collection (it is also performed in a lazy way on
* lookup).
* - Software watchdog.
* - Update some statistic.
* - Incremental rehashing of the DBs hash tables.
* - Triggering BGSAVE / AOF rewrite, and handling of terminated children.
* - Clients timeout of different kinds.
* - Replication reconnection.
* - Many more...
*
* Everything directly called here will be called server.hz times per second,
* so in order to throttle execution of things we want to do less frequently
* a macro is used: run_with_period(milliseconds) { .... }
*/
int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) {
int j;
UNUSED(eventLoop);
UNUSED(id);
UNUSED(clientData);
/* Software watchdog: deliver the SIGALRM that will reach the signal
* handler if we don't return here fast enough. */
if (server.watchdog_period) watchdogScheduleSignal(server.watchdog_period);
/* Update the time cache. */
updateCachedTime(1);
server.hz = server.config_hz;
/* Adapt the server.hz value to the number of configured clients. If we have
* many clients, we want to call serverCron() with an higher frequency. */
if (server.dynamic_hz) {
while (listLength(server.clients) / server.hz >
MAX_CLIENTS_PER_CLOCK_TICK)
{
server.hz *= 2;
if (server.hz > CONFIG_MAX_HZ) {
server.hz = CONFIG_MAX_HZ;
break;
}
}
}
run_with_period(100) {
trackInstantaneousMetric(STATS_METRIC_COMMAND,server.stat_numcommands);
trackInstantaneousMetric(STATS_METRIC_NET_INPUT,
server.stat_net_input_bytes);
trackInstantaneousMetric(STATS_METRIC_NET_OUTPUT,
server.stat_net_output_bytes);
}
/* We have just LRU_BITS bits per object for LRU information.
* So we use an (eventually wrapping) LRU clock.
*
* Note that even if the counter wraps it's not a big problem,
* everything will still work but some object will appear younger
* to Redis. However for this to happen a given object should never be
* touched for all the time needed to the counter to wrap, which is
* not likely.
*
* Note that you can change the resolution altering the
* LRU_CLOCK_RESOLUTION define. */
server.lruclock = getLRUClock();
/* Record the max memory used since the server was started. */
if (zmalloc_used_memory() > server.stat_peak_memory)
server.stat_peak_memory = zmalloc_used_memory();
run_with_period(100) {
/* Sample the RSS and other metrics here since this is a relatively slow call.
* We must sample the zmalloc_used at the same time we take the rss, otherwise
* the frag ratio calculate may be off (ratio of two samples at different times) */
server.cron_malloc_stats.process_rss = zmalloc_get_rss();
server.cron_malloc_stats.zmalloc_used = zmalloc_used_memory();
/* Sampling the allcator info can be slow too.
* The fragmentation ratio it'll show is potentically more accurate
* it excludes other RSS pages such as: shared libraries, LUA and other non-zmalloc
* allocations, and allocator reserved pages that can be pursed (all not actual frag) */
zmalloc_get_allocator_info(&server.cron_malloc_stats.allocator_allocated,
&server.cron_malloc_stats.allocator_active,
&server.cron_malloc_stats.allocator_resident);
/* in case the allocator isn't providing these stats, fake them so that
* fragmention info still shows some (inaccurate metrics) */
if (!server.cron_malloc_stats.allocator_resident) {
/* LUA memory isn't part of zmalloc_used, but it is part of the process RSS,
* so we must desuct it in order to be able to calculate correct
* "allocator fragmentation" ratio */
size_t lua_memory = lua_gc(server.lua,LUA_GCCOUNT,0)*1024LL;
server.cron_malloc_stats.allocator_resident = server.cron_malloc_stats.process_rss - lua_memory;
}
if (!server.cron_malloc_stats.allocator_active)
server.cron_malloc_stats.allocator_active = server.cron_malloc_stats.allocator_resident;
if (!server.cron_malloc_stats.allocator_allocated)
server.cron_malloc_stats.allocator_allocated = server.cron_malloc_stats.zmalloc_used;
}
/* We received a SIGTERM, shutting down here in a safe way, as it is
* not ok doing so inside the signal handler. */
if (server.shutdown_asap) {
if (prepareForShutdown(SHUTDOWN_NOFLAGS) == C_OK) exit(0);
serverLog(LL_WARNING,"SIGTERM received but errors trying to shut down the server, check the logs for more information");
server.shutdown_asap = 0;
}
/* Show some info about non-empty databases */
run_with_period(5000) {
for (j = 0; j < server.dbnum; j++) {
long long size, used, vkeys;
size = dictSlots(server.db[j].dict);
used = dictSize(server.db[j].dict);
vkeys = dictSize(server.db[j].expires);
if (used || vkeys) {
serverLog(LL_VERBOSE,"DB %d: %lld keys (%lld volatile) in %lld slots HT.",j,used,vkeys,size);
/* dictPrintStats(server.dict); */
}
}
}
/* Show information about connected clients */
if (!server.sentinel_mode) {
run_with_period(5000) {
serverLog(LL_DEBUG,
"%lu clients connected (%lu replicas), %zu bytes in use",
listLength(server.clients)-listLength(server.slaves),
listLength(server.slaves),
zmalloc_used_memory());
}
}
/* We need to do a few operations on clients asynchronously. */
clientsCron();
/* Handle background operations on Redis databases. */
databasesCron();
/* Start a scheduled AOF rewrite if this was requested by the user while
* a BGSAVE was in progress. */
if (!hasActiveChildProcess() &&
server.aof_rewrite_scheduled)
{
rewriteAppendOnlyFileBackground();
}
/* Check if a background saving or AOF rewrite in progress terminated. */
if (hasActiveChildProcess() || ldbPendingChildren())
{
checkChildrenDone();
} else {
/* If there is not a background saving/rewrite in progress check if
* we have to save/rewrite now. */
for (j = 0; j < server.saveparamslen; j++) {
struct saveparam *sp = server.saveparams+j;
/* Save if we reached the given amount of changes,
* the given amount of seconds, and if the latest bgsave was
* successful or if, in case of an error, at least
* CONFIG_BGSAVE_RETRY_DELAY seconds already elapsed. */
if (server.dirty >= sp->changes &&
server.unixtime-server.lastsave > sp->seconds &&
(server.unixtime-server.lastbgsave_try >
CONFIG_BGSAVE_RETRY_DELAY ||
server.lastbgsave_status == C_OK))
{
serverLog(LL_NOTICE,"%d changes in %d seconds. Saving...",
sp->changes, (int)sp->seconds);
rdbSaveInfo rsi, *rsiptr;
rsiptr = rdbPopulateSaveInfo(&rsi);
rdbSaveBackground(server.rdb_filename,rsiptr);
break;
}
}
/* Trigger an AOF rewrite if needed. */
if (server.aof_state == AOF_ON &&
!hasActiveChildProcess() &&
server.aof_rewrite_perc &&
server.aof_current_size > server.aof_rewrite_min_size)
{
long long base = server.aof_rewrite_base_size ?
server.aof_rewrite_base_size : 1;
long long growth = (server.aof_current_size*100/base) - 100;
if (growth >= server.aof_rewrite_perc) {
serverLog(LL_NOTICE,"Starting automatic rewriting of AOF on %lld%% growth",growth);
rewriteAppendOnlyFileBackground();
}
}
}
/* AOF postponed flush: Try at every cron cycle if the slow fsync
* completed. */
if (server.aof_flush_postponed_start) flushAppendOnlyFile(0);
/* AOF write errors: in this case we have a buffer to flush as well and
* clear the AOF error in case of success to make the DB writable again,
* however to try every second is enough in case of 'hz' is set to
* an higher frequency. */
run_with_period(1000) {
if (server.aof_last_write_status == C_ERR)
flushAppendOnlyFile(0);
}
/* Clear the paused clients flag if needed. */
clientsArePaused(); /* Don't check return value, just use the side effect.*/
/* Replication cron function -- used to reconnect to master,
* detect transfer failures, start background RDB transfers and so forth. */
run_with_period(1000) replicationCron();
/* Run the Redis Cluster cron. */
run_with_period(100) {
if (server.cluster_enabled) clusterCron();
}
/* Run the Sentinel timer if we are in sentinel mode. */
if (server.sentinel_mode) sentinelTimer();
/* Cleanup expired MIGRATE cached sockets. */
run_with_period(1000) {
migrateCloseTimedoutSockets();
}
/* Stop the I/O threads if we don't have enough pending work. */
stopThreadedIOIfNeeded();
/* Start a scheduled BGSAVE if the corresponding flag is set. This is
* useful when we are forced to postpone a BGSAVE because an AOF
* rewrite is in progress.
*
* Note: this code must be after the replicationCron() call above so
* make sure when refactoring this file to keep this order. This is useful
* because we want to give priority to RDB savings for replication. */
if (!hasActiveChildProcess() &&
server.rdb_bgsave_scheduled &&
(server.unixtime-server.lastbgsave_try > CONFIG_BGSAVE_RETRY_DELAY ||
server.lastbgsave_status == C_OK))
{
rdbSaveInfo rsi, *rsiptr;
rsiptr = rdbPopulateSaveInfo(&rsi);
if (rdbSaveBackground(server.rdb_filename,rsiptr) == C_OK)
server.rdb_bgsave_scheduled = 0;
}
/* Fire the cron loop modules event. */
RedisModuleCronLoopV1 ei = {REDISMODULE_CRON_LOOP_VERSION,server.hz};
moduleFireServerEvent(REDISMODULE_EVENT_CRON_LOOP,
0,
&ei);
server.cronloops++;
return 1000/server.hz;
}
/* This function gets called every time Redis is entering the
* main loop of the event driven library, that is, before to sleep
* for ready file descriptors. */
void beforeSleep(struct aeEventLoop *eventLoop) {
UNUSED(eventLoop);
/* Handle precise timeouts of blocked clients. */
handleBlockedClientsTimeout();
/* We should handle pending reads clients ASAP after event loop. */
handleClientsWithPendingReadsUsingThreads();
/* Handle TLS pending data. (must be done before flushAppendOnlyFile) */
tlsProcessPendingData();
/* If tls still has pending unread data don't sleep at all. */
aeSetDontWait(server.el, tlsHasPendingData());
/* Call the Redis Cluster before sleep function. Note that this function
* may change the state of Redis Cluster (from ok to fail or vice versa),
* so it's a good idea to call it before serving the unblocked clients
* later in this function. */
if (server.cluster_enabled) clusterBeforeSleep();
/* Run a fast expire cycle (the called function will return
* ASAP if a fast cycle is not needed). */
if (server.active_expire_enabled && server.masterhost == NULL)
activeExpireCycle(ACTIVE_EXPIRE_CYCLE_FAST);
/* Send all the slaves an ACK request if at least one client blocked
* during the previous event loop iteration. */
if (server.get_ack_from_slaves) {
robj *argv[3];
argv[0] = createStringObject("REPLCONF",8);
argv[1] = createStringObject("GETACK",6);
argv[2] = createStringObject("*",1); /* Not used argument. */
replicationFeedSlaves(server.slaves, server.slaveseldb, argv, 3);
decrRefCount(argv[0]);
decrRefCount(argv[1]);
decrRefCount(argv[2]);
server.get_ack_from_slaves = 0;
}
/* Unblock all the clients blocked for synchronous replication
* in WAIT. */
if (listLength(server.clients_waiting_acks))
processClientsWaitingReplicas();
/* Check if there are clients unblocked by modules that implement
* blocking commands. */
if (moduleCount()) moduleHandleBlockedClients();
/* Try to process pending commands for clients that were just unblocked. */
if (listLength(server.unblocked_clients))
processUnblockedClients();
/* Send the invalidation messages to clients participating to the
* client side caching protocol in broadcasting (BCAST) mode. */
trackingBroadcastInvalidationMessages();
/* Write the AOF buffer on disk */
flushAppendOnlyFile(0);
/* Handle writes with pending output buffers. */
handleClientsWithPendingWritesUsingThreads();
/* Close clients that need to be closed asynchronous */
freeClientsInAsyncFreeQueue();
/* Before we are going to sleep, let the threads access the dataset by
* releasing the GIL. Redis main thread will not touch anything at this
* time. */
if (moduleCount()) moduleReleaseGIL();
}
/* This function is called immadiately after the event loop multiplexing
* API returned, and the control is going to soon return to Redis by invoking
* the different events callbacks. */
void afterSleep(struct aeEventLoop *eventLoop) {
UNUSED(eventLoop);
if (moduleCount()) moduleAcquireGIL();
}
/* =========================== Server initialization ======================== */
void createSharedObjects(void) {
int j;
shared.crlf = createObject(OBJ_STRING,sdsnew("\r\n"));
shared.ok = createObject(OBJ_STRING,sdsnew("+OK\r\n"));
shared.err = createObject(OBJ_STRING,sdsnew("-ERR\r\n"));
shared.emptybulk = createObject(OBJ_STRING,sdsnew("$0\r\n\r\n"));
shared.czero = createObject(OBJ_STRING,sdsnew(":0\r\n"));
shared.cone = createObject(OBJ_STRING,sdsnew(":1\r\n"));
shared.emptyarray = createObject(OBJ_STRING,sdsnew("*0\r\n"));
shared.pong = createObject(OBJ_STRING,sdsnew("+PONG\r\n"));
shared.queued = createObject(OBJ_STRING,sdsnew("+QUEUED\r\n"));
shared.emptyscan = createObject(OBJ_STRING,sdsnew("*2\r\n$1\r\n0\r\n*0\r\n"));
shared.wrongtypeerr = createObject(OBJ_STRING,sdsnew(
"-WRONGTYPE Operation against a key holding the wrong kind of value\r\n"));
shared.nokeyerr = createObject(OBJ_STRING,sdsnew(
"-ERR no such key\r\n"));
shared.syntaxerr = createObject(OBJ_STRING,sdsnew(
"-ERR syntax error\r\n"));
shared.sameobjecterr = createObject(OBJ_STRING,sdsnew(
"-ERR source and destination objects are the same\r\n"));
shared.outofrangeerr = createObject(OBJ_STRING,sdsnew(
"-ERR index out of range\r\n"));
shared.noscripterr = createObject(OBJ_STRING,sdsnew(
"-NOSCRIPT No matching script. Please use EVAL.\r\n"));
shared.loadingerr = createObject(OBJ_STRING,sdsnew(
"-LOADING Redis is loading the dataset in memory\r\n"));
shared.slowscripterr = createObject(OBJ_STRING,sdsnew(
"-BUSY Redis is busy running a script. You can only call SCRIPT KILL or SHUTDOWN NOSAVE.\r\n"));
shared.masterdownerr = createObject(OBJ_STRING,sdsnew(
"-MASTERDOWN Link with MASTER is down and replica-serve-stale-data is set to 'no'.\r\n"));
shared.bgsaveerr = createObject(OBJ_STRING,sdsnew(
"-MISCONF Redis is configured to save RDB snapshots, but it is currently not able to persist on disk. Commands that may modify the data set are disabled, because this instance is configured to report errors during writes if RDB snapshotting fails (stop-writes-on-bgsave-error option). Please check the Redis logs for details about the RDB error.\r\n"));
shared.roslaveerr = createObject(OBJ_STRING,sdsnew(
"-READONLY You can't write against a read only replica.\r\n"));
shared.noautherr = createObject(OBJ_STRING,sdsnew(
"-NOAUTH Authentication required.\r\n"));
shared.oomerr = createObject(OBJ_STRING,sdsnew(
"-OOM command not allowed when used memory > 'maxmemory'.\r\n"));
shared.execaborterr = createObject(OBJ_STRING,sdsnew(
"-EXECABORT Transaction discarded because of previous errors.\r\n"));
shared.noreplicaserr = createObject(OBJ_STRING,sdsnew(
"-NOREPLICAS Not enough good replicas to write.\r\n"));
shared.busykeyerr = createObject(OBJ_STRING,sdsnew(
"-BUSYKEY Target key name already exists.\r\n"));
shared.space = createObject(OBJ_STRING,sdsnew(" "));
shared.colon = createObject(OBJ_STRING,sdsnew(":"));
shared.plus = createObject(OBJ_STRING,sdsnew("+"));
/* The shared NULL depends on the protocol version. */
shared.null[0] = NULL;
shared.null[1] = NULL;
shared.null[2] = createObject(OBJ_STRING,sdsnew("$-1\r\n"));
shared.null[3] = createObject(OBJ_STRING,sdsnew("_\r\n"));
shared.nullarray[0] = NULL;
shared.nullarray[1] = NULL;
shared.nullarray[2] = createObject(OBJ_STRING,sdsnew("*-1\r\n"));
shared.nullarray[3] = createObject(OBJ_STRING,sdsnew("_\r\n"));
shared.emptymap[0] = NULL;
shared.emptymap[1] = NULL;
shared.emptymap[2] = createObject(OBJ_STRING,sdsnew("*0\r\n"));
shared.emptymap[3] = createObject(OBJ_STRING,sdsnew("%0\r\n"));
shared.emptyset[0] = NULL;
shared.emptyset[1] = NULL;
shared.emptyset[2] = createObject(OBJ_STRING,sdsnew("*0\r\n"));
shared.emptyset[3] = createObject(OBJ_STRING,sdsnew("~0\r\n"));
for (j = 0; j < PROTO_SHARED_SELECT_CMDS; j++) {
char dictid_str[64];
int dictid_len;
dictid_len = ll2string(dictid_str,sizeof(dictid_str),j);
shared.select[j] = createObject(OBJ_STRING,
sdscatprintf(sdsempty(),
"*2\r\n$6\r\nSELECT\r\n$%d\r\n%s\r\n",
dictid_len, dictid_str));
}
shared.messagebulk = createStringObject("$7\r\nmessage\r\n",13);
shared.pmessagebulk = createStringObject("$8\r\npmessage\r\n",14);
shared.subscribebulk = createStringObject("$9\r\nsubscribe\r\n",15);
shared.unsubscribebulk = createStringObject("$11\r\nunsubscribe\r\n",18);
shared.psubscribebulk = createStringObject("$10\r\npsubscribe\r\n",17);
shared.punsubscribebulk = createStringObject("$12\r\npunsubscribe\r\n",19);
shared.del = createStringObject("DEL",3);
shared.unlink = createStringObject("UNLINK",6);
shared.rpop = createStringObject("RPOP",4);
shared.lpop = createStringObject("LPOP",4);
shared.lpush = createStringObject("LPUSH",5);
shared.rpoplpush = createStringObject("RPOPLPUSH",9);
shared.zpopmin = createStringObject("ZPOPMIN",7);
shared.zpopmax = createStringObject("ZPOPMAX",7);
shared.multi = createStringObject("MULTI",5);
shared.exec = createStringObject("EXEC",4);
for (j = 0; j < OBJ_SHARED_INTEGERS; j++) {
shared.integers[j] =
makeObjectShared(createObject(OBJ_STRING,(void*)(long)j));
shared.integers[j]->encoding = OBJ_ENCODING_INT;
}
for (j = 0; j < OBJ_SHARED_BULKHDR_LEN; j++) {
shared.mbulkhdr[j] = createObject(OBJ_STRING,
sdscatprintf(sdsempty(),"*%d\r\n",j));
shared.bulkhdr[j] = createObject(OBJ_STRING,
sdscatprintf(sdsempty(),"$%d\r\n",j));
}
/* The following two shared objects, minstring and maxstrings, are not
* actually used for their value but as a special object meaning
* respectively the minimum possible string and the maximum possible
* string in string comparisons for the ZRANGEBYLEX command. */
shared.minstring = sdsnew("minstring");
shared.maxstring = sdsnew("maxstring");
}
void initServerConfig(void) {
int j;
updateCachedTime(1);
getRandomHexChars(server.runid,CONFIG_RUN_ID_SIZE);
server.runid[CONFIG_RUN_ID_SIZE] = '\0';
changeReplicationId();
clearReplicationId2();
server.hz = CONFIG_DEFAULT_HZ; /* Initialize it ASAP, even if it may get
updated later after loading the config.
This value may be used before the server
is initialized. */
server.timezone = getTimeZone(); /* Initialized by tzset(). */
server.configfile = NULL;
server.executable = NULL;
server.arch_bits = (sizeof(long) == 8) ? 64 : 32;
server.bindaddr_count = 0;
server.unixsocketperm = CONFIG_DEFAULT_UNIX_SOCKET_PERM;
server.ipfd_count = 0;
server.tlsfd_count = 0;
server.sofd = -1;
server.active_expire_enabled = 1;
server.client_max_querybuf_len = PROTO_MAX_QUERYBUF_LEN;
server.saveparams = NULL;
server.loading = 0;
server.logfile = zstrdup(CONFIG_DEFAULT_LOGFILE);
server.aof_state = AOF_OFF;
server.aof_rewrite_base_size = 0;
server.aof_rewrite_scheduled = 0;
server.aof_flush_sleep = 0;
server.aof_last_fsync = time(NULL);
server.aof_rewrite_time_last = -1;
server.aof_rewrite_time_start = -1;
server.aof_lastbgrewrite_status = C_OK;
server.aof_delayed_fsync = 0;
server.aof_fd = -1;
server.aof_selected_db = -1; /* Make sure the first time will not match */
server.aof_flush_postponed_start = 0;
server.pidfile = NULL;
server.active_defrag_running = 0;
server.notify_keyspace_events = 0;
server.blocked_clients = 0;
memset(server.blocked_clients_by_type,0,
sizeof(server.blocked_clients_by_type));
server.shutdown_asap = 0;
server.cluster_configfile = zstrdup(CONFIG_DEFAULT_CLUSTER_CONFIG_FILE);
server.cluster_module_flags = CLUSTER_MODULE_FLAG_NONE;
server.migrate_cached_sockets = dictCreate(&migrateCacheDictType,NULL);
server.next_client_id = 1; /* Client IDs, start from 1 .*/
server.loading_process_events_interval_bytes = (1024*1024*2);
server.lruclock = getLRUClock();
resetServerSaveParams();
appendServerSaveParams(60*60,1); /* save after 1 hour and 1 change */
appendServerSaveParams(300,100); /* save after 5 minutes and 100 changes */
appendServerSaveParams(60,10000); /* save after 1 minute and 10000 changes */
/* Replication related */
server.masterauth = NULL;
server.masterhost = NULL;
server.masterport = 6379;
server.master = NULL;
server.cached_master = NULL;
server.master_initial_offset = -1;
server.repl_state = REPL_STATE_NONE;
server.repl_transfer_tmpfile = NULL;
server.repl_transfer_fd = -1;
server.repl_transfer_s = NULL;
server.repl_syncio_timeout = CONFIG_REPL_SYNCIO_TIMEOUT;
server.repl_down_since = 0; /* Never connected, repl is down since EVER. */
server.master_repl_offset = 0;
server.master_repl_meaningful_offset = 0;
/* Replication partial resync backlog */
server.repl_backlog = NULL;
server.repl_backlog_histlen = 0;
server.repl_backlog_idx = 0;
server.repl_backlog_off = 0;
server.repl_no_slaves_since = time(NULL);
/* Client output buffer limits */
for (j = 0; j < CLIENT_TYPE_OBUF_COUNT; j++)
server.client_obuf_limits[j] = clientBufferLimitsDefaults[j];
/* Double constants initialization */
R_Zero = 0.0;
R_PosInf = 1.0/R_Zero;
R_NegInf = -1.0/R_Zero;
R_Nan = R_Zero/R_Zero;
/* Command table -- we initiialize it here as it is part of the
* initial configuration, since command names may be changed via
* redis.conf using the rename-command directive. */
server.commands = dictCreate(&commandTableDictType,NULL);
server.orig_commands = dictCreate(&commandTableDictType,NULL);
populateCommandTable();
server.delCommand = lookupCommandByCString("del");
server.multiCommand = lookupCommandByCString("multi");
server.lpushCommand = lookupCommandByCString("lpush");
server.lpopCommand = lookupCommandByCString("lpop");
server.rpopCommand = lookupCommandByCString("rpop");
server.zpopminCommand = lookupCommandByCString("zpopmin");
server.zpopmaxCommand = lookupCommandByCString("zpopmax");
server.sremCommand = lookupCommandByCString("srem");
server.execCommand = lookupCommandByCString("exec");
server.expireCommand = lookupCommandByCString("expire");
server.pexpireCommand = lookupCommandByCString("pexpire");
server.xclaimCommand = lookupCommandByCString("xclaim");
server.xgroupCommand = lookupCommandByCString("xgroup");
server.rpoplpushCommand = lookupCommandByCString("rpoplpush");
/* Debugging */
server.assert_failed = "<no assertion failed>";
server.assert_file = "<no file>";
server.assert_line = 0;
server.bug_report_start = 0;
server.watchdog_period = 0;
/* By default we want scripts to be always replicated by effects
* (single commands executed by the script), and not by sending the
* script to the slave / AOF. This is the new way starting from
* Redis 5. However it is possible to revert it via redis.conf. */
server.lua_always_replicate_commands = 1;
initConfigValues();
}
extern char **environ;
/* Restart the server, executing the same executable that started this
* instance, with the same arguments and configuration file.
*
* The function is designed to directly call execve() so that the new
* server instance will retain the PID of the previous one.
*
* The list of flags, that may be bitwise ORed together, alter the
* behavior of this function:
*
* RESTART_SERVER_NONE No flags.
* RESTART_SERVER_GRACEFULLY Do a proper shutdown before restarting.
* RESTART_SERVER_CONFIG_REWRITE Rewrite the config file before restarting.
*
* On success the function does not return, because the process turns into
* a different process. On error C_ERR is returned. */
int restartServer(int flags, mstime_t delay) {
int j;
/* Check if we still have accesses to the executable that started this
* server instance. */
if (access(server.executable,X_OK) == -1) {
serverLog(LL_WARNING,"Can't restart: this process has no "
"permissions to execute %s", server.executable);
return C_ERR;
}
/* Config rewriting. */
if (flags & RESTART_SERVER_CONFIG_REWRITE &&
server.configfile &&
rewriteConfig(server.configfile) == -1)
{
serverLog(LL_WARNING,"Can't restart: configuration rewrite process "
"failed");
return C_ERR;
}
/* Perform a proper shutdown. */
if (flags & RESTART_SERVER_GRACEFULLY &&
prepareForShutdown(SHUTDOWN_NOFLAGS) != C_OK)
{
serverLog(LL_WARNING,"Can't restart: error preparing for shutdown");
return C_ERR;
}
/* Close all file descriptors, with the exception of stdin, stdout, strerr
* which are useful if we restart a Redis server which is not daemonized. */
for (j = 3; j < (int)server.maxclients + 1024; j++) {
/* Test the descriptor validity before closing it, otherwise
* Valgrind issues a warning on close(). */
if (fcntl(j,F_GETFD) != -1) close(j);
}
/* Execute the server with the original command line. */
if (delay) usleep(delay*1000);
zfree(server.exec_argv[0]);
server.exec_argv[0] = zstrdup(server.executable);
execve(server.executable,server.exec_argv,environ);
/* If an error occurred here, there is nothing we can do, but exit. */
_exit(1);
return C_ERR; /* Never reached. */
}
/* This function will try to raise the max number of open files accordingly to
* the configured max number of clients. It also reserves a number of file
* descriptors (CONFIG_MIN_RESERVED_FDS) for extra operations of
* persistence, listening sockets, log files and so forth.
*
* If it will not be possible to set the limit accordingly to the configured
* max number of clients, the function will do the reverse setting
* server.maxclients to the value that we can actually handle. */
void adjustOpenFilesLimit(void) {
rlim_t maxfiles = server.maxclients+CONFIG_MIN_RESERVED_FDS;
struct rlimit limit;
if (getrlimit(RLIMIT_NOFILE,&limit) == -1) {
serverLog(LL_WARNING,"Unable to obtain the current NOFILE limit (%s), assuming 1024 and setting the max clients configuration accordingly.",
strerror(errno));
server.maxclients = 1024-CONFIG_MIN_RESERVED_FDS;
} else {
rlim_t oldlimit = limit.rlim_cur;
/* Set the max number of files if the current limit is not enough
* for our needs. */
if (oldlimit < maxfiles) {
rlim_t bestlimit;
int setrlimit_error = 0;
/* Try to set the file limit to match 'maxfiles' or at least
* to the higher value supported less than maxfiles. */
bestlimit = maxfiles;
while(bestlimit > oldlimit) {
rlim_t decr_step = 16;
limit.rlim_cur = bestlimit;
limit.rlim_max = bestlimit;
if (setrlimit(RLIMIT_NOFILE,&limit) != -1) break;
setrlimit_error = errno;
/* We failed to set file limit to 'bestlimit'. Try with a
* smaller limit decrementing by a few FDs per iteration. */
if (bestlimit < decr_step) break;
bestlimit -= decr_step;
}
/* Assume that the limit we get initially is still valid if
* our last try was even lower. */
if (bestlimit < oldlimit) bestlimit = oldlimit;
if (bestlimit < maxfiles) {
unsigned int old_maxclients = server.maxclients;
server.maxclients = bestlimit-CONFIG_MIN_RESERVED_FDS;
/* maxclients is unsigned so may overflow: in order
* to check if maxclients is now logically less than 1
* we test indirectly via bestlimit. */
if (bestlimit <= CONFIG_MIN_RESERVED_FDS) {
serverLog(LL_WARNING,"Your current 'ulimit -n' "
"of %llu is not enough for the server to start. "
"Please increase your open file limit to at least "
"%llu. Exiting.",
(unsigned long long) oldlimit,
(unsigned long long) maxfiles);
exit(1);
}
serverLog(LL_WARNING,"You requested maxclients of %d "
"requiring at least %llu max file descriptors.",
old_maxclients,
(unsigned long long) maxfiles);
serverLog(LL_WARNING,"Server can't set maximum open files "
"to %llu because of OS error: %s.",
(unsigned long long) maxfiles, strerror(setrlimit_error));
serverLog(LL_WARNING,"Current maximum open files is %llu. "
"maxclients has been reduced to %d to compensate for "
"low ulimit. "
"If you need higher maxclients increase 'ulimit -n'.",
(unsigned long long) bestlimit, server.maxclients);
} else {
serverLog(LL_NOTICE,"Increased maximum number of open files "
"to %llu (it was originally set to %llu).",
(unsigned long long) maxfiles,
(unsigned long long) oldlimit);
}
}
}
}
/* Check that server.tcp_backlog can be actually enforced in Linux according
* to the value of /proc/sys/net/core/somaxconn, or warn about it. */
void checkTcpBacklogSettings(void) {
#ifdef HAVE_PROC_SOMAXCONN
FILE *fp = fopen("/proc/sys/net/core/somaxconn","r");
char buf[1024];
if (!fp) return;
if (fgets(buf,sizeof(buf),fp) != NULL) {
int somaxconn = atoi(buf);
if (somaxconn > 0 && somaxconn < server.tcp_backlog) {
serverLog(LL_WARNING,"WARNING: The TCP backlog setting of %d cannot be enforced because /proc/sys/net/core/somaxconn is set to the lower value of %d.", server.tcp_backlog, somaxconn);
}
}
fclose(fp);
#endif
}
/* Initialize a set of file descriptors to listen to the specified 'port'
* binding the addresses specified in the Redis server configuration.
*
* The listening file descriptors are stored in the integer array 'fds'
* and their number is set in '*count'.
*
* The addresses to bind are specified in the global server.bindaddr array
* and their number is server.bindaddr_count. If the server configuration
* contains no specific addresses to bind, this function will try to
* bind * (all addresses) for both the IPv4 and IPv6 protocols.
*
* On success the function returns C_OK.
*
* On error the function returns C_ERR. For the function to be on
* error, at least one of the server.bindaddr addresses was
* impossible to bind, or no bind addresses were specified in the server
* configuration but the function is not able to bind * for at least
* one of the IPv4 or IPv6 protocols. */
int listenToPort(int port, int *fds, int *count) {
int j;
/* Force binding of 0.0.0.0 if no bind address is specified, always
* entering the loop if j == 0. */
if (server.bindaddr_count == 0) server.bindaddr[0] = NULL;
for (j = 0; j < server.bindaddr_count || j == 0; j++) {
if (server.bindaddr[j] == NULL) {
int unsupported = 0;
/* Bind * for both IPv6 and IPv4, we enter here only if
* server.bindaddr_count == 0. */
fds[*count] = anetTcp6Server(server.neterr,port,NULL,
server.tcp_backlog);
if (fds[*count] != ANET_ERR) {
anetNonBlock(NULL,fds[*count]);
(*count)++;
} else if (errno == EAFNOSUPPORT) {
unsupported++;
serverLog(LL_WARNING,"Not listening to IPv6: unsupported");
}
if (*count == 1 || unsupported) {
/* Bind the IPv4 address as well. */
fds[*count] = anetTcpServer(server.neterr,port,NULL,
server.tcp_backlog);
if (fds[*count] != ANET_ERR) {
anetNonBlock(NULL,fds[*count]);
(*count)++;
} else if (errno == EAFNOSUPPORT) {
unsupported++;
serverLog(LL_WARNING,"Not listening to IPv4: unsupported");
}
}
/* Exit the loop if we were able to bind * on IPv4 and IPv6,
* otherwise fds[*count] will be ANET_ERR and we'll print an
* error and return to the caller with an error. */
if (*count + unsupported == 2) break;
} else if (strchr(server.bindaddr[j],':')) {
/* Bind IPv6 address. */
fds[*count] = anetTcp6Server(server.neterr,port,server.bindaddr[j],
server.tcp_backlog);
} else {
/* Bind IPv4 address. */
fds[*count] = anetTcpServer(server.neterr,port,server.bindaddr[j],
server.tcp_backlog);
}
if (fds[*count] == ANET_ERR) {
serverLog(LL_WARNING,
"Could not create server TCP listening socket %s:%d: %s",
server.bindaddr[j] ? server.bindaddr[j] : "*",
port, server.neterr);
if (errno == ENOPROTOOPT || errno == EPROTONOSUPPORT ||
errno == ESOCKTNOSUPPORT || errno == EPFNOSUPPORT ||
errno == EAFNOSUPPORT || errno == EADDRNOTAVAIL)
continue;
return C_ERR;
}
anetNonBlock(NULL,fds[*count]);
(*count)++;
}
return C_OK;
}
/* Resets the stats that we expose via INFO or other means that we want
* to reset via CONFIG RESETSTAT. The function is also used in order to
* initialize these fields in initServer() at server startup. */
void resetServerStats(void) {
int j;
server.stat_numcommands = 0;
server.stat_numconnections = 0;
server.stat_expiredkeys = 0;
server.stat_expired_stale_perc = 0;
server.stat_expired_time_cap_reached_count = 0;
server.stat_expire_cycle_time_used = 0;
server.stat_evictedkeys = 0;
server.stat_keyspace_misses = 0;
server.stat_keyspace_hits = 0;
server.stat_active_defrag_hits = 0;
server.stat_active_defrag_misses = 0;
server.stat_active_defrag_key_hits = 0;
server.stat_active_defrag_key_misses = 0;
server.stat_active_defrag_scanned = 0;
server.stat_fork_time = 0;
server.stat_fork_rate = 0;
server.stat_rejected_conn = 0;
server.stat_sync_full = 0;
server.stat_sync_partial_ok = 0;
server.stat_sync_partial_err = 0;
for (j = 0; j < STATS_METRIC_COUNT; j++) {
server.inst_metric[j].idx = 0;
server.inst_metric[j].last_sample_time = mstime();
server.inst_metric[j].last_sample_count = 0;
memset(server.inst_metric[j].samples,0,
sizeof(server.inst_metric[j].samples));
}
server.stat_net_input_bytes = 0;
server.stat_net_output_bytes = 0;
server.stat_unexpected_error_replies = 0;
server.aof_delayed_fsync = 0;
}
void initServer(void) {
int j;
signal(SIGHUP, SIG_IGN);
signal(SIGPIPE, SIG_IGN);
setupSignalHandlers();
if (server.syslog_enabled) {
openlog(server.syslog_ident, LOG_PID | LOG_NDELAY | LOG_NOWAIT,
server.syslog_facility);
}
/* Initialization after setting defaults from the config system. */
server.aof_state = server.aof_enabled ? AOF_ON : AOF_OFF;
server.hz = server.config_hz;
server.pid = getpid();
server.current_client = NULL;
server.fixed_time_expire = 0;
server.clients = listCreate();
server.clients_index = raxNew();
server.clients_to_close = listCreate();
server.slaves = listCreate();
server.monitors = listCreate();
server.clients_pending_write = listCreate();
server.clients_pending_read = listCreate();
server.clients_timeout_table = raxNew();
server.slaveseldb = -1; /* Force to emit the first SELECT command. */
server.unblocked_clients = listCreate();
server.ready_keys = listCreate();
server.clients_waiting_acks = listCreate();
server.get_ack_from_slaves = 0;
server.clients_paused = 0;
server.system_memory_size = zmalloc_get_memory_size();
if (server.tls_port && tlsConfigure(&server.tls_ctx_config) == C_ERR) {
serverLog(LL_WARNING, "Failed to configure TLS. Check logs for more info.");
exit(1);
}
createSharedObjects();
adjustOpenFilesLimit();
server.el = aeCreateEventLoop(server.maxclients+CONFIG_FDSET_INCR);
if (server.el == NULL) {
serverLog(LL_WARNING,
"Failed creating the event loop. Error message: '%s'",
strerror(errno));
exit(1);
}
server.db = zmalloc(sizeof(redisDb)*server.dbnum);
/* Open the TCP listening socket for the user commands. */
if (server.port != 0 &&
listenToPort(server.port,server.ipfd,&server.ipfd_count) == C_ERR)
exit(1);
if (server.tls_port != 0 &&
listenToPort(server.tls_port,server.tlsfd,&server.tlsfd_count) == C_ERR)
exit(1);
/* Open the listening Unix domain socket. */
if (server.unixsocket != NULL) {
unlink(server.unixsocket); /* don't care if this fails */
server.sofd = anetUnixServer(server.neterr,server.unixsocket,
server.unixsocketperm, server.tcp_backlog);
if (server.sofd == ANET_ERR) {
serverLog(LL_WARNING, "Opening Unix socket: %s", server.neterr);
exit(1);
}
anetNonBlock(NULL,server.sofd);
}
/* Abort if there are no listening sockets at all. */
if (server.ipfd_count == 0 && server.tlsfd_count == 0 && server.sofd < 0) {
serverLog(LL_WARNING, "Configured to not listen anywhere, exiting.");
exit(1);
}
/* Create the Redis databases, and initialize other internal state. */
for (j = 0; j < server.dbnum; j++) {
server.db[j].dict = dictCreate(&dbDictType,NULL);
server.db[j].expires = dictCreate(&keyptrDictType,NULL);
server.db[j].expires_cursor = 0;
server.db[j].blocking_keys = dictCreate(&keylistDictType,NULL);
server.db[j].ready_keys = dictCreate(&objectKeyPointerValueDictType,NULL);
server.db[j].watched_keys = dictCreate(&keylistDictType,NULL);
server.db[j].id = j;
server.db[j].avg_ttl = 0;
server.db[j].defrag_later = listCreate();
listSetFreeMethod(server.db[j].defrag_later,(void (*)(void*))sdsfree);
}
evictionPoolAlloc(); /* Initialize the LRU keys pool. */
server.pubsub_channels = dictCreate(&keylistDictType,NULL);
server.pubsub_patterns = listCreate();
server.pubsub_patterns_dict = dictCreate(&keylistDictType,NULL);
listSetFreeMethod(server.pubsub_patterns,freePubsubPattern);
listSetMatchMethod(server.pubsub_patterns,listMatchPubsubPattern);
server.cronloops = 0;
server.rdb_child_pid = -1;
server.aof_child_pid = -1;
server.module_child_pid = -1;
server.rdb_child_type = RDB_CHILD_TYPE_NONE;
server.rdb_pipe_conns = NULL;
server.rdb_pipe_numconns = 0;
server.rdb_pipe_numconns_writing = 0;
server.rdb_pipe_buff = NULL;
server.rdb_pipe_bufflen = 0;
server.rdb_bgsave_scheduled = 0;
server.child_info_pipe[0] = -1;
server.child_info_pipe[1] = -1;
server.child_info_data.magic = 0;
aofRewriteBufferReset();
server.aof_buf = sdsempty();
server.lastsave = time(NULL); /* At startup we consider the DB saved. */
server.lastbgsave_try = 0; /* At startup we never tried to BGSAVE. */
server.rdb_save_time_last = -1;
server.rdb_save_time_start = -1;
server.dirty = 0;
resetServerStats();
/* A few stats we don't want to reset: server startup time, and peak mem. */
server.stat_starttime = time(NULL);
server.stat_peak_memory = 0;
server.stat_rdb_cow_bytes = 0;
server.stat_aof_cow_bytes = 0;
server.stat_module_cow_bytes = 0;
server.cron_malloc_stats.zmalloc_used = 0;
server.cron_malloc_stats.process_rss = 0;
server.cron_malloc_stats.allocator_allocated = 0;
server.cron_malloc_stats.allocator_active = 0;
server.cron_malloc_stats.allocator_resident = 0;
server.lastbgsave_status = C_OK;
server.aof_last_write_status = C_OK;
server.aof_last_write_errno = 0;
server.repl_good_slaves_count = 0;
/* Create the timer callback, this is our way to process many background
* operations incrementally, like clients timeout, eviction of unaccessed
* expired keys and so forth. */
if (aeCreateTimeEvent(server.el, 1, serverCron, NULL, NULL) == AE_ERR) {
serverPanic("Can't create event loop timers.");
exit(1);
}
/* Create an event handler for accepting new connections in TCP and Unix
* domain sockets. */
for (j = 0; j < server.ipfd_count; j++) {
if (aeCreateFileEvent(server.el, server.ipfd[j], AE_READABLE,
acceptTcpHandler,NULL) == AE_ERR)
{
serverPanic(
"Unrecoverable error creating server.ipfd file event.");
}
}
for (j = 0; j < server.tlsfd_count; j++) {
if (aeCreateFileEvent(server.el, server.tlsfd[j], AE_READABLE,
acceptTLSHandler,NULL) == AE_ERR)
{
serverPanic(
"Unrecoverable error creating server.tlsfd file event.");
}
}
if (server.sofd > 0 && aeCreateFileEvent(server.el,server.sofd,AE_READABLE,
acceptUnixHandler,NULL) == AE_ERR) serverPanic("Unrecoverable error creating server.sofd file event.");
/* Register a readable event for the pipe used to awake the event loop
* when a blocked client in a module needs attention. */
if (aeCreateFileEvent(server.el, server.module_blocked_pipe[0], AE_READABLE,
moduleBlockedClientPipeReadable,NULL) == AE_ERR) {
serverPanic(
"Error registering the readable event for the module "
"blocked clients subsystem.");
}
/* Open the AOF file if needed. */
if (server.aof_state == AOF_ON) {
server.aof_fd = open(server.aof_filename,
O_WRONLY|O_APPEND|O_CREAT,0644);
if (server.aof_fd == -1) {
serverLog(LL_WARNING, "Can't open the append-only file: %s",
strerror(errno));
exit(1);
}
}
/* 32 bit instances are limited to 4GB of address space, so if there is
* no explicit limit in the user provided configuration we set a limit
* at 3 GB using maxmemory with 'noeviction' policy'. This avoids
* useless crashes of the Redis instance for out of memory. */
if (server.arch_bits == 32 && server.maxmemory == 0) {
serverLog(LL_WARNING,"Warning: 32 bit instance detected but no memory limit set. Setting 3 GB maxmemory limit with 'noeviction' policy now.");
server.maxmemory = 3072LL*(1024*1024); /* 3 GB */
server.maxmemory_policy = MAXMEMORY_NO_EVICTION;
}
if (server.cluster_enabled) clusterInit();
replicationScriptCacheInit();
scriptingInit(1);
slowlogInit();
latencyMonitorInit();
}
/* Some steps in server initialization need to be done last (after modules
* are loaded).
* Specifically, creation of threads due to a race bug in ld.so, in which
* Thread Local Storage initialization collides with dlopen call.
* see: https://sourceware.org/bugzilla/show_bug.cgi?id=19329 */
void InitServerLast() {
bioInit();
initThreadedIO();
set_jemalloc_bg_thread(server.jemalloc_bg_thread);
server.initial_memory_usage = zmalloc_used_memory();
}
/* Parse the flags string description 'strflags' and set them to the
* command 'c'. If the flags are all valid C_OK is returned, otherwise
* C_ERR is returned (yet the recognized flags are set in the command). */
int populateCommandTableParseFlags(struct redisCommand *c, char *strflags) {
int argc;
sds *argv;
/* Split the line into arguments for processing. */
argv = sdssplitargs(strflags,&argc);
if (argv == NULL) return C_ERR;
for (int j = 0; j < argc; j++) {
char *flag = argv[j];
if (!strcasecmp(flag,"write")) {
c->flags |= CMD_WRITE|CMD_CATEGORY_WRITE;
} else if (!strcasecmp(flag,"read-only")) {
c->flags |= CMD_READONLY|CMD_CATEGORY_READ;
} else if (!strcasecmp(flag,"use-memory")) {
c->flags |= CMD_DENYOOM;
} else if (!strcasecmp(flag,"admin")) {
c->flags |= CMD_ADMIN|CMD_CATEGORY_ADMIN|CMD_CATEGORY_DANGEROUS;
} else if (!strcasecmp(flag,"pub-sub")) {
c->flags |= CMD_PUBSUB|CMD_CATEGORY_PUBSUB;
} else if (!strcasecmp(flag,"no-script")) {
c->flags |= CMD_NOSCRIPT;
} else if (!strcasecmp(flag,"random")) {
c->flags |= CMD_RANDOM;
} else if (!strcasecmp(flag,"to-sort")) {
c->flags |= CMD_SORT_FOR_SCRIPT;
} else if (!strcasecmp(flag,"ok-loading")) {
c->flags |= CMD_LOADING;
} else if (!strcasecmp(flag,"ok-stale")) {
c->flags |= CMD_STALE;
} else if (!strcasecmp(flag,"no-monitor")) {
c->flags |= CMD_SKIP_MONITOR;
} else if (!strcasecmp(flag,"no-slowlog")) {
c->flags |= CMD_SKIP_SLOWLOG;
} else if (!strcasecmp(flag,"cluster-asking")) {
c->flags |= CMD_ASKING;
} else if (!strcasecmp(flag,"fast")) {
c->flags |= CMD_FAST | CMD_CATEGORY_FAST;
} else if (!strcasecmp(flag,"no-auth")) {
c->flags |= CMD_NO_AUTH;
} else {
/* Parse ACL categories here if the flag name starts with @. */
uint64_t catflag;
if (flag[0] == '@' &&
(catflag = ACLGetCommandCategoryFlagByName(flag+1)) != 0)
{
c->flags |= catflag;
} else {
sdsfreesplitres(argv,argc);
return C_ERR;
}
}
}
/* If it's not @fast is @slow in this binary world. */
if (!(c->flags & CMD_CATEGORY_FAST)) c->flags |= CMD_CATEGORY_SLOW;
sdsfreesplitres(argv,argc);
return C_OK;
}
/* Populates the Redis Command Table starting from the hard coded list
* we have on top of redis.c file. */
void populateCommandTable(void) {
int j;
int numcommands = sizeof(redisCommandTable)/sizeof(struct redisCommand);
for (j = 0; j < numcommands; j++) {
struct redisCommand *c = redisCommandTable+j;
int retval1, retval2;
/* Translate the command string flags description into an actual
* set of flags. */
if (populateCommandTableParseFlags(c,c->sflags) == C_ERR)
serverPanic("Unsupported command flag");
c->id = ACLGetCommandID(c->name); /* Assign the ID used for ACL. */
retval1 = dictAdd(server.commands, sdsnew(c->name), c);
/* Populate an additional dictionary that will be unaffected
* by rename-command statements in redis.conf. */
retval2 = dictAdd(server.orig_commands, sdsnew(c->name), c);
serverAssert(retval1 == DICT_OK && retval2 == DICT_OK);
}
}
void resetCommandTableStats(void) {
struct redisCommand *c;
dictEntry *de;
dictIterator *di;
di = dictGetSafeIterator(server.commands);
while((de = dictNext(di)) != NULL) {
c = (struct redisCommand *) dictGetVal(de);
c->microseconds = 0;
c->calls = 0;
}
dictReleaseIterator(di);
}
/* ========================== Redis OP Array API ============================ */
void redisOpArrayInit(redisOpArray *oa) {
oa->ops = NULL;
oa->numops = 0;
}
int redisOpArrayAppend(redisOpArray *oa, struct redisCommand *cmd, int dbid,
robj **argv, int argc, int target)
{
redisOp *op;
oa->ops = zrealloc(oa->ops,sizeof(redisOp)*(oa->numops+1));
op = oa->ops+oa->numops;
op->cmd = cmd;
op->dbid = dbid;
op->argv = argv;
op->argc = argc;
op->target = target;
oa->numops++;
return oa->numops;
}
void redisOpArrayFree(redisOpArray *oa) {
while(oa->numops) {
int j;
redisOp *op;
oa->numops--;
op = oa->ops+oa->numops;
for (j = 0; j < op->argc; j++)
decrRefCount(op->argv[j]);
zfree(op->argv);
}
zfree(oa->ops);
}
/* ====================== Commands lookup and execution ===================== */
struct redisCommand *lookupCommand(sds name) {
return dictFetchValue(server.commands, name);
}
struct redisCommand *lookupCommandByCString(char *s) {
struct redisCommand *cmd;
sds name = sdsnew(s);
cmd = dictFetchValue(server.commands, name);
sdsfree(name);
return cmd;
}
/* Lookup the command in the current table, if not found also check in
* the original table containing the original command names unaffected by
* redis.conf rename-command statement.
*
* This is used by functions rewriting the argument vector such as
* rewriteClientCommandVector() in order to set client->cmd pointer
* correctly even if the command was renamed. */
struct redisCommand *lookupCommandOrOriginal(sds name) {
struct redisCommand *cmd = dictFetchValue(server.commands, name);
if (!cmd) cmd = dictFetchValue(server.orig_commands,name);
return cmd;
}
/* Propagate the specified command (in the context of the specified database id)
* to AOF and Slaves.
*
* flags are an xor between:
* + PROPAGATE_NONE (no propagation of command at all)
* + PROPAGATE_AOF (propagate into the AOF file if is enabled)
* + PROPAGATE_REPL (propagate into the replication link)
*
* This should not be used inside commands implementation. Use instead
* alsoPropagate(), preventCommandPropagation(), forceCommandPropagation().
*/
void propagate(struct redisCommand *cmd, int dbid, robj **argv, int argc,
int flags)
{
if (server.aof_state != AOF_OFF && flags & PROPAGATE_AOF)
feedAppendOnlyFile(cmd,dbid,argv,argc);
if (flags & PROPAGATE_REPL)
replicationFeedSlaves(server.slaves,dbid,argv,argc);
}
/* Used inside commands to schedule the propagation of additional commands
* after the current command is propagated to AOF / Replication.
*
* 'cmd' must be a pointer to the Redis command to replicate, dbid is the
* database ID the command should be propagated into.
* Arguments of the command to propagte are passed as an array of redis
* objects pointers of len 'argc', using the 'argv' vector.
*
* The function does not take a reference to the passed 'argv' vector,
* so it is up to the caller to release the passed argv (but it is usually
* stack allocated). The function autoamtically increments ref count of
* passed objects, so the caller does not need to. */
void alsoPropagate(struct redisCommand *cmd, int dbid, robj **argv, int argc,
int target)
{
robj **argvcopy;
int j;
if (server.loading) return; /* No propagation during loading. */
argvcopy = zmalloc(sizeof(robj*)*argc);
for (j = 0; j < argc; j++) {
argvcopy[j] = argv[j];
incrRefCount(argv[j]);
}
redisOpArrayAppend(&server.also_propagate,cmd,dbid,argvcopy,argc,target);
}
/* It is possible to call the function forceCommandPropagation() inside a
* Redis command implementation in order to to force the propagation of a
* specific command execution into AOF / Replication. */
void forceCommandPropagation(client *c, int flags) {
if (flags & PROPAGATE_REPL) c->flags |= CLIENT_FORCE_REPL;
if (flags & PROPAGATE_AOF) c->flags |= CLIENT_FORCE_AOF;
}
/* Avoid that the executed command is propagated at all. This way we
* are free to just propagate what we want using the alsoPropagate()
* API. */
void preventCommandPropagation(client *c) {
c->flags |= CLIENT_PREVENT_PROP;
}
/* AOF specific version of preventCommandPropagation(). */
void preventCommandAOF(client *c) {
c->flags |= CLIENT_PREVENT_AOF_PROP;
}
/* Replication specific version of preventCommandPropagation(). */
void preventCommandReplication(client *c) {
c->flags |= CLIENT_PREVENT_REPL_PROP;
}
/* Call() is the core of Redis execution of a command.
*
* The following flags can be passed:
* CMD_CALL_NONE No flags.
* CMD_CALL_SLOWLOG Check command speed and log in the slow log if needed.
* CMD_CALL_STATS Populate command stats.
* CMD_CALL_PROPAGATE_AOF Append command to AOF if it modified the dataset
* or if the client flags are forcing propagation.
* CMD_CALL_PROPAGATE_REPL Send command to slaves if it modified the dataset
* or if the client flags are forcing propagation.
* CMD_CALL_PROPAGATE Alias for PROPAGATE_AOF|PROPAGATE_REPL.
* CMD_CALL_FULL Alias for SLOWLOG|STATS|PROPAGATE.
*
* The exact propagation behavior depends on the client flags.
* Specifically:
*
* 1. If the client flags CLIENT_FORCE_AOF or CLIENT_FORCE_REPL are set
* and assuming the corresponding CMD_CALL_PROPAGATE_AOF/REPL is set
* in the call flags, then the command is propagated even if the
* dataset was not affected by the command.
* 2. If the client flags CLIENT_PREVENT_REPL_PROP or CLIENT_PREVENT_AOF_PROP
* are set, the propagation into AOF or to slaves is not performed even
* if the command modified the dataset.
*
* Note that regardless of the client flags, if CMD_CALL_PROPAGATE_AOF
* or CMD_CALL_PROPAGATE_REPL are not set, then respectively AOF or
* slaves propagation will never occur.
*
* Client flags are modified by the implementation of a given command
* using the following API:
*
* forceCommandPropagation(client *c, int flags);
* preventCommandPropagation(client *c);
* preventCommandAOF(client *c);
* preventCommandReplication(client *c);
*
*/
void call(client *c, int flags) {
long long dirty;
ustime_t start, duration;
int client_old_flags = c->flags;
struct redisCommand *real_cmd = c->cmd;
server.fixed_time_expire++;
/* Sent the command to clients in MONITOR mode, only if the commands are
* not generated from reading an AOF. */
if (listLength(server.monitors) &&
!server.loading &&
!(c->cmd->flags & (CMD_SKIP_MONITOR|CMD_ADMIN)))
{
replicationFeedMonitors(c,server.monitors,c->db->id,c->argv,c->argc);
}
/* Initialization: clear the flags that must be set by the command on
* demand, and initialize the array for additional commands propagation. */
c->flags &= ~(CLIENT_FORCE_AOF|CLIENT_FORCE_REPL|CLIENT_PREVENT_PROP);
redisOpArray prev_also_propagate = server.also_propagate;
redisOpArrayInit(&server.also_propagate);
/* Call the command. */
dirty = server.dirty;
updateCachedTime(0);
start = server.ustime;
c->cmd->proc(c);
duration = ustime()-start;
dirty = server.dirty-dirty;
if (dirty < 0) dirty = 0;
/* When EVAL is called loading the AOF we don't want commands called
* from Lua to go into the slowlog or to populate statistics. */
if (server.loading && c->flags & CLIENT_LUA)
flags &= ~(CMD_CALL_SLOWLOG | CMD_CALL_STATS);
/* If the caller is Lua, we want to force the EVAL caller to propagate
* the script if the command flag or client flag are forcing the
* propagation. */
if (c->flags & CLIENT_LUA && server.lua_caller) {
if (c->flags & CLIENT_FORCE_REPL)
server.lua_caller->flags |= CLIENT_FORCE_REPL;
if (c->flags & CLIENT_FORCE_AOF)
server.lua_caller->flags |= CLIENT_FORCE_AOF;
}
/* Log the command into the Slow log if needed, and populate the
* per-command statistics that we show in INFO commandstats. */
if (flags & CMD_CALL_SLOWLOG && !(c->cmd->flags & CMD_SKIP_SLOWLOG)) {
char *latency_event = (c->cmd->flags & CMD_FAST) ?
"fast-command" : "command";
latencyAddSampleIfNeeded(latency_event,duration/1000);
slowlogPushEntryIfNeeded(c,c->argv,c->argc,duration);
}
if (flags & CMD_CALL_STATS) {
/* use the real command that was executed (cmd and lastamc) may be
* different, in case of MULTI-EXEC or re-written commands such as
* EXPIRE, GEOADD, etc. */
real_cmd->microseconds += duration;
real_cmd->calls++;
}
/* Propagate the command into the AOF and replication link */
if (flags & CMD_CALL_PROPAGATE &&
(c->flags & CLIENT_PREVENT_PROP) != CLIENT_PREVENT_PROP)
{
int propagate_flags = PROPAGATE_NONE;
/* Check if the command operated changes in the data set. If so
* set for replication / AOF propagation. */
if (dirty) propagate_flags |= (PROPAGATE_AOF|PROPAGATE_REPL);
/* If the client forced AOF / replication of the command, set
* the flags regardless of the command effects on the data set. */
if (c->flags & CLIENT_FORCE_REPL) propagate_flags |= PROPAGATE_REPL;
if (c->flags & CLIENT_FORCE_AOF) propagate_flags |= PROPAGATE_AOF;
/* However prevent AOF / replication propagation if the command
* implementations called preventCommandPropagation() or similar,
* or if we don't have the call() flags to do so. */
if (c->flags & CLIENT_PREVENT_REPL_PROP ||
!(flags & CMD_CALL_PROPAGATE_REPL))
propagate_flags &= ~PROPAGATE_REPL;
if (c->flags & CLIENT_PREVENT_AOF_PROP ||
!(flags & CMD_CALL_PROPAGATE_AOF))
propagate_flags &= ~PROPAGATE_AOF;
/* Call propagate() only if at least one of AOF / replication
* propagation is needed. Note that modules commands handle replication
* in an explicit way, so we never replicate them automatically. */
if (propagate_flags != PROPAGATE_NONE && !(c->cmd->flags & CMD_MODULE))
propagate(c->cmd,c->db->id,c->argv,c->argc,propagate_flags);
}
/* Restore the old replication flags, since call() can be executed
* recursively. */
c->flags &= ~(CLIENT_FORCE_AOF|CLIENT_FORCE_REPL|CLIENT_PREVENT_PROP);
c->flags |= client_old_flags &
(CLIENT_FORCE_AOF|CLIENT_FORCE_REPL|CLIENT_PREVENT_PROP);
/* Handle the alsoPropagate() API to handle commands that want to propagate
* multiple separated commands. Note that alsoPropagate() is not affected
* by CLIENT_PREVENT_PROP flag. */
if (server.also_propagate.numops) {
int j;
redisOp *rop;
if (flags & CMD_CALL_PROPAGATE) {
int multi_emitted = 0;
/* Wrap the commands in server.also_propagate array,
* but don't wrap it if we are already in MULTI context,
* in case the nested MULTI/EXEC.
*
* And if the array contains only one command, no need to
* wrap it, since the single command is atomic. */
if (server.also_propagate.numops > 1 &&
!(c->cmd->flags & CMD_MODULE) &&
!(c->flags & CLIENT_MULTI) &&
!(flags & CMD_CALL_NOWRAP))
{
execCommandPropagateMulti(c);
multi_emitted = 1;
}
for (j = 0; j < server.also_propagate.numops; j++) {
rop = &server.also_propagate.ops[j];
int target = rop->target;
/* Whatever the command wish is, we honor the call() flags. */
if (!(flags&CMD_CALL_PROPAGATE_AOF)) target &= ~PROPAGATE_AOF;
if (!(flags&CMD_CALL_PROPAGATE_REPL)) target &= ~PROPAGATE_REPL;
if (target)
propagate(rop->cmd,rop->dbid,rop->argv,rop->argc,target);
}
if (multi_emitted) {
execCommandPropagateExec(c);
}
}
redisOpArrayFree(&server.also_propagate);
}
server.also_propagate = prev_also_propagate;
/* If the client has keys tracking enabled for client side caching,
* make sure to remember the keys it fetched via this command. */
if (c->cmd->flags & CMD_READONLY) {
client *caller = (c->flags & CLIENT_LUA && server.lua_caller) ?
server.lua_caller : c;
if (caller->flags & CLIENT_TRACKING &&
!(caller->flags & CLIENT_TRACKING_BCAST))
{
trackingRememberKeys(caller);
}
}
server.fixed_time_expire--;
server.stat_numcommands++;
}
/* If this function gets called we already read a whole
* command, arguments are in the client argv/argc fields.
* processCommand() execute the command or prepare the
* server for a bulk read from the client.
*
* If C_OK is returned the client is still alive and valid and
* other operations can be performed by the caller. Otherwise
* if C_ERR is returned the client was destroyed (i.e. after QUIT). */
int processCommand(client *c) {
moduleCallCommandFilters(c);
/* The QUIT command is handled separately. Normal command procs will
* go through checking for replication and QUIT will cause trouble
* when FORCE_REPLICATION is enabled and would be implemented in
* a regular command proc. */
if (!strcasecmp(c->argv[0]->ptr,"quit")) {
addReply(c,shared.ok);
c->flags |= CLIENT_CLOSE_AFTER_REPLY;
return C_ERR;
}
/* Now lookup the command and check ASAP about trivial error conditions
* such as wrong arity, bad command name and so forth. */
c->cmd = c->lastcmd = lookupCommand(c->argv[0]->ptr);
if (!c->cmd) {
flagTransaction(c);
sds args = sdsempty();
int i;
for (i=1; i < c->argc && sdslen(args) < 128; i++)
args = sdscatprintf(args, "`%.*s`, ", 128-(int)sdslen(args), (char*)c->argv[i]->ptr);
addReplyErrorFormat(c,"unknown command `%s`, with args beginning with: %s",
(char*)c->argv[0]->ptr, args);
sdsfree(args);
return C_OK;
} else if ((c->cmd->arity > 0 && c->cmd->arity != c->argc) ||
(c->argc < -c->cmd->arity)) {
flagTransaction(c);
addReplyErrorFormat(c,"wrong number of arguments for '%s' command",
c->cmd->name);
return C_OK;
}
/* Check if the user is authenticated. This check is skipped in case
* the default user is flagged as "nopass" and is active. */
int auth_required = (!(DefaultUser->flags & USER_FLAG_NOPASS) ||
(DefaultUser->flags & USER_FLAG_DISABLED)) &&
!c->authenticated;
if (auth_required) {
/* AUTH and HELLO and no auth modules are valid even in
* non-authenticated state. */
if (!(c->cmd->flags & CMD_NO_AUTH)) {
flagTransaction(c);
addReply(c,shared.noautherr);
return C_OK;
}
}
/* Check if the user can run this command according to the current
* ACLs. */
int acl_keypos;
int acl_retval = ACLCheckCommandPerm(c,&acl_keypos);
if (acl_retval != ACL_OK) {
addACLLogEntry(c,acl_retval,acl_keypos,NULL);
flagTransaction(c);
if (acl_retval == ACL_DENIED_CMD)
addReplyErrorFormat(c,
"-NOPERM this user has no permissions to run "
"the '%s' command or its subcommand", c->cmd->name);
else
addReplyErrorFormat(c,
"-NOPERM this user has no permissions to access "
"one of the keys used as arguments");
return C_OK;
}
/* If cluster is enabled perform the cluster redirection here.
* However we don't perform the redirection if:
* 1) The sender of this command is our master.
* 2) The command has no key arguments. */
if (server.cluster_enabled &&
!(c->flags & CLIENT_MASTER) &&
!(c->flags & CLIENT_LUA &&
server.lua_caller->flags & CLIENT_MASTER) &&
!(c->cmd->getkeys_proc == NULL && c->cmd->firstkey == 0 &&
c->cmd->proc != execCommand))
{
int hashslot;
int error_code;
clusterNode *n = getNodeByQuery(c,c->cmd,c->argv,c->argc,
&hashslot,&error_code);
if (n == NULL || n != server.cluster->myself) {
if (c->cmd->proc == execCommand) {
discardTransaction(c);
} else {
flagTransaction(c);
}
clusterRedirectClient(c,n,hashslot,error_code);
return C_OK;
}
}
/* Handle the maxmemory directive.
*
* Note that we do not want to reclaim memory if we are here re-entering
* the event loop since there is a busy Lua script running in timeout
* condition, to avoid mixing the propagation of scripts with the
* propagation of DELs due to eviction. */
if (server.maxmemory && !server.lua_timedout) {
int out_of_memory = freeMemoryIfNeededAndSafe() == C_ERR;
/* freeMemoryIfNeeded may flush slave output buffers. This may result