X-Git-Url: https://git.saurik.com/redis.git/blobdiff_plain/e2641e09cc0daf44f63f654230f72d22acf3a9af..d4d208595c78d78196c926c55799bc81ae06f230:/src/redis.c?ds=inline diff --git a/src/redis.c b/src/redis.c index 5f539216..3914003d 100644 --- a/src/redis.c +++ b/src/redis.c @@ -28,11 +28,8 @@ */ #include "redis.h" - -#ifdef HAVE_BACKTRACE -#include -#include -#endif /* HAVE_BACKTRACE */ +#include "slowlog.h" +#include "bio.h" #include #include @@ -41,7 +38,6 @@ #include #include #include -#include #include #include #include @@ -51,13 +47,13 @@ #include #include #include -#include +#include /* Our shared "common" objects */ struct sharedObjectsStruct shared; -/* Global vars that are actally used as constants. The following double +/* 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. */ @@ -68,145 +64,229 @@ double R_Zero, R_PosInf, R_NegInf, R_Nan; /* Global vars */ struct redisServer server; /* server global state */ struct redisCommand *commandTable; -struct redisCommand readonlyCommandTable[] = { - {"get",getCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"set",setCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,0,0,0}, - {"setnx",setnxCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,0,0,0}, - {"setex",setexCommand,4,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,0,0,0}, - {"append",appendCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"substr",substrCommand,4,REDIS_CMD_INLINE,NULL,1,1,1}, - {"del",delCommand,-2,REDIS_CMD_INLINE,NULL,0,0,0}, - {"exists",existsCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"incr",incrCommand,2,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"decr",decrCommand,2,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"mget",mgetCommand,-2,REDIS_CMD_INLINE,NULL,1,-1,1}, - {"rpush",rpushCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"lpush",lpushCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"rpushx",rpushxCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"lpushx",lpushxCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"linsert",linsertCommand,5,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"rpop",rpopCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"lpop",lpopCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"brpop",brpopCommand,-3,REDIS_CMD_INLINE,NULL,1,1,1}, - {"blpop",blpopCommand,-3,REDIS_CMD_INLINE,NULL,1,1,1}, - {"llen",llenCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"lindex",lindexCommand,3,REDIS_CMD_INLINE,NULL,1,1,1}, - {"lset",lsetCommand,4,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"lrange",lrangeCommand,4,REDIS_CMD_INLINE,NULL,1,1,1}, - {"ltrim",ltrimCommand,4,REDIS_CMD_INLINE,NULL,1,1,1}, - {"lrem",lremCommand,4,REDIS_CMD_BULK,NULL,1,1,1}, - {"rpoplpush",rpoplpushcommand,3,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,2,1}, - {"sadd",saddCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"srem",sremCommand,3,REDIS_CMD_BULK,NULL,1,1,1}, - {"smove",smoveCommand,4,REDIS_CMD_BULK,NULL,1,2,1}, - {"sismember",sismemberCommand,3,REDIS_CMD_BULK,NULL,1,1,1}, - {"scard",scardCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"spop",spopCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"srandmember",srandmemberCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"sinter",sinterCommand,-2,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,-1,1}, - {"sinterstore",sinterstoreCommand,-3,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,2,-1,1}, - {"sunion",sunionCommand,-2,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,-1,1}, - {"sunionstore",sunionstoreCommand,-3,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,2,-1,1}, - {"sdiff",sdiffCommand,-2,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,-1,1}, - {"sdiffstore",sdiffstoreCommand,-3,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,2,-1,1}, - {"smembers",sinterCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"zadd",zaddCommand,4,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"zincrby",zincrbyCommand,4,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"zrem",zremCommand,3,REDIS_CMD_BULK,NULL,1,1,1}, - {"zremrangebyscore",zremrangebyscoreCommand,4,REDIS_CMD_INLINE,NULL,1,1,1}, - {"zremrangebyrank",zremrangebyrankCommand,4,REDIS_CMD_INLINE,NULL,1,1,1}, - {"zunionstore",zunionstoreCommand,-4,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,zunionInterBlockClientOnSwappedKeys,0,0,0}, - {"zinterstore",zinterstoreCommand,-4,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,zunionInterBlockClientOnSwappedKeys,0,0,0}, - {"zrange",zrangeCommand,-4,REDIS_CMD_INLINE,NULL,1,1,1}, - {"zrangebyscore",zrangebyscoreCommand,-4,REDIS_CMD_INLINE,NULL,1,1,1}, - {"zcount",zcountCommand,4,REDIS_CMD_INLINE,NULL,1,1,1}, - {"zrevrange",zrevrangeCommand,-4,REDIS_CMD_INLINE,NULL,1,1,1}, - {"zcard",zcardCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"zscore",zscoreCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"zrank",zrankCommand,3,REDIS_CMD_BULK,NULL,1,1,1}, - {"zrevrank",zrevrankCommand,3,REDIS_CMD_BULK,NULL,1,1,1}, - {"hset",hsetCommand,4,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"hsetnx",hsetnxCommand,4,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"hget",hgetCommand,3,REDIS_CMD_BULK,NULL,1,1,1}, - {"hmset",hmsetCommand,-4,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"hmget",hmgetCommand,-3,REDIS_CMD_BULK,NULL,1,1,1}, - {"hincrby",hincrbyCommand,4,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"hdel",hdelCommand,3,REDIS_CMD_BULK,NULL,1,1,1}, - {"hlen",hlenCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"hkeys",hkeysCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"hvals",hvalsCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"hgetall",hgetallCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"hexists",hexistsCommand,3,REDIS_CMD_BULK,NULL,1,1,1}, - {"incrby",incrbyCommand,3,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"decrby",decrbyCommand,3,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"getset",getsetCommand,3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"mset",msetCommand,-3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,-1,2}, - {"msetnx",msetnxCommand,-3,REDIS_CMD_BULK|REDIS_CMD_DENYOOM,NULL,1,-1,2}, - {"randomkey",randomkeyCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"select",selectCommand,2,REDIS_CMD_INLINE,NULL,0,0,0}, - {"move",moveCommand,3,REDIS_CMD_INLINE,NULL,1,1,1}, - {"rename",renameCommand,3,REDIS_CMD_INLINE,NULL,1,1,1}, - {"renamenx",renamenxCommand,3,REDIS_CMD_INLINE,NULL,1,1,1}, - {"expire",expireCommand,3,REDIS_CMD_INLINE,NULL,0,0,0}, - {"expireat",expireatCommand,3,REDIS_CMD_INLINE,NULL,0,0,0}, - {"keys",keysCommand,2,REDIS_CMD_INLINE,NULL,0,0,0}, - {"dbsize",dbsizeCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"auth",authCommand,2,REDIS_CMD_INLINE,NULL,0,0,0}, - {"ping",pingCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"echo",echoCommand,2,REDIS_CMD_BULK,NULL,0,0,0}, - {"save",saveCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"bgsave",bgsaveCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"bgrewriteaof",bgrewriteaofCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"shutdown",shutdownCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"lastsave",lastsaveCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"type",typeCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"multi",multiCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"exec",execCommand,1,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,execBlockClientOnSwappedKeys,0,0,0}, - {"discard",discardCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"sync",syncCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"flushdb",flushdbCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"flushall",flushallCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"sort",sortCommand,-2,REDIS_CMD_INLINE|REDIS_CMD_DENYOOM,NULL,1,1,1}, - {"info",infoCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"monitor",monitorCommand,1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"ttl",ttlCommand,2,REDIS_CMD_INLINE,NULL,1,1,1}, - {"slaveof",slaveofCommand,3,REDIS_CMD_INLINE,NULL,0,0,0}, - {"debug",debugCommand,-2,REDIS_CMD_INLINE,NULL,0,0,0}, - {"config",configCommand,-2,REDIS_CMD_BULK,NULL,0,0,0}, - {"subscribe",subscribeCommand,-2,REDIS_CMD_INLINE,NULL,0,0,0}, - {"unsubscribe",unsubscribeCommand,-1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"psubscribe",psubscribeCommand,-2,REDIS_CMD_INLINE,NULL,0,0,0}, - {"punsubscribe",punsubscribeCommand,-1,REDIS_CMD_INLINE,NULL,0,0,0}, - {"publish",publishCommand,3,REDIS_CMD_BULK|REDIS_CMD_FORCE_REPLICATION,NULL,0,0,0}, - {"watch",watchCommand,-2,REDIS_CMD_INLINE,NULL,0,0,0}, - {"unwatch",unwatchCommand,1,REDIS_CMD_INLINE,NULL,0,0,0} + +/* 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. + * + * The flags, microseconds and calls fields are computed by Redis and should + * always be set to zero. + * + * Command flags are expressed using strings where every character represents + * a flag. Later the populateCommandTable() function will take care of + * populating the real 'flags' field using this characters. + * + * This is the meaning of the flags: + * + * w: write command (may modify the key space). + * r: read command (will never modify the key space). + * m: may increase memory usage once called. Don't allow if out of memory. + * a: admin command, like SAVE or SHUTDOWN. + * p: Pub/Sub related command. + * f: force replication of this command, regarless of server.dirty. + * s: command not allowed in scripts. + * R: 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. */ +struct redisCommand redisCommandTable[] = { + {"get",getCommand,2,"r",0,NULL,1,1,1,0,0}, + {"set",setCommand,3,"wm",0,noPreloadGetKeys,1,1,1,0,0}, + {"setnx",setnxCommand,3,"wm",0,noPreloadGetKeys,1,1,1,0,0}, + {"setex",setexCommand,4,"wm",0,noPreloadGetKeys,1,1,1,0,0}, + {"psetex",psetexCommand,4,"wm",0,noPreloadGetKeys,1,1,1,0,0}, + {"append",appendCommand,3,"wm",0,NULL,1,1,1,0,0}, + {"strlen",strlenCommand,2,"r",0,NULL,1,1,1,0,0}, + {"del",delCommand,-2,"w",0,noPreloadGetKeys,1,-1,1,0,0}, + {"exists",existsCommand,2,"r",0,NULL,1,1,1,0,0}, + {"setbit",setbitCommand,4,"wm",0,NULL,1,1,1,0,0}, + {"getbit",getbitCommand,3,"r",0,NULL,1,1,1,0,0}, + {"setrange",setrangeCommand,4,"wm",0,NULL,1,1,1,0,0}, + {"getrange",getrangeCommand,4,"r",0,NULL,1,1,1,0,0}, + {"substr",getrangeCommand,4,"r",0,NULL,1,1,1,0,0}, + {"incr",incrCommand,2,"wm",0,NULL,1,1,1,0,0}, + {"decr",decrCommand,2,"wm",0,NULL,1,1,1,0,0}, + {"mget",mgetCommand,-2,"r",0,NULL,1,-1,1,0,0}, + {"rpush",rpushCommand,-3,"wm",0,NULL,1,1,1,0,0}, + {"lpush",lpushCommand,-3,"wm",0,NULL,1,1,1,0,0}, + {"rpushx",rpushxCommand,3,"wm",0,NULL,1,1,1,0,0}, + {"lpushx",lpushxCommand,3,"wm",0,NULL,1,1,1,0,0}, + {"linsert",linsertCommand,5,"wm",0,NULL,1,1,1,0,0}, + {"rpop",rpopCommand,2,"w",0,NULL,1,1,1,0,0}, + {"lpop",lpopCommand,2,"w",0,NULL,1,1,1,0,0}, + {"brpop",brpopCommand,-3,"ws",0,NULL,1,1,1,0,0}, + {"brpoplpush",brpoplpushCommand,4,"wms",0,NULL,1,2,1,0,0}, + {"blpop",blpopCommand,-3,"ws",0,NULL,1,-2,1,0,0}, + {"llen",llenCommand,2,"r",0,NULL,1,1,1,0,0}, + {"lindex",lindexCommand,3,"r",0,NULL,1,1,1,0,0}, + {"lset",lsetCommand,4,"wm",0,NULL,1,1,1,0,0}, + {"lrange",lrangeCommand,4,"r",0,NULL,1,1,1,0,0}, + {"ltrim",ltrimCommand,4,"w",0,NULL,1,1,1,0,0}, + {"lrem",lremCommand,4,"w",0,NULL,1,1,1,0,0}, + {"rpoplpush",rpoplpushCommand,3,"wm",0,NULL,1,2,1,0,0}, + {"sadd",saddCommand,-3,"wm",0,NULL,1,1,1,0,0}, + {"srem",sremCommand,-3,"w",0,NULL,1,1,1,0,0}, + {"smove",smoveCommand,4,"w",0,NULL,1,2,1,0,0}, + {"sismember",sismemberCommand,3,"r",0,NULL,1,1,1,0,0}, + {"scard",scardCommand,2,"r",0,NULL,1,1,1,0,0}, + {"spop",spopCommand,2,"wRs",0,NULL,1,1,1,0,0}, + {"srandmember",srandmemberCommand,2,"rR",0,NULL,1,1,1,0,0}, + {"sinter",sinterCommand,-2,"r",0,NULL,1,-1,1,0,0}, + {"sinterstore",sinterstoreCommand,-3,"wm",0,NULL,1,-1,1,0,0}, + {"sunion",sunionCommand,-2,"r",0,NULL,1,-1,1,0,0}, + {"sunionstore",sunionstoreCommand,-3,"wm",0,NULL,1,-1,1,0,0}, + {"sdiff",sdiffCommand,-2,"r",0,NULL,1,-1,1,0,0}, + {"sdiffstore",sdiffstoreCommand,-3,"wm",0,NULL,1,-1,1,0,0}, + {"smembers",sinterCommand,2,"r",0,NULL,1,1,1,0,0}, + {"zadd",zaddCommand,-4,"wm",0,NULL,1,1,1,0,0}, + {"zincrby",zincrbyCommand,4,"wm",0,NULL,1,1,1,0,0}, + {"zrem",zremCommand,-3,"w",0,NULL,1,1,1,0,0}, + {"zremrangebyscore",zremrangebyscoreCommand,4,"w",0,NULL,1,1,1,0,0}, + {"zremrangebyrank",zremrangebyrankCommand,4,"w",0,NULL,1,1,1,0,0}, + {"zunionstore",zunionstoreCommand,-4,"wm",0,zunionInterGetKeys,0,0,0,0,0}, + {"zinterstore",zinterstoreCommand,-4,"wm",0,zunionInterGetKeys,0,0,0,0,0}, + {"zrange",zrangeCommand,-4,"r",0,NULL,1,1,1,0,0}, + {"zrangebyscore",zrangebyscoreCommand,-4,"r",0,NULL,1,1,1,0,0}, + {"zrevrangebyscore",zrevrangebyscoreCommand,-4,"r",0,NULL,1,1,1,0,0}, + {"zcount",zcountCommand,4,"r",0,NULL,1,1,1,0,0}, + {"zrevrange",zrevrangeCommand,-4,"r",0,NULL,1,1,1,0,0}, + {"zcard",zcardCommand,2,"r",0,NULL,1,1,1,0,0}, + {"zscore",zscoreCommand,3,"r",0,NULL,1,1,1,0,0}, + {"zrank",zrankCommand,3,"r",0,NULL,1,1,1,0,0}, + {"zrevrank",zrevrankCommand,3,"r",0,NULL,1,1,1,0,0}, + {"hset",hsetCommand,4,"wm",0,NULL,1,1,1,0,0}, + {"hsetnx",hsetnxCommand,4,"wm",0,NULL,1,1,1,0,0}, + {"hget",hgetCommand,3,"r",0,NULL,1,1,1,0,0}, + {"hmset",hmsetCommand,-4,"wm",0,NULL,1,1,1,0,0}, + {"hmget",hmgetCommand,-3,"r",0,NULL,1,1,1,0,0}, + {"hincrby",hincrbyCommand,4,"wm",0,NULL,1,1,1,0,0}, + {"hincrbyfloat",hincrbyfloatCommand,4,"wm",0,NULL,1,1,1,0,0}, + {"hdel",hdelCommand,-3,"w",0,NULL,1,1,1,0,0}, + {"hlen",hlenCommand,2,"r",0,NULL,1,1,1,0,0}, + {"hkeys",hkeysCommand,2,"r",0,NULL,1,1,1,0,0}, + {"hvals",hvalsCommand,2,"r",0,NULL,1,1,1,0,0}, + {"hgetall",hgetallCommand,2,"r",0,NULL,1,1,1,0,0}, + {"hexists",hexistsCommand,3,"r",0,NULL,1,1,1,0,0}, + {"incrby",incrbyCommand,3,"wm",0,NULL,1,1,1,0,0}, + {"decrby",decrbyCommand,3,"wm",0,NULL,1,1,1,0,0}, + {"incrbyfloat",incrbyfloatCommand,3,"wm",0,NULL,1,1,1,0,0}, + {"getset",getsetCommand,3,"wm",0,NULL,1,1,1,0,0}, + {"mset",msetCommand,-3,"wm",0,NULL,1,-1,2,0,0}, + {"msetnx",msetnxCommand,-3,"wm",0,NULL,1,-1,2,0,0}, + {"randomkey",randomkeyCommand,1,"rR",0,NULL,0,0,0,0,0}, + {"select",selectCommand,2,"r",0,NULL,0,0,0,0,0}, + {"move",moveCommand,3,"w",0,NULL,1,1,1,0,0}, + {"rename",renameCommand,3,"w",0,renameGetKeys,1,2,1,0,0}, + {"renamenx",renamenxCommand,3,"w",0,renameGetKeys,1,2,1,0,0}, + {"expire",expireCommand,3,"w",0,NULL,1,1,1,0,0}, + {"expireat",expireatCommand,3,"w",0,NULL,1,1,1,0,0}, + {"pexpire",pexpireCommand,3,"w",0,NULL,1,1,1,0,0}, + {"pexpireat",pexpireatCommand,3,"w",0,NULL,1,1,1,0,0}, + {"keys",keysCommand,2,"r",0,NULL,0,0,0,0,0}, + {"dbsize",dbsizeCommand,1,"r",0,NULL,0,0,0,0,0}, + {"auth",authCommand,2,"rs",0,NULL,0,0,0,0,0}, + {"ping",pingCommand,1,"r",0,NULL,0,0,0,0,0}, + {"echo",echoCommand,2,"r",0,NULL,0,0,0,0,0}, + {"save",saveCommand,1,"ars",0,NULL,0,0,0,0,0}, + {"bgsave",bgsaveCommand,1,"ar",0,NULL,0,0,0,0,0}, + {"bgrewriteaof",bgrewriteaofCommand,1,"ar",0,NULL,0,0,0,0,0}, + {"shutdown",shutdownCommand,-1,"ar",0,NULL,0,0,0,0,0}, + {"lastsave",lastsaveCommand,1,"r",0,NULL,0,0,0,0,0}, + {"type",typeCommand,2,"r",0,NULL,1,1,1,0,0}, + {"multi",multiCommand,1,"rs",0,NULL,0,0,0,0,0}, + {"exec",execCommand,1,"wms",0,NULL,0,0,0,0,0}, + {"discard",discardCommand,1,"rs",0,NULL,0,0,0,0,0}, + {"sync",syncCommand,1,"ars",0,NULL,0,0,0,0,0}, + {"flushdb",flushdbCommand,1,"w",0,NULL,0,0,0,0,0}, + {"flushall",flushallCommand,1,"w",0,NULL,0,0,0,0,0}, + {"sort",sortCommand,-2,"wm",0,NULL,1,1,1,0,0}, + {"info",infoCommand,-1,"r",0,NULL,0,0,0,0,0}, + {"monitor",monitorCommand,1,"ars",0,NULL,0,0,0,0,0}, + {"ttl",ttlCommand,2,"r",0,NULL,1,1,1,0,0}, + {"pttl",pttlCommand,2,"r",0,NULL,1,1,1,0,0}, + {"persist",persistCommand,2,"w",0,NULL,1,1,1,0,0}, + {"slaveof",slaveofCommand,3,"aws",0,NULL,0,0,0,0,0}, + {"debug",debugCommand,-2,"aws",0,NULL,0,0,0,0,0}, + {"config",configCommand,-2,"ar",0,NULL,0,0,0,0,0}, + {"subscribe",subscribeCommand,-2,"rps",0,NULL,0,0,0,0,0}, + {"unsubscribe",unsubscribeCommand,-1,"rps",0,NULL,0,0,0,0,0}, + {"psubscribe",psubscribeCommand,-2,"rps",0,NULL,0,0,0,0,0}, + {"punsubscribe",punsubscribeCommand,-1,"rps",0,NULL,0,0,0,0,0}, + {"publish",publishCommand,3,"rpf",0,NULL,0,0,0,0,0}, + {"watch",watchCommand,-2,"rs",0,noPreloadGetKeys,1,-1,1,0,0}, + {"unwatch",unwatchCommand,1,"rs",0,NULL,0,0,0,0,0}, + {"cluster",clusterCommand,-2,"ar",0,NULL,0,0,0,0,0}, + {"restore",restoreCommand,4,"awm",0,NULL,1,1,1,0,0}, + {"migrate",migrateCommand,6,"aw",0,NULL,0,0,0,0,0}, + {"asking",askingCommand,1,"r",0,NULL,0,0,0,0,0}, + {"dump",dumpCommand,2,"ar",0,NULL,1,1,1,0,0}, + {"object",objectCommand,-2,"r",0,NULL,2,2,2,0,0}, + {"client",clientCommand,-2,"ar",0,NULL,0,0,0,0,0}, + {"eval",evalCommand,-3,"wms",0,zunionInterGetKeys,0,0,0,0,0}, + {"evalsha",evalShaCommand,-3,"wms",0,zunionInterGetKeys,0,0,0,0,0}, + {"slowlog",slowlogCommand,-2,"r",0,NULL,0,0,0,0,0}, + {"script",scriptCommand,-2,"ras",0,NULL,0,0,0,0,0} }; /*============================ Utility functions ============================ */ -void redisLog(int level, const char *fmt, ...) { - va_list ap; +/* Low level logging. To use only for very big messages, otherwise + * redisLog() is to prefer. */ +void redisLogRaw(int level, const char *msg) { + const int syslogLevelMap[] = { LOG_DEBUG, LOG_INFO, LOG_NOTICE, LOG_WARNING }; + const char *c = ".-*#"; + time_t now = time(NULL); FILE *fp; + char buf[64]; + int rawmode = (level & REDIS_LOG_RAW); + + level &= 0xff; /* clear flags */ + if (level < server.verbosity) return; fp = (server.logfile == NULL) ? stdout : fopen(server.logfile,"a"); if (!fp) return; - va_start(ap, fmt); - if (level >= server.verbosity) { - char *c = ".-*#"; - char buf[64]; - time_t now; - - now = time(NULL); - strftime(buf,64,"%d %b %H:%M:%S",localtime(&now)); - fprintf(fp,"[%d] %s %c ",(int)getpid(),buf,c[level]); - vfprintf(fp, fmt, ap); - fprintf(fp,"\n"); - fflush(fp); + if (rawmode) { + fprintf(fp,"%s",msg); + } else { + strftime(buf,sizeof(buf),"%d %b %H:%M:%S",localtime(&now)); + fprintf(fp,"[%d] %s %c %s\n",(int)getpid(),buf,c[level],msg); } - va_end(ap); + fflush(fp); if (server.logfile) fclose(fp); + + if (server.syslog_enabled) syslog(syslogLevelMap[level], "%s", msg); +} + +/* Like redisLogRaw() but with printf-alike support. This is the funciton that + * is used across the code. The raw version is only used in order to dump + * the INFO output on crash. */ +void redisLog(int level, const char *fmt, ...) { + va_list ap; + char msg[REDIS_MAX_LOGMSG_LEN]; + + if ((level&0xff) < server.verbosity) return; + + va_start(ap, fmt); + vsnprintf(msg, sizeof(msg), fmt, ap); + va_end(ap); + + redisLogRaw(level,msg); } /* Redis generally does not try to recover from out of memory conditions @@ -220,6 +300,22 @@ void oom(const char *msg) { abort(); } +/* 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 */ +long long mstime(void) { + return ustime()/1000; +} + /*====================== Hash table type implementation ==================== */ /* This is an hash table type that uses the SDS dynamic strings libary as @@ -250,6 +346,15 @@ int dictSdsKeyCompare(void *privdata, const void *key1, return memcmp(key1, key2, l1) == 0; } +/* A case insensitive version used for the command lookup table. */ +int dictSdsKeyCaseCompare(void *privdata, const void *key1, + const void *key2) +{ + DICT_NOTUSED(privdata); + + return strcasecmp(key1, key2) == 0; +} + void dictRedisObjectDestructor(void *privdata, void *val) { DICT_NOTUSED(privdata); @@ -281,6 +386,10 @@ unsigned int dictSdsHash(const void *key) { return dictGenHashFunction((unsigned char*)key, sdslen((char*)key)); } +unsigned int dictSdsCaseHash(const void *key) { + return dictGenCaseHashFunction((unsigned char*)key, sdslen((char*)key)); +} + int dictEncObjKeyCompare(void *privdata, const void *key1, const void *key2) { @@ -322,7 +431,7 @@ unsigned int dictEncObjHash(const void *key) { } } -/* Sets type */ +/* Sets type hash table */ dictType setDictType = { dictEncObjHash, /* hash function */ NULL, /* key dup */ @@ -339,7 +448,7 @@ dictType zsetDictType = { NULL, /* val dup */ dictEncObjKeyCompare, /* key compare */ dictRedisObjectDestructor, /* key destructor */ - dictVanillaFree /* val destructor of malloc(sizeof(double)) */ + NULL /* val destructor */ }; /* Db->dict, keys are sds strings, vals are Redis objects. */ @@ -362,6 +471,16 @@ dictType keyptrDictType = { 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 zimpaps) */ dictType hashDictType = { dictEncObjHash, /* hash function */ @@ -384,6 +503,17 @@ dictType keylistDictType = { 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 */ +}; + int htNeedsResize(dict *dict) { long long size, used; @@ -428,7 +558,7 @@ void incrementallyRehash(void) { * for dict.c to resize the hash tables accordingly to the fact we have o not * running childs. */ void updateDictResizePolicy(void) { - if (server.bgsavechildpid == -1 && server.bgrewritechildpid == -1) + if (server.rdb_child_pid == -1 && server.aof_child_pid == -1) dictEnableResize(); else dictDisableResize(); @@ -436,8 +566,54 @@ void updateDictResizePolicy(void) { /* ======================= Cron: called every 100 ms ======================== */ +/* Try to expire a few timed out keys. The algorithm used is adaptive and + * will use few CPU cycles if there are few expiring keys, otherwise + * it will get more aggressive to avoid that too much memory is used by + * keys that can be removed from the keyspace. */ +void activeExpireCycle(void) { + int j; + + for (j = 0; j < server.dbnum; j++) { + int expired; + redisDb *db = server.db+j; + + /* Continue to expire if at the end of the cycle more than 25% + * of the keys were expired. */ + do { + long num = dictSize(db->expires); + long long now = mstime(); + + expired = 0; + if (num > REDIS_EXPIRELOOKUPS_PER_CRON) + num = REDIS_EXPIRELOOKUPS_PER_CRON; + while (num--) { + dictEntry *de; + long long t; + + if ((de = dictGetRandomKey(db->expires)) == NULL) break; + t = dictGetSignedIntegerVal(de); + if (now > t) { + sds key = dictGetKey(de); + robj *keyobj = createStringObject(key,sdslen(key)); + + propagateExpire(db,keyobj); + dbDelete(db,keyobj); + decrRefCount(keyobj); + expired++; + server.stat_expiredkeys++; + } + } + } while (expired > REDIS_EXPIRELOOKUPS_PER_CRON/4); + } +} + +void updateLRUClock(void) { + server.lruclock = (time(NULL)/REDIS_LRU_CLOCK_RESOLUTION) & + REDIS_LRU_CLOCK_MAX; +} + int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) { - int j, loops = server.cronloops++; + int j, loops = server.cronloops; REDIS_NOTUSED(eventLoop); REDIS_NOTUSED(id); REDIS_NOTUSED(clientData); @@ -447,24 +623,29 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) { * in objects at every object access, and accuracy is not needed. * To access a global var is faster than calling time(NULL) */ server.unixtime = time(NULL); - /* We have just 21 bits per object for LRU information. - * So we use an (eventually wrapping) LRU clock with minutes resolution. + + /* We have just 22 bits per object for LRU information. + * So we use an (eventually wrapping) LRU clock with 10 seconds resolution. + * 2^22 bits with 10 seconds resoluton is more or less 1.5 years. * - * When we need to select what object to swap, we compute the minimum - * time distance between the current lruclock and the object last access - * lruclock info. Even if clocks will wrap on overflow, there is - * the interesting property that we are sure that at least - * ABS(A-B) minutes passed between current time and timestamp B. + * Note that even if this will wrap after 1.5 years it's not a problem, + * everything will still work but just some object will appear younger + * to Redis. But for this to happen a given object should never be touched + * for 1.5 years. * - * This is not precise but we don't need at all precision, but just - * something statistically reasonable. + * Note that you can change the resolution altering the + * REDIS_LRU_CLOCK_RESOLUTION define. */ - server.lruclock = (time(NULL)/60)&((1<<21)-1); + updateLRUClock(); + + /* 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(); /* 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() == REDIS_OK) exit(0); + if (prepareForShutdown(0) == REDIS_OK) exit(0); redisLog(REDIS_WARNING,"SIGTERM received but errors trying to shut down the server, check the logs for more information"); } @@ -487,7 +668,7 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) { * if we resize the HT while there is the saving child at work actually * a lot of memory movements in the parent will cause a lot of pages * copied. */ - if (server.bgsavechildpid == -1 && server.bgrewritechildpid == -1) { + if (server.rdb_child_pid == -1 && server.aof_child_pid == -1) { if (!(loops % 10)) tryResizeHashTables(); if (server.activerehashing) incrementallyRehash(); } @@ -501,26 +682,40 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) { } /* Close connections of timedout clients */ - if ((server.maxidletime && !(loops % 100)) || server.blpop_blocked_clients) + if ((server.maxidletime && !(loops % 100)) || server.bpop_blocked_clients) closeTimedoutClients(); - /* Check if a background saving or AOF rewrite in progress terminated */ - if (server.bgsavechildpid != -1 || server.bgrewritechildpid != -1) { + /* Start a scheduled AOF rewrite if this was requested by the user while + * a BGSAVE was in progress. */ + if (server.rdb_child_pid == -1 && server.aof_child_pid == -1 && + server.aof_rewrite_scheduled) + { + rewriteAppendOnlyFileBackground(); + } + + /* Check if a background saving or AOF rewrite in progress terminated. */ + if (server.rdb_child_pid != -1 || server.aof_child_pid != -1) { int statloc; pid_t pid; if ((pid = wait3(&statloc,WNOHANG,NULL)) != 0) { - if (pid == server.bgsavechildpid) { - backgroundSaveDoneHandler(statloc); + int exitcode = WEXITSTATUS(statloc); + int bysignal = 0; + + if (WIFSIGNALED(statloc)) bysignal = WTERMSIG(statloc); + + if (pid == server.rdb_child_pid) { + backgroundSaveDoneHandler(exitcode,bysignal); } else { - backgroundRewriteDoneHandler(statloc); + backgroundRewriteDoneHandler(exitcode,bysignal); } updateDictResizePolicy(); } } else { - /* If there is not a background saving in progress check if - * we have to save now */ time_t now = time(NULL); + + /* 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; @@ -528,82 +723,45 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) { now-server.lastsave > sp->seconds) { redisLog(REDIS_NOTICE,"%d changes in %d seconds. Saving...", sp->changes, sp->seconds); - rdbSaveBackground(server.dbfilename); + rdbSaveBackground(server.rdb_filename); break; } } + + /* Trigger an AOF rewrite if needed */ + if (server.rdb_child_pid == -1 && + server.aof_child_pid == -1 && + 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) { + redisLog(REDIS_NOTICE,"Starting automatic rewriting of AOF on %lld%% growth",growth); + rewriteAppendOnlyFileBackground(); + } + } } - /* Try to expire a few timed out keys. The algorithm used is adaptive and - * will use few CPU cycles if there are few expiring keys, otherwise - * it will get more aggressive to avoid that too much memory is used by - * keys that can be removed from the keyspace. */ - for (j = 0; j < server.dbnum; j++) { - int expired; - redisDb *db = server.db+j; - /* Continue to expire if at the end of the cycle more than 25% - * of the keys were expired. */ - do { - long num = dictSize(db->expires); - time_t now = time(NULL); + /* If we postponed an AOF buffer flush, let's try to do it every time the + * cron function is called. */ + if (server.aof_flush_postponed_start) flushAppendOnlyFile(0); - expired = 0; - if (num > REDIS_EXPIRELOOKUPS_PER_CRON) - num = REDIS_EXPIRELOOKUPS_PER_CRON; - while (num--) { - dictEntry *de; - time_t t; + /* Expire a few keys per cycle, only if this is a master. + * On slaves we wait for DEL operations synthesized by the master + * in order to guarantee a strict consistency. */ + if (server.masterhost == NULL) activeExpireCycle(); - if ((de = dictGetRandomKey(db->expires)) == NULL) break; - t = (time_t) dictGetEntryVal(de); - if (now > t) { - sds key = dictGetEntryKey(de); - robj *keyobj = createStringObject(key,sdslen(key)); - - dbDelete(db,keyobj); - decrRefCount(keyobj); - expired++; - server.stat_expiredkeys++; - } - } - } while (expired > REDIS_EXPIRELOOKUPS_PER_CRON/4); - } + /* Replication cron function -- used to reconnect to master and + * to detect transfer failures. */ + if (!(loops % 10)) replicationCron(); - /* Swap a few keys on disk if we are over the memory limit and VM - * is enbled. Try to free objects from the free list first. */ - if (vmCanSwapOut()) { - while (server.vm_enabled && zmalloc_used_memory() > - server.vm_max_memory) - { - int retval; - - if (tryFreeOneObjectFromFreelist() == REDIS_OK) continue; - retval = (server.vm_max_threads == 0) ? - vmSwapOneObjectBlocking() : - vmSwapOneObjectThreaded(); - if (retval == REDIS_ERR && !(loops % 300) && - zmalloc_used_memory() > - (server.vm_max_memory+server.vm_max_memory/10)) - { - redisLog(REDIS_WARNING,"WARNING: vm-max-memory limit exceeded by more than 10%% but unable to swap more objects out!"); - } - /* Note that when using threade I/O we free just one object, - * because anyway when the I/O thread in charge to swap this - * object out will finish, the handler of completed jobs - * will try to swap more objects if we are still out of memory. */ - if (retval == REDIS_ERR || server.vm_max_threads > 0) break; - } - } + /* Run other sub-systems specific cron jobs */ + if (server.cluster_enabled && !(loops % 10)) clusterCron(); - /* Check if we should connect to a MASTER */ - if (server.replstate == REDIS_REPL_CONNECT && !(loops % 10)) { - redisLog(REDIS_NOTICE,"Connecting to MASTER..."); - if (syncWithMaster() == REDIS_OK) { - redisLog(REDIS_NOTICE,"MASTER <-> SLAVE sync succeeded"); - if (server.appendonly) rewriteAppendOnlyFileBackground(); - } - } + server.cronloops++; return 100; } @@ -612,34 +770,27 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) { * for ready file descriptors. */ void beforeSleep(struct aeEventLoop *eventLoop) { REDIS_NOTUSED(eventLoop); - - /* Awake clients that got all the swapped keys they requested */ - if (server.vm_enabled && listLength(server.io_ready_clients)) { - listIter li; - listNode *ln; - - listRewind(server.io_ready_clients,&li); - while((ln = listNext(&li))) { - redisClient *c = ln->value; - struct redisCommand *cmd; - - /* Resume the client. */ - listDelNode(server.io_ready_clients,ln); - c->flags &= (~REDIS_IO_WAIT); - server.vm_blocked_clients--; - aeCreateFileEvent(server.el, c->fd, AE_READABLE, - readQueryFromClient, c); - cmd = lookupCommand(c->argv[0]->ptr); - redisAssert(cmd != NULL); - call(c,cmd); - resetClient(c); - /* There may be more data to process in the input buffer. */ - if (c->querybuf && sdslen(c->querybuf) > 0) - processInputBuffer(c); + listNode *ln; + redisClient *c; + + /* Try to process pending commands for clients that were just unblocked. */ + while (listLength(server.unblocked_clients)) { + ln = listFirst(server.unblocked_clients); + redisAssert(ln != NULL); + c = ln->value; + listDelNode(server.unblocked_clients,ln); + c->flags &= ~REDIS_UNBLOCKED; + + /* Process remaining data in the input buffer. */ + if (c->querybuf && sdslen(c->querybuf) > 0) { + server.current_client = c; + processInputBuffer(c); + server.current_client = NULL; } } + /* Write the AOF buffer on disk */ - flushAppendOnlyFile(); + flushAppendOnlyFile(0); } /* =========================== Server initialization ======================== */ @@ -669,6 +820,12 @@ void createSharedObjects(void) { "-ERR source and destination objects are the same\r\n")); shared.outofrangeerr = createObject(REDIS_STRING,sdsnew( "-ERR index out of range\r\n")); + shared.noscripterr = createObject(REDIS_STRING,sdsnew( + "-NOSCRIPT No matching script. Please use EVAL.\r\n")); + shared.loadingerr = createObject(REDIS_STRING,sdsnew( + "-LOADING Redis is loading the dataset in memory\r\n")); + shared.slowscripterr = createObject(REDIS_STRING,sdsnew( + "-BUSY Redis is busy running a script. You can only call SCRIPT KILL or SHUTDOWN NOSAVE.\r\n")); shared.space = createObject(REDIS_STRING,sdsnew(" ")); shared.colon = createObject(REDIS_STRING,sdsnew(":")); shared.plus = createObject(REDIS_STRING,sdsnew("+")); @@ -697,61 +854,138 @@ void createSharedObjects(void) { } void initServerConfig() { - server.dbnum = REDIS_DEFAULT_DBNUM; server.port = REDIS_SERVERPORT; - server.verbosity = REDIS_VERBOSE; + server.bindaddr = NULL; + server.unixsocket = NULL; + server.unixsocketperm = 0; + server.ipfd = -1; + server.sofd = -1; + server.dbnum = REDIS_DEFAULT_DBNUM; + server.verbosity = REDIS_NOTICE; server.maxidletime = REDIS_MAXIDLETIME; + server.client_max_querybuf_len = REDIS_MAX_QUERYBUF_LEN; server.saveparams = NULL; + server.loading = 0; server.logfile = NULL; /* NULL = log on standard output */ - server.bindaddr = NULL; - server.glueoutputbuf = 1; + server.syslog_enabled = 0; + server.syslog_ident = zstrdup("redis"); + server.syslog_facility = LOG_LOCAL0; server.daemonize = 0; - server.appendonly = 0; - server.appendfsync = APPENDFSYNC_EVERYSEC; - server.no_appendfsync_on_rewrite = 0; - server.lastfsync = time(NULL); - server.appendfd = -1; - server.appendseldb = -1; /* Make sure the first time will not match */ + server.aof_state = REDIS_AOF_OFF; + server.aof_fsync = AOF_FSYNC_EVERYSEC; + server.aof_no_fsync_on_rewrite = 0; + server.aof_rewrite_perc = REDIS_AOF_REWRITE_PERC; + server.aof_rewrite_min_size = REDIS_AOF_REWRITE_MIN_SIZE; + server.aof_rewrite_base_size = 0; + server.aof_rewrite_scheduled = 0; + server.aof_last_fsync = time(NULL); + 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 = zstrdup("/var/run/redis.pid"); - server.dbfilename = zstrdup("dump.rdb"); - server.appendfilename = zstrdup("appendonly.aof"); + server.rdb_filename = zstrdup("dump.rdb"); + server.aof_filename = zstrdup("appendonly.aof"); server.requirepass = NULL; - server.rdbcompression = 1; + server.rdb_compression = 1; server.activerehashing = 1; - server.maxclients = 0; - server.blpop_blocked_clients = 0; + server.maxclients = REDIS_MAX_CLIENTS; + server.bpop_blocked_clients = 0; server.maxmemory = 0; - server.vm_enabled = 0; - server.vm_swap_file = zstrdup("/tmp/redis-%p.vm"); - server.vm_page_size = 256; /* 256 bytes per page */ - server.vm_pages = 1024*1024*100; /* 104 millions of pages */ - server.vm_max_memory = 1024LL*1024*1024*1; /* 1 GB of RAM */ - server.vm_max_threads = 4; - server.vm_blocked_clients = 0; + server.maxmemory_policy = REDIS_MAXMEMORY_VOLATILE_LRU; + server.maxmemory_samples = 3; server.hash_max_zipmap_entries = REDIS_HASH_MAX_ZIPMAP_ENTRIES; server.hash_max_zipmap_value = REDIS_HASH_MAX_ZIPMAP_VALUE; server.list_max_ziplist_entries = REDIS_LIST_MAX_ZIPLIST_ENTRIES; server.list_max_ziplist_value = REDIS_LIST_MAX_ZIPLIST_VALUE; + server.set_max_intset_entries = REDIS_SET_MAX_INTSET_ENTRIES; + server.zset_max_ziplist_entries = REDIS_ZSET_MAX_ZIPLIST_ENTRIES; + server.zset_max_ziplist_value = REDIS_ZSET_MAX_ZIPLIST_VALUE; server.shutdown_asap = 0; - + server.repl_ping_slave_period = REDIS_REPL_PING_SLAVE_PERIOD; + server.repl_timeout = REDIS_REPL_TIMEOUT; + server.cluster_enabled = 0; + server.cluster.configfile = zstrdup("nodes.conf"); + server.lua_caller = NULL; + server.lua_time_limit = REDIS_LUA_TIME_LIMIT; + server.lua_client = NULL; + server.lua_timedout = 0; + + updateLRUClock(); 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.isslave = 0; server.masterauth = NULL; server.masterhost = NULL; server.masterport = 6379; server.master = NULL; - server.replstate = REDIS_REPL_NONE; + server.repl_state = REDIS_REPL_NONE; + server.repl_syncio_timeout = REDIS_REPL_SYNCIO_TIMEOUT; + server.repl_serve_stale_data = 1; + server.repl_down_since = -1; /* 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 intiialize 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); + populateCommandTable(); + server.delCommand = lookupCommandByCString("del"); + server.multiCommand = lookupCommandByCString("multi"); + + /* Slow log */ + server.slowlog_log_slower_than = REDIS_SLOWLOG_LOG_SLOWER_THAN; + server.slowlog_max_len = REDIS_SLOWLOG_MAX_LEN; + + /* Assert */ + server.assert_failed = ""; + server.assert_file = ""; + server.assert_line = 0; + server.bug_report_start = 0; +} + +/* This function will try to raise the max number of open files accordingly to + * the configured max number of clients. It will also account for 32 additional + * file descriptors as we need a few more for 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+32; + struct rlimit limit; + + if (maxfiles < 1024) maxfiles = 1024; + if (getrlimit(RLIMIT_NOFILE,&limit) == -1) { + redisLog(REDIS_WARNING,"Unable to obtain the current NOFILE limit (%s), assuming 1024 and setting the max clients configuration accordingly.", + strerror(errno)); + server.maxclients = 1024-32; + } 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) { + limit.rlim_cur = maxfiles; + limit.rlim_max = maxfiles; + if (setrlimit(RLIMIT_NOFILE,&limit) == -1) { + server.maxclients = oldlimit-32; + redisLog(REDIS_WARNING,"Unable to set the max number of files limit to %d (%s), setting the max clients configuration to %d.", + (int) maxfiles, strerror(errno), (int) server.maxclients); + } else { + redisLog(REDIS_NOTICE,"Max number of open files set to %d", + (int) maxfiles); + } + } + } } void initServer() { @@ -759,23 +993,42 @@ void initServer() { signal(SIGHUP, SIG_IGN); signal(SIGPIPE, SIG_IGN); - setupSigSegvAction(); + setupSignalHandlers(); - server.devnull = fopen("/dev/null","w"); - if (server.devnull == NULL) { - redisLog(REDIS_WARNING, "Can't open /dev/null: %s", server.neterr); - exit(1); + if (server.syslog_enabled) { + openlog(server.syslog_ident, LOG_PID | LOG_NDELAY | LOG_NOWAIT, + server.syslog_facility); } + + server.current_client = NULL; server.clients = listCreate(); server.slaves = listCreate(); server.monitors = listCreate(); - server.objfreelist = listCreate(); + server.unblocked_clients = listCreate(); + createSharedObjects(); - server.el = aeCreateEventLoop(); + adjustOpenFilesLimit(); + server.el = aeCreateEventLoop(server.maxclients+1024); server.db = zmalloc(sizeof(redisDb)*server.dbnum); - server.fd = anetTcpServer(server.neterr, server.port, server.bindaddr); - if (server.fd == -1) { - redisLog(REDIS_WARNING, "Opening TCP port: %s", server.neterr); + + if (server.port != 0) { + server.ipfd = anetTcpServer(server.neterr,server.port,server.bindaddr); + if (server.ipfd == ANET_ERR) { + redisLog(REDIS_WARNING, "Opening port %d: %s", + server.port, server.neterr); + exit(1); + } + } + if (server.unixsocket != NULL) { + unlink(server.unixsocket); /* don't care if this fails */ + server.sofd = anetUnixServer(server.neterr,server.unixsocket,server.unixsocketperm); + if (server.sofd == ANET_ERR) { + redisLog(REDIS_WARNING, "Opening socket: %s", server.neterr); + exit(1); + } + } + if (server.ipfd < 0 && server.sofd < 0) { + redisLog(REDIS_WARNING, "Configured to not listen anywhere, exiting."); exit(1); } for (j = 0; j < server.dbnum; j++) { @@ -783,8 +1036,6 @@ void initServer() { server.db[j].expires = dictCreate(&keyptrDictType,NULL); server.db[j].blocking_keys = dictCreate(&keylistDictType,NULL); server.db[j].watched_keys = dictCreate(&keylistDictType,NULL); - if (server.vm_enabled) - server.db[j].io_keys = dictCreate(&keylistDictType,NULL); server.db[j].id = j; } server.pubsub_channels = dictCreate(&keylistDictType,NULL); @@ -792,71 +1043,120 @@ void initServer() { listSetFreeMethod(server.pubsub_patterns,freePubsubPattern); listSetMatchMethod(server.pubsub_patterns,listMatchPubsubPattern); server.cronloops = 0; - server.bgsavechildpid = -1; - server.bgrewritechildpid = -1; - server.bgrewritebuf = sdsempty(); - server.aofbuf = sdsempty(); + server.rdb_child_pid = -1; + server.aof_child_pid = -1; + server.aof_rewrite_buf = sdsempty(); + server.aof_buf = sdsempty(); server.lastsave = time(NULL); server.dirty = 0; server.stat_numcommands = 0; server.stat_numconnections = 0; server.stat_expiredkeys = 0; + server.stat_evictedkeys = 0; server.stat_starttime = time(NULL); + server.stat_keyspace_misses = 0; + server.stat_keyspace_hits = 0; + server.stat_peak_memory = 0; + server.stat_fork_time = 0; + server.stat_rejected_conn = 0; server.unixtime = time(NULL); aeCreateTimeEvent(server.el, 1, serverCron, NULL, NULL); - if (aeCreateFileEvent(server.el, server.fd, AE_READABLE, - acceptHandler, NULL) == AE_ERR) oom("creating file event"); - - if (server.appendonly) { - server.appendfd = open(server.appendfilename,O_WRONLY|O_APPEND|O_CREAT,0644); - if (server.appendfd == -1) { + if (server.ipfd > 0 && aeCreateFileEvent(server.el,server.ipfd,AE_READABLE, + acceptTcpHandler,NULL) == AE_ERR) oom("creating file event"); + if (server.sofd > 0 && aeCreateFileEvent(server.el,server.sofd,AE_READABLE, + acceptUnixHandler,NULL) == AE_ERR) oom("creating file event"); + + if (server.aof_state == REDIS_AOF_ON) { + server.aof_fd = open(server.aof_filename, + O_WRONLY|O_APPEND|O_CREAT,0644); + if (server.aof_fd == -1) { redisLog(REDIS_WARNING, "Can't open the append-only file: %s", strerror(errno)); exit(1); } } - if (server.vm_enabled) vmInit(); + if (server.cluster_enabled) clusterInit(); + scriptingInit(); + slowlogInit(); + bioInit(); + srand(time(NULL)^getpid()); + } -int qsortRedisCommands(const void *r1, const void *r2) { - return strcasecmp( - ((struct redisCommand*)r1)->name, - ((struct redisCommand*)r2)->name); +/* 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; + char *f = c->sflags; + int retval; + + while(*f != '\0') { + switch(*f) { + case 'w': c->flags |= REDIS_CMD_WRITE; break; + case 'r': c->flags |= REDIS_CMD_READONLY; break; + case 'm': c->flags |= REDIS_CMD_DENYOOM; break; + case 'a': c->flags |= REDIS_CMD_ADMIN; break; + case 'p': c->flags |= REDIS_CMD_PUBSUB; break; + case 'f': c->flags |= REDIS_CMD_FORCE_REPLICATION; break; + case 's': c->flags |= REDIS_CMD_NOSCRIPT; break; + case 'R': c->flags |= REDIS_CMD_RANDOM; break; + default: redisPanic("Unsupported command flag"); break; + } + f++; + } + + retval = dictAdd(server.commands, sdsnew(c->name), c); + assert(retval == DICT_OK); + } } -void sortCommandTable() { - /* Copy and sort the read-only version of the command table */ - commandTable = (struct redisCommand*)malloc(sizeof(readonlyCommandTable)); - memcpy(commandTable,readonlyCommandTable,sizeof(readonlyCommandTable)); - qsort(commandTable, - sizeof(readonlyCommandTable)/sizeof(struct redisCommand), - sizeof(struct redisCommand),qsortRedisCommands); +void resetCommandTableStats(void) { + int numcommands = sizeof(redisCommandTable)/sizeof(struct redisCommand); + int j; + + for (j = 0; j < numcommands; j++) { + struct redisCommand *c = redisCommandTable+j; + + c->microseconds = 0; + c->calls = 0; + } } /* ====================== Commands lookup and execution ===================== */ -struct redisCommand *lookupCommand(char *name) { - struct redisCommand tmp = {name,NULL,0,0,NULL,0,0,0}; - return bsearch( - &tmp, - commandTable, - sizeof(readonlyCommandTable)/sizeof(struct redisCommand), - sizeof(struct redisCommand), - qsortRedisCommands); +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; } /* Call() is the core of Redis execution of a command */ -void call(redisClient *c, struct redisCommand *cmd) { - long long dirty; +void call(redisClient *c) { + long long dirty, start = ustime(), duration; dirty = server.dirty; - cmd->proc(c); + c->cmd->proc(c); dirty = server.dirty-dirty; - - if (server.appendonly && dirty) - feedAppendOnlyFile(cmd,c->db->id,c->argv,c->argc); - if ((dirty || cmd->flags & REDIS_CMD_FORCE_REPLICATION) && + duration = ustime()-start; + c->cmd->microseconds += duration; + slowlogPushEntryIfNeeded(c->argv,c->argc,duration); + c->cmd->calls++; + + if (server.aof_state != REDIS_AOF_OFF && dirty > 0) + feedAppendOnlyFile(c->cmd,c->db->id,c->argv,c->argc); + if ((dirty > 0 || c->cmd->flags & REDIS_CMD_FORCE_REPLICATION) && listLength(server.slaves)) replicationFeedSlaves(server.slaves,c->db->id,c->argv,c->argc); if (listLength(server.monitors)) @@ -873,224 +1173,194 @@ void call(redisClient *c, struct redisCommand *cmd) { * and other operations can be performed by the caller. Otherwise * if 0 is returned the client was destroied (i.e. after QUIT). */ int processCommand(redisClient *c) { - struct redisCommand *cmd; - - /* Free some memory if needed (maxmemory setting) */ - if (server.maxmemory) freeMemoryIfNeeded(); - - /* Handle the multi bulk command type. This is an alternative protocol - * supported by Redis in order to receive commands that are composed of - * multiple binary-safe "bulk" arguments. The latency of processing is - * a bit higher but this allows things like multi-sets, so if this - * protocol is used only for MSET and similar commands this is a big win. */ - if (c->multibulk == 0 && c->argc == 1 && ((char*)(c->argv[0]->ptr))[0] == '*') { - c->multibulk = atoi(((char*)c->argv[0]->ptr)+1); - if (c->multibulk <= 0) { - resetClient(c); - return 1; - } else { - decrRefCount(c->argv[c->argc-1]); - c->argc--; - return 1; - } - } else if (c->multibulk) { - if (c->bulklen == -1) { - if (((char*)c->argv[0]->ptr)[0] != '$') { - addReplySds(c,sdsnew("-ERR multi bulk protocol error\r\n")); - resetClient(c); - return 1; - } else { - int bulklen = atoi(((char*)c->argv[0]->ptr)+1); - decrRefCount(c->argv[0]); - if (bulklen < 0 || bulklen > 1024*1024*1024) { - c->argc--; - addReplySds(c,sdsnew("-ERR invalid bulk write count\r\n")); - resetClient(c); - return 1; - } - c->argc--; - c->bulklen = bulklen+2; /* add two bytes for CR+LF */ - return 1; - } - } else { - c->mbargv = zrealloc(c->mbargv,(sizeof(robj*))*(c->mbargc+1)); - c->mbargv[c->mbargc] = c->argv[0]; - c->mbargc++; - c->argc--; - c->multibulk--; - if (c->multibulk == 0) { - robj **auxargv; - int auxargc; - - /* Here we need to swap the multi-bulk argc/argv with the - * normal argc/argv of the client structure. */ - auxargv = c->argv; - c->argv = c->mbargv; - c->mbargv = auxargv; - - auxargc = c->argc; - c->argc = c->mbargc; - c->mbargc = auxargc; - - /* We need to set bulklen to something different than -1 - * in order for the code below to process the command without - * to try to read the last argument of a bulk command as - * a special argument. */ - c->bulklen = 0; - /* continue below and process the command */ - } else { - c->bulklen = -1; - return 1; - } - } - } - /* -- end of multi bulk commands processing -- */ - - /* The QUIT command is handled as a special case. Normal command - * procs are unable to close the client connection safely */ + /* 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")) { - freeClient(c); - return 0; + addReply(c,shared.ok); + c->flags |= REDIS_CLOSE_AFTER_REPLY; + return REDIS_ERR; } /* Now lookup the command and check ASAP about trivial error conditions - * such wrong arity, bad command name and so forth. */ - cmd = lookupCommand(c->argv[0]->ptr); - if (!cmd) { - addReplySds(c, - sdscatprintf(sdsempty(), "-ERR unknown command '%s'\r\n", - (char*)c->argv[0]->ptr)); - resetClient(c); - return 1; - } else if ((cmd->arity > 0 && cmd->arity != c->argc) || - (c->argc < -cmd->arity)) { - addReplySds(c, - sdscatprintf(sdsempty(), - "-ERR wrong number of arguments for '%s' command\r\n", - cmd->name)); - resetClient(c); - return 1; - } else if (cmd->flags & REDIS_CMD_BULK && c->bulklen == -1) { - /* This is a bulk command, we have to read the last argument yet. */ - int bulklen = atoi(c->argv[c->argc-1]->ptr); - - decrRefCount(c->argv[c->argc-1]); - if (bulklen < 0 || bulklen > 1024*1024*1024) { - c->argc--; - addReplySds(c,sdsnew("-ERR invalid bulk write count\r\n")); - resetClient(c); - return 1; - } - c->argc--; - c->bulklen = bulklen+2; /* add two bytes for CR+LF */ - /* It is possible that the bulk read is already in the - * buffer. Check this condition and handle it accordingly. - * This is just a fast path, alternative to call processInputBuffer(). - * It's a good idea since the code is small and this condition - * happens most of the times. */ - if ((signed)sdslen(c->querybuf) >= c->bulklen) { - c->argv[c->argc] = createStringObject(c->querybuf,c->bulklen-2); - c->argc++; - c->querybuf = sdsrange(c->querybuf,c->bulklen,-1); - } else { - /* Otherwise return... there is to read the last argument - * from the socket. */ - return 1; - } + * such as wrong arity, bad command name and so forth. */ + c->cmd = c->lastcmd = lookupCommand(c->argv[0]->ptr); + if (!c->cmd) { + addReplyErrorFormat(c,"unknown command '%s'", + (char*)c->argv[0]->ptr); + return REDIS_OK; + } else if ((c->cmd->arity > 0 && c->cmd->arity != c->argc) || + (c->argc < -c->cmd->arity)) { + addReplyErrorFormat(c,"wrong number of arguments for '%s' command", + c->cmd->name); + return REDIS_OK; } - /* Let's try to encode the bulk object to save space. */ - if (cmd->flags & REDIS_CMD_BULK) - c->argv[c->argc-1] = tryObjectEncoding(c->argv[c->argc-1]); /* Check if the user is authenticated */ - if (server.requirepass && !c->authenticated && cmd->proc != authCommand) { - addReplySds(c,sdsnew("-ERR operation not permitted\r\n")); - resetClient(c); - return 1; + if (server.requirepass && !c->authenticated && c->cmd->proc != authCommand) + { + addReplyError(c,"operation not permitted"); + return REDIS_OK; } - /* Handle the maxmemory directive */ - if (server.maxmemory && (cmd->flags & REDIS_CMD_DENYOOM) && + /* If cluster is enabled, redirect here */ + if (server.cluster_enabled && + !(c->cmd->getkeys_proc == NULL && c->cmd->firstkey == 0)) { + int hashslot; + + if (server.cluster.state != REDIS_CLUSTER_OK) { + addReplyError(c,"The cluster is down. Check with CLUSTER INFO for more information"); + return REDIS_OK; + } else { + int ask; + clusterNode *n = getNodeByQuery(c,c->cmd,c->argv,c->argc,&hashslot,&ask); + if (n == NULL) { + addReplyError(c,"Multi keys request invalid in cluster"); + return REDIS_OK; + } else if (n != server.cluster.myself) { + addReplySds(c,sdscatprintf(sdsempty(), + "-%s %d %s:%d\r\n", ask ? "ASK" : "MOVED", + hashslot,n->ip,n->port)); + return REDIS_OK; + } + } + } + + /* Handle the maxmemory directive. + * + * First we try to free some memory if possible (if there are volatile + * keys in the dataset). If there are not the only thing we can do + * is returning an error. */ + if (server.maxmemory) freeMemoryIfNeeded(); + if (server.maxmemory && (c->cmd->flags & REDIS_CMD_DENYOOM) && zmalloc_used_memory() > server.maxmemory) { - addReplySds(c,sdsnew("-ERR command not allowed when used memory > 'maxmemory'\r\n")); - resetClient(c); - return 1; + addReplyError(c,"command not allowed when used memory > 'maxmemory'"); + return REDIS_OK; } /* Only allow SUBSCRIBE and UNSUBSCRIBE in the context of Pub/Sub */ if ((dictSize(c->pubsub_channels) > 0 || listLength(c->pubsub_patterns) > 0) && - cmd->proc != subscribeCommand && cmd->proc != unsubscribeCommand && - cmd->proc != psubscribeCommand && cmd->proc != punsubscribeCommand) { - addReplySds(c,sdsnew("-ERR only (P)SUBSCRIBE / (P)UNSUBSCRIBE / QUIT allowed in this context\r\n")); - resetClient(c); - return 1; + c->cmd->proc != subscribeCommand && + c->cmd->proc != unsubscribeCommand && + c->cmd->proc != psubscribeCommand && + c->cmd->proc != punsubscribeCommand) { + addReplyError(c,"only (P)SUBSCRIBE / (P)UNSUBSCRIBE / QUIT allowed in this context"); + return REDIS_OK; + } + + /* Only allow INFO and SLAVEOF when slave-serve-stale-data is no and + * we are a slave with a broken link with master. */ + if (server.masterhost && server.repl_state != REDIS_REPL_CONNECTED && + server.repl_serve_stale_data == 0 && + c->cmd->proc != infoCommand && c->cmd->proc != slaveofCommand) + { + addReplyError(c, + "link with MASTER is down and slave-serve-stale-data is set to no"); + return REDIS_OK; + } + + /* Loading DB? Return an error if the command is not INFO */ + if (server.loading && c->cmd->proc != infoCommand) { + addReply(c, shared.loadingerr); + return REDIS_OK; + } + + /* Lua script too slow? Only allow SHUTDOWN NOSAVE and SCRIPT KILL. */ + if (server.lua_timedout && + !(c->cmd->proc != shutdownCommand && + c->argc == 2 && + tolower(((char*)c->argv[1]->ptr)[0]) == 'n') && + !(c->cmd->proc == scriptCommand && + c->argc == 2 && + tolower(((char*)c->argv[1]->ptr)[0]) == 'k')) + { + addReply(c, shared.slowscripterr); + return REDIS_OK; } /* Exec the command */ if (c->flags & REDIS_MULTI && - cmd->proc != execCommand && cmd->proc != discardCommand && - cmd->proc != multiCommand && cmd->proc != watchCommand) + c->cmd->proc != execCommand && c->cmd->proc != discardCommand && + c->cmd->proc != multiCommand && c->cmd->proc != watchCommand) { - queueMultiCommand(c,cmd); + queueMultiCommand(c); addReply(c,shared.queued); } else { - if (server.vm_enabled && server.vm_max_threads > 0 && - blockClientOnSwappedKeys(c,cmd)) return 1; - call(c,cmd); + call(c); } - - /* Prepare the client for the next command */ - resetClient(c); - return 1; + return REDIS_OK; } /*================================== Shutdown =============================== */ -int prepareForShutdown() { - redisLog(REDIS_WARNING,"User requested shutdown, saving DB..."); +int prepareForShutdown(int flags) { + int save = flags & REDIS_SHUTDOWN_SAVE; + int nosave = flags & REDIS_SHUTDOWN_NOSAVE; + + redisLog(REDIS_WARNING,"User requested shutdown..."); /* Kill the saving child if there is a background saving in progress. We want to avoid race conditions, for instance our saving child may overwrite the synchronous saving did by SHUTDOWN. */ - if (server.bgsavechildpid != -1) { - redisLog(REDIS_WARNING,"There is a live saving child. Killing it!"); - kill(server.bgsavechildpid,SIGKILL); - rdbRemoveTempFile(server.bgsavechildpid); + if (server.rdb_child_pid != -1) { + redisLog(REDIS_WARNING,"There is a child saving an .rdb. Killing it!"); + kill(server.rdb_child_pid,SIGKILL); + rdbRemoveTempFile(server.rdb_child_pid); } - if (server.appendonly) { + if (server.aof_state != REDIS_AOF_OFF) { + /* Kill the AOF saving child as the AOF we already have may be longer + * but contains the full dataset anyway. */ + if (server.aof_child_pid != -1) { + redisLog(REDIS_WARNING, + "There is a child rewriting the AOF. Killing it!"); + kill(server.aof_child_pid,SIGKILL); + } /* Append only file: fsync() the AOF and exit */ - aof_fsync(server.appendfd); - if (server.vm_enabled) unlink(server.vm_swap_file); - } else { + redisLog(REDIS_NOTICE,"Calling fsync() on the AOF file."); + aof_fsync(server.aof_fd); + } + if ((server.saveparamslen > 0 && !nosave) || save) { + redisLog(REDIS_NOTICE,"Saving the final RDB snapshot before exiting."); /* Snapshotting. Perform a SYNC SAVE and exit */ - if (rdbSave(server.dbfilename) == REDIS_OK) { - if (server.daemonize) - unlink(server.pidfile); - redisLog(REDIS_WARNING,"%zu bytes used at exit",zmalloc_used_memory()); - } else { + if (rdbSave(server.rdb_filename) != REDIS_OK) { /* Ooops.. error saving! The best we can do is to continue * operating. Note that if there was a background saving process, * in the next cron() Redis will be notified that the background * saving aborted, handling special stuff like slaves pending for * synchronization... */ - redisLog(REDIS_WARNING,"Error trying to save the DB, can't exit"); + redisLog(REDIS_WARNING,"Error trying to save the DB, can't exit."); return REDIS_ERR; } } - redisLog(REDIS_WARNING,"Server exit now, bye bye..."); + if (server.daemonize) { + redisLog(REDIS_NOTICE,"Removing the pid file."); + unlink(server.pidfile); + } + /* Close the listening sockets. Apparently this allows faster restarts. */ + if (server.ipfd != -1) close(server.ipfd); + if (server.sofd != -1) close(server.sofd); + if (server.unixsocket) { + redisLog(REDIS_NOTICE,"Removing the unix socket file."); + unlink(server.unixsocket); /* don't care if this fails */ + } + + redisLog(REDIS_WARNING,"Redis is now ready to exit, bye bye..."); return REDIS_OK; } /*================================== Commands =============================== */ void authCommand(redisClient *c) { - if (!server.requirepass || !strcmp(c->argv[1]->ptr, server.requirepass)) { + if (!server.requirepass) { + addReplyError(c,"Client sent AUTH, but no password is set"); + } else if (!strcmp(c->argv[1]->ptr, server.requirepass)) { c->authenticated = 1; addReply(c,shared.ok); } else { c->authenticated = 0; - addReplySds(c,sdscatprintf(sdsempty(),"-ERR invalid password\r\n")); + addReplyError(c,"invalid password"); } } @@ -1126,125 +1396,331 @@ void bytesToHuman(char *s, unsigned long long n) { /* Create the string returned by the INFO command. This is decoupled * by the INFO command itself as we need to report the same information * on memory corruption problems. */ -sds genRedisInfoString(void) { - sds info; +sds genRedisInfoString(char *section) { + sds info = sdsempty(); time_t uptime = time(NULL)-server.stat_starttime; - int j; - char hmem[64]; - - bytesToHuman(hmem,zmalloc_used_memory()); - info = sdscatprintf(sdsempty(), - "redis_version:%s\r\n" - "redis_git_sha1:%s\r\n" - "redis_git_dirty:%d\r\n" - "arch_bits:%s\r\n" - "multiplexing_api:%s\r\n" - "process_id:%ld\r\n" - "uptime_in_seconds:%ld\r\n" - "uptime_in_days:%ld\r\n" - "connected_clients:%d\r\n" - "connected_slaves:%d\r\n" - "blocked_clients:%d\r\n" - "used_memory:%zu\r\n" - "used_memory_human:%s\r\n" - "changes_since_last_save:%lld\r\n" - "bgsave_in_progress:%d\r\n" - "last_save_time:%ld\r\n" - "bgrewriteaof_in_progress:%d\r\n" - "total_connections_received:%lld\r\n" - "total_commands_processed:%lld\r\n" - "expired_keys:%lld\r\n" - "hash_max_zipmap_entries:%zu\r\n" - "hash_max_zipmap_value:%zu\r\n" - "pubsub_channels:%ld\r\n" - "pubsub_patterns:%u\r\n" - "vm_enabled:%d\r\n" - "role:%s\r\n" - ,REDIS_VERSION, - redisGitSHA1(), - strtol(redisGitDirty(),NULL,10) > 0, - (sizeof(long) == 8) ? "64" : "32", - aeGetApiName(), - (long) getpid(), - uptime, - uptime/(3600*24), - listLength(server.clients)-listLength(server.slaves), - listLength(server.slaves), - server.blpop_blocked_clients, - zmalloc_used_memory(), - hmem, - server.dirty, - server.bgsavechildpid != -1, - server.lastsave, - server.bgrewritechildpid != -1, - server.stat_numconnections, - server.stat_numcommands, - server.stat_expiredkeys, - server.hash_max_zipmap_entries, - server.hash_max_zipmap_value, - dictSize(server.pubsub_channels), - listLength(server.pubsub_patterns), - server.vm_enabled != 0, - server.masterhost == NULL ? "master" : "slave" - ); - if (server.masterhost) { + int j, numcommands; + struct rusage self_ru, c_ru; + unsigned long lol, bib; + int allsections = 0, defsections = 0; + int sections = 0; + + if (section) { + allsections = strcasecmp(section,"all") == 0; + defsections = strcasecmp(section,"default") == 0; + } + + getrusage(RUSAGE_SELF, &self_ru); + getrusage(RUSAGE_CHILDREN, &c_ru); + getClientsMaxBuffers(&lol,&bib); + + /* Server */ + if (allsections || defsections || !strcasecmp(section,"server")) { + if (sections++) info = sdscat(info,"\r\n"); info = sdscatprintf(info, - "master_host:%s\r\n" - "master_port:%d\r\n" - "master_link_status:%s\r\n" - "master_last_io_seconds_ago:%d\r\n" - ,server.masterhost, - server.masterport, - (server.replstate == REDIS_REPL_CONNECTED) ? - "up" : "down", - server.master ? ((int)(time(NULL)-server.master->lastinteraction)) : -1 - ); + "# Server\r\n" + "redis_version:%s\r\n" + "redis_git_sha1:%s\r\n" + "redis_git_dirty:%d\r\n" + "arch_bits:%s\r\n" + "multiplexing_api:%s\r\n" + "gcc_version:%d.%d.%d\r\n" + "process_id:%ld\r\n" + "tcp_port:%d\r\n" + "uptime_in_seconds:%ld\r\n" + "uptime_in_days:%ld\r\n" + "lru_clock:%ld\r\n", + REDIS_VERSION, + redisGitSHA1(), + strtol(redisGitDirty(),NULL,10) > 0, + (sizeof(long) == 8) ? "64" : "32", + aeGetApiName(), +#ifdef __GNUC__ + __GNUC__,__GNUC_MINOR__,__GNUC_PATCHLEVEL__, +#else + 0,0,0, +#endif + (long) getpid(), + server.port, + uptime, + uptime/(3600*24), + (unsigned long) server.lruclock); } - if (server.vm_enabled) { - lockThreadedIO(); + + /* Clients */ + if (allsections || defsections || !strcasecmp(section,"clients")) { + if (sections++) info = sdscat(info,"\r\n"); info = sdscatprintf(info, - "vm_conf_max_memory:%llu\r\n" - "vm_conf_page_size:%llu\r\n" - "vm_conf_pages:%llu\r\n" - "vm_stats_used_pages:%llu\r\n" - "vm_stats_swapped_objects:%llu\r\n" - "vm_stats_swappin_count:%llu\r\n" - "vm_stats_swappout_count:%llu\r\n" - "vm_stats_io_newjobs_len:%lu\r\n" - "vm_stats_io_processing_len:%lu\r\n" - "vm_stats_io_processed_len:%lu\r\n" - "vm_stats_io_active_threads:%lu\r\n" - "vm_stats_blocked_clients:%lu\r\n" - ,(unsigned long long) server.vm_max_memory, - (unsigned long long) server.vm_page_size, - (unsigned long long) server.vm_pages, - (unsigned long long) server.vm_stats_used_pages, - (unsigned long long) server.vm_stats_swapped_objects, - (unsigned long long) server.vm_stats_swapins, - (unsigned long long) server.vm_stats_swapouts, - (unsigned long) listLength(server.io_newjobs), - (unsigned long) listLength(server.io_processing), - (unsigned long) listLength(server.io_processed), - (unsigned long) server.io_active_threads, - (unsigned long) server.vm_blocked_clients - ); - unlockThreadedIO(); + "# Clients\r\n" + "connected_clients:%d\r\n" + "client_longest_output_list:%lu\r\n" + "client_biggest_input_buf:%lu\r\n" + "blocked_clients:%d\r\n", + listLength(server.clients)-listLength(server.slaves), + lol, bib, + server.bpop_blocked_clients); } - for (j = 0; j < server.dbnum; j++) { - long long keys, vkeys; - keys = dictSize(server.db[j].dict); - vkeys = dictSize(server.db[j].expires); - if (keys || vkeys) { - info = sdscatprintf(info, "db%d:keys=%lld,expires=%lld\r\n", - j, keys, vkeys); + /* Memory */ + if (allsections || defsections || !strcasecmp(section,"memory")) { + char hmem[64]; + char peak_hmem[64]; + + bytesToHuman(hmem,zmalloc_used_memory()); + bytesToHuman(peak_hmem,server.stat_peak_memory); + if (sections++) info = sdscat(info,"\r\n"); + info = sdscatprintf(info, + "# Memory\r\n" + "used_memory:%zu\r\n" + "used_memory_human:%s\r\n" + "used_memory_rss:%zu\r\n" + "used_memory_peak:%zu\r\n" + "used_memory_peak_human:%s\r\n" + "used_memory_lua:%lld\r\n" + "mem_fragmentation_ratio:%.2f\r\n" + "mem_allocator:%s\r\n", + zmalloc_used_memory(), + hmem, + zmalloc_get_rss(), + server.stat_peak_memory, + peak_hmem, + ((long long)lua_gc(server.lua,LUA_GCCOUNT,0))*1024LL, + zmalloc_get_fragmentation_ratio(), + ZMALLOC_LIB + ); + } + + /* Persistence */ + if (allsections || defsections || !strcasecmp(section,"persistence")) { + if (sections++) info = sdscat(info,"\r\n"); + info = sdscatprintf(info, + "# Persistence\r\n" + "loading:%d\r\n" + "aof_enabled:%d\r\n" + "changes_since_last_save:%lld\r\n" + "bgsave_in_progress:%d\r\n" + "last_save_time:%ld\r\n" + "bgrewriteaof_in_progress:%d\r\n", + server.loading, + server.aof_state != REDIS_AOF_OFF, + server.dirty, + server.rdb_child_pid != -1, + server.lastsave, + server.aof_child_pid != -1); + + if (server.aof_state != REDIS_AOF_OFF) { + info = sdscatprintf(info, + "aof_current_size:%lld\r\n" + "aof_base_size:%lld\r\n" + "aof_pending_rewrite:%d\r\n" + "aof_buffer_length:%zu\r\n" + "aof_pending_bio_fsync:%llu\r\n", + (long long) server.aof_current_size, + (long long) server.aof_rewrite_base_size, + server.aof_rewrite_scheduled, + sdslen(server.aof_buf), + bioPendingJobsOfType(REDIS_BIO_AOF_FSYNC)); + } + + if (server.loading) { + double perc; + time_t eta, elapsed; + off_t remaining_bytes = server.loading_total_bytes- + server.loading_loaded_bytes; + + perc = ((double)server.loading_loaded_bytes / + server.loading_total_bytes) * 100; + + elapsed = time(NULL)-server.loading_start_time; + if (elapsed == 0) { + eta = 1; /* A fake 1 second figure if we don't have + enough info */ + } else { + eta = (elapsed*remaining_bytes)/server.loading_loaded_bytes; + } + + info = sdscatprintf(info, + "loading_start_time:%ld\r\n" + "loading_total_bytes:%llu\r\n" + "loading_loaded_bytes:%llu\r\n" + "loading_loaded_perc:%.2f\r\n" + "loading_eta_seconds:%ld\r\n" + ,(unsigned long) server.loading_start_time, + (unsigned long long) server.loading_total_bytes, + (unsigned long long) server.loading_loaded_bytes, + perc, + eta + ); + } + } + + /* Stats */ + if (allsections || defsections || !strcasecmp(section,"stats")) { + if (sections++) info = sdscat(info,"\r\n"); + info = sdscatprintf(info, + "# Stats\r\n" + "total_connections_received:%lld\r\n" + "total_commands_processed:%lld\r\n" + "rejected_connections:%lld\r\n" + "expired_keys:%lld\r\n" + "evicted_keys:%lld\r\n" + "keyspace_hits:%lld\r\n" + "keyspace_misses:%lld\r\n" + "pubsub_channels:%ld\r\n" + "pubsub_patterns:%u\r\n" + "latest_fork_usec:%lld\r\n", + server.stat_numconnections, + server.stat_numcommands, + server.stat_rejected_conn, + server.stat_expiredkeys, + server.stat_evictedkeys, + server.stat_keyspace_hits, + server.stat_keyspace_misses, + dictSize(server.pubsub_channels), + listLength(server.pubsub_patterns), + server.stat_fork_time); + } + + /* Replication */ + if (allsections || defsections || !strcasecmp(section,"replication")) { + if (sections++) info = sdscat(info,"\r\n"); + info = sdscatprintf(info, + "# Replication\r\n" + "role:%s\r\n", + server.masterhost == NULL ? "master" : "slave"); + if (server.masterhost) { + info = sdscatprintf(info, + "master_host:%s\r\n" + "master_port:%d\r\n" + "master_link_status:%s\r\n" + "master_last_io_seconds_ago:%d\r\n" + "master_sync_in_progress:%d\r\n" + ,server.masterhost, + server.masterport, + (server.repl_state == REDIS_REPL_CONNECTED) ? + "up" : "down", + server.master ? + ((int)(time(NULL)-server.master->lastinteraction)) : -1, + server.repl_state == REDIS_REPL_TRANSFER + ); + + if (server.repl_state == REDIS_REPL_TRANSFER) { + info = sdscatprintf(info, + "master_sync_left_bytes:%ld\r\n" + "master_sync_last_io_seconds_ago:%d\r\n" + ,(long)server.repl_transfer_left, + (int)(time(NULL)-server.repl_transfer_lastio) + ); + } + + if (server.repl_state != REDIS_REPL_CONNECTED) { + info = sdscatprintf(info, + "master_link_down_since_seconds:%ld\r\n", + (long)time(NULL)-server.repl_down_since); + } + } + info = sdscatprintf(info, + "connected_slaves:%d\r\n", + listLength(server.slaves)); + if (listLength(server.slaves)) { + int slaveid = 0; + listNode *ln; + listIter li; + + listRewind(server.slaves,&li); + while((ln = listNext(&li))) { + redisClient *slave = listNodeValue(ln); + char *state = NULL; + char ip[32]; + int port; + + if (anetPeerToString(slave->fd,ip,&port) == -1) continue; + switch(slave->replstate) { + case REDIS_REPL_WAIT_BGSAVE_START: + case REDIS_REPL_WAIT_BGSAVE_END: + state = "wait_bgsave"; + break; + case REDIS_REPL_SEND_BULK: + state = "send_bulk"; + break; + case REDIS_REPL_ONLINE: + state = "online"; + break; + } + if (state == NULL) continue; + info = sdscatprintf(info,"slave%d:%s,%d,%s\r\n", + slaveid,ip,port,state); + slaveid++; + } + } + } + + /* CPU */ + if (allsections || defsections || !strcasecmp(section,"cpu")) { + if (sections++) info = sdscat(info,"\r\n"); + info = sdscatprintf(info, + "# CPU\r\n" + "used_cpu_sys:%.2f\r\n" + "used_cpu_user:%.2f\r\n" + "used_cpu_sys_children:%.2f\r\n" + "used_cpu_user_children:%.2f\r\n", + (float)self_ru.ru_stime.tv_sec+(float)self_ru.ru_stime.tv_usec/1000000, + (float)self_ru.ru_utime.tv_sec+(float)self_ru.ru_utime.tv_usec/1000000, + (float)c_ru.ru_stime.tv_sec+(float)c_ru.ru_stime.tv_usec/1000000, + (float)c_ru.ru_utime.tv_sec+(float)c_ru.ru_utime.tv_usec/1000000); + } + + /* cmdtime */ + if (allsections || !strcasecmp(section,"commandstats")) { + if (sections++) info = sdscat(info,"\r\n"); + info = sdscatprintf(info, "# Commandstats\r\n"); + numcommands = sizeof(redisCommandTable)/sizeof(struct redisCommand); + for (j = 0; j < numcommands; j++) { + struct redisCommand *c = redisCommandTable+j; + + if (!c->calls) continue; + info = sdscatprintf(info, + "cmdstat_%s:calls=%lld,usec=%lld,usec_per_call=%.2f\r\n", + c->name, c->calls, c->microseconds, + (c->calls == 0) ? 0 : ((float)c->microseconds/c->calls)); + } + } + + /* Clusetr */ + if (allsections || defsections || !strcasecmp(section,"cluster")) { + if (sections++) info = sdscat(info,"\r\n"); + info = sdscatprintf(info, + "# Cluster\r\n" + "cluster_enabled:%d\r\n", + server.cluster_enabled); + } + + /* Key space */ + if (allsections || defsections || !strcasecmp(section,"keyspace")) { + if (sections++) info = sdscat(info,"\r\n"); + info = sdscatprintf(info, "# Keyspace\r\n"); + for (j = 0; j < server.dbnum; j++) { + long long keys, vkeys; + + keys = dictSize(server.db[j].dict); + vkeys = dictSize(server.db[j].expires); + if (keys || vkeys) { + info = sdscatprintf(info, "db%d:keys=%lld,expires=%lld\r\n", + j, keys, vkeys); + } } } return info; } void infoCommand(redisClient *c) { - sds info = genRedisInfoString(); + char *section = c->argc == 2 ? c->argv[1]->ptr : "default"; + + if (c->argc > 2) { + addReply(c,shared.syntaxerr); + return; + } + sds info = genRedisInfoString(section); addReplySds(c,sdscatprintf(sdsempty(),"$%lu\r\n", (unsigned long)sdslen(info))); addReplySds(c,info); @@ -1263,27 +1739,6 @@ void monitorCommand(redisClient *c) { /* ============================ Maxmemory directive ======================== */ -/* Try to free one object form the pre-allocated objects free list. - * This is useful under low mem conditions as by default we take 1 million - * free objects allocated. On success REDIS_OK is returned, otherwise - * REDIS_ERR. */ -int tryFreeOneObjectFromFreelist(void) { - robj *o; - - if (server.vm_enabled) pthread_mutex_lock(&server.obj_freelist_mutex); - if (listLength(server.objfreelist)) { - listNode *head = listFirst(server.objfreelist); - o = listNodeValue(head); - listDelNode(server.objfreelist,head); - if (server.vm_enabled) pthread_mutex_unlock(&server.obj_freelist_mutex); - zfree(o); - return REDIS_OK; - } else { - if (server.vm_enabled) pthread_mutex_unlock(&server.obj_freelist_mutex); - return REDIS_ERR; - } -} - /* This function gets called when 'maxmemory' is set on the config file to limit * the max memory used by the server, and we are out of memory. * This function will try to, in order: @@ -1296,30 +1751,90 @@ int tryFreeOneObjectFromFreelist(void) { * memory usage. */ void freeMemoryIfNeeded(void) { + /* Remove keys accordingly to the active policy as long as we are + * over the memory limit. */ + if (server.maxmemory_policy == REDIS_MAXMEMORY_NO_EVICTION) return; + while (server.maxmemory && zmalloc_used_memory() > server.maxmemory) { int j, k, freed = 0; - if (tryFreeOneObjectFromFreelist() == REDIS_OK) continue; for (j = 0; j < server.dbnum; j++) { - int minttl = -1; - robj *minkey = NULL; + long bestval = 0; /* just to prevent warning */ + sds bestkey = NULL; struct dictEntry *de; + redisDb *db = server.db+j; + dict *dict; + + if (server.maxmemory_policy == REDIS_MAXMEMORY_ALLKEYS_LRU || + server.maxmemory_policy == REDIS_MAXMEMORY_ALLKEYS_RANDOM) + { + dict = server.db[j].dict; + } else { + dict = server.db[j].expires; + } + if (dictSize(dict) == 0) continue; + + /* volatile-random and allkeys-random policy */ + if (server.maxmemory_policy == REDIS_MAXMEMORY_ALLKEYS_RANDOM || + server.maxmemory_policy == REDIS_MAXMEMORY_VOLATILE_RANDOM) + { + de = dictGetRandomKey(dict); + bestkey = dictGetKey(de); + } + + /* volatile-lru and allkeys-lru policy */ + else if (server.maxmemory_policy == REDIS_MAXMEMORY_ALLKEYS_LRU || + server.maxmemory_policy == REDIS_MAXMEMORY_VOLATILE_LRU) + { + for (k = 0; k < server.maxmemory_samples; k++) { + sds thiskey; + long thisval; + robj *o; + + de = dictGetRandomKey(dict); + thiskey = dictGetKey(de); + /* When policy is volatile-lru we need an additonal lookup + * to locate the real key, as dict is set to db->expires. */ + if (server.maxmemory_policy == REDIS_MAXMEMORY_VOLATILE_LRU) + de = dictFind(db->dict, thiskey); + o = dictGetVal(de); + thisval = estimateObjectIdleTime(o); + + /* Higher idle time is better candidate for deletion */ + if (bestkey == NULL || thisval > bestval) { + bestkey = thiskey; + bestval = thisval; + } + } + } - if (dictSize(server.db[j].expires)) { - freed = 1; - /* From a sample of three keys drop the one nearest to - * the natural expire */ - for (k = 0; k < 3; k++) { - time_t t; - - de = dictGetRandomKey(server.db[j].expires); - t = (time_t) dictGetEntryVal(de); - if (minttl == -1 || t < minttl) { - minkey = dictGetEntryKey(de); - minttl = t; + /* volatile-ttl */ + else if (server.maxmemory_policy == REDIS_MAXMEMORY_VOLATILE_TTL) { + for (k = 0; k < server.maxmemory_samples; k++) { + sds thiskey; + long thisval; + + de = dictGetRandomKey(dict); + thiskey = dictGetKey(de); + thisval = (long) dictGetVal(de); + + /* Expire sooner (minor expire unix timestamp) is better + * candidate for deletion */ + if (bestkey == NULL || thisval < bestval) { + bestkey = thiskey; + bestval = thisval; } } - dbDelete(server.db+j,minkey); + } + + /* Finally remove the selected key. */ + if (bestkey) { + robj *keyobj = createStringObject(bestkey,sdslen(bestkey)); + propagateExpire(db,keyobj); + dbDelete(db,keyobj); + server.stat_evictedkeys++; + decrRefCount(keyobj); + freed++; } } if (!freed) return; /* nothing to free... */ @@ -1350,9 +1865,17 @@ void linuxOvercommitMemoryWarning(void) { } #endif /* __linux__ */ +void createPidFile(void) { + /* Try to write the pid file in a best-effort way. */ + FILE *fp = fopen(server.pidfile,"w"); + if (fp) { + fprintf(fp,"%d\n",(int)getpid()); + fclose(fp); + } +} + void daemonize(void) { int fd; - FILE *fp; if (fork() != 0) exit(0); /* parent exits */ setsid(); /* create a new session */ @@ -1366,12 +1889,6 @@ void daemonize(void) { dup2(fd, STDERR_FILENO); if (fd > STDERR_FILENO) close(fd); } - /* Try to write the pid file */ - fp = fopen(server.pidfile,"w"); - if (fp) { - fprintf(fp,"%d\n",getpid()); - fclose(fp); - } } void version() { @@ -1381,136 +1898,135 @@ void version() { } void usage() { - fprintf(stderr,"Usage: ./redis-server [/path/to/redis.conf]\n"); + fprintf(stderr,"Usage: ./redis-server [/path/to/redis.conf] [options]\n"); fprintf(stderr," ./redis-server - (read config from stdin)\n"); + fprintf(stderr," ./redis-server -v or --version\n"); + fprintf(stderr," ./redis-server -h or --help\n\n"); + fprintf(stderr,"Examples:\n"); + fprintf(stderr," ./redis-server (run the server with default conf)\n"); + fprintf(stderr," ./redis-server /etc/redis/6379.conf\n"); + fprintf(stderr," ./redis-server --port 7777\n"); + fprintf(stderr," ./redis-server --port 7777 --slaveof 127.0.0.1 8888\n"); + fprintf(stderr," ./redis-server /etc/myredis.conf --loglevel verbose\n"); exit(1); } +void redisAsciiArt(void) { +#include "asciilogo.h" + char *buf = zmalloc(1024*16); + + snprintf(buf,1024*16,ascii_logo, + REDIS_VERSION, + redisGitSHA1(), + strtol(redisGitDirty(),NULL,10) > 0, + (sizeof(long) == 8) ? "64" : "32", + server.cluster_enabled ? "cluster" : "stand alone", + server.port, + (long) getpid() + ); + redisLogRaw(REDIS_NOTICE|REDIS_LOG_RAW,buf); + zfree(buf); +} + +static void sigtermHandler(int sig) { + REDIS_NOTUSED(sig); + + redisLog(REDIS_WARNING,"Received SIGTERM, scheduling shutdown..."); + server.shutdown_asap = 1; +} + +void setupSignalHandlers(void) { + struct sigaction act; + + /* When the SA_SIGINFO flag is set in sa_flags then sa_sigaction is used. + * Otherwise, sa_handler is used. */ + sigemptyset(&act.sa_mask); + act.sa_flags = SA_NODEFER | SA_ONSTACK | SA_RESETHAND; + act.sa_handler = sigtermHandler; + sigaction(SIGTERM, &act, NULL); + +#ifdef HAVE_BACKTRACE + sigemptyset(&act.sa_mask); + act.sa_flags = SA_NODEFER | SA_ONSTACK | SA_RESETHAND | SA_SIGINFO; + act.sa_sigaction = sigsegvHandler; + sigaction(SIGSEGV, &act, NULL); + sigaction(SIGBUS, &act, NULL); + sigaction(SIGFPE, &act, NULL); + sigaction(SIGILL, &act, NULL); +#endif + return; +} + int main(int argc, char **argv) { - time_t start; + long long start; + zmalloc_enable_thread_safeness(); initServerConfig(); - sortCommandTable(); - if (argc == 2) { + if (argc >= 2) { + int j = 1; /* First option to parse in argv[] */ + sds options = sdsempty(); + char *configfile = NULL; + + /* Handle special options --help and --version */ if (strcmp(argv[1], "-v") == 0 || strcmp(argv[1], "--version") == 0) version(); - if (strcmp(argv[1], "--help") == 0) usage(); + if (strcmp(argv[1], "--help") == 0 || + strcmp(argv[1], "-h") == 0) usage(); + /* First argument is the config file name? */ + if (argv[j][0] != '-' || argv[j][1] != '-') + configfile = argv[j++]; + /* All the other options are parsed and conceptually appended to the + * configuration file. For instance --port 6380 will generate the + * string "port 6380\n" to be parsed after the actual file name + * is parsed, if any. */ + while(j != argc) { + if (argv[j][0] == '-' && argv[j][1] == '-') { + /* Option name */ + if (sdslen(options)) options = sdscat(options,"\n"); + options = sdscat(options,argv[j]+2); + options = sdscat(options," "); + } else { + /* Option argument */ + options = sdscatrepr(options,argv[j],strlen(argv[j])); + options = sdscat(options," "); + } + j++; + } resetServerSaveParams(); - loadServerConfig(argv[1]); - } else if ((argc > 2)) { - usage(); + loadServerConfig(configfile,options); + sdsfree(options); } else { redisLog(REDIS_WARNING,"Warning: no config file specified, using the default config. In order to specify a config file use 'redis-server /path/to/redis.conf'"); } if (server.daemonize) daemonize(); initServer(); - redisLog(REDIS_NOTICE,"Server started, Redis version " REDIS_VERSION); + if (server.daemonize) createPidFile(); + redisAsciiArt(); + redisLog(REDIS_WARNING,"Server started, Redis version " REDIS_VERSION); #ifdef __linux__ linuxOvercommitMemoryWarning(); #endif - start = time(NULL); - if (server.appendonly) { - if (loadAppendOnlyFile(server.appendfilename) == REDIS_OK) - redisLog(REDIS_NOTICE,"DB loaded from append only file: %ld seconds",time(NULL)-start); + start = ustime(); + if (server.aof_state == REDIS_AOF_ON) { + if (loadAppendOnlyFile(server.aof_filename) == REDIS_OK) + redisLog(REDIS_NOTICE,"DB loaded from append only file: %.3f seconds",(float)(ustime()-start)/1000000); } else { - if (rdbLoad(server.dbfilename) == REDIS_OK) - redisLog(REDIS_NOTICE,"DB loaded from disk: %ld seconds",time(NULL)-start); + if (rdbLoad(server.rdb_filename) == REDIS_OK) { + redisLog(REDIS_NOTICE,"DB loaded from disk: %.3f seconds", + (float)(ustime()-start)/1000000); + } else if (errno != ENOENT) { + redisLog(REDIS_WARNING,"Fatal error loading the DB. Exiting."); + exit(1); + } } - redisLog(REDIS_NOTICE,"The server is now ready to accept connections on port %d", server.port); + if (server.ipfd > 0) + redisLog(REDIS_NOTICE,"The server is now ready to accept connections on port %d", server.port); + if (server.sofd > 0) + redisLog(REDIS_NOTICE,"The server is now ready to accept connections at %s", server.unixsocket); aeSetBeforeSleepProc(server.el,beforeSleep); aeMain(server.el); aeDeleteEventLoop(server.el); return 0; } -/* ============================= Backtrace support ========================= */ - -#ifdef HAVE_BACKTRACE -void *getMcontextEip(ucontext_t *uc) { -#if defined(__FreeBSD__) - return (void*) uc->uc_mcontext.mc_eip; -#elif defined(__dietlibc__) - return (void*) uc->uc_mcontext.eip; -#elif defined(__APPLE__) && !defined(MAC_OS_X_VERSION_10_6) - #if __x86_64__ - return (void*) uc->uc_mcontext->__ss.__rip; - #else - return (void*) uc->uc_mcontext->__ss.__eip; - #endif -#elif defined(__APPLE__) && defined(MAC_OS_X_VERSION_10_6) - #if defined(_STRUCT_X86_THREAD_STATE64) && !defined(__i386__) - return (void*) uc->uc_mcontext->__ss.__rip; - #else - return (void*) uc->uc_mcontext->__ss.__eip; - #endif -#elif defined(__i386__) || defined(__X86_64__) || defined(__x86_64__) - return (void*) uc->uc_mcontext.gregs[REG_EIP]; /* Linux 32/64 bit */ -#elif defined(__ia64__) /* Linux IA64 */ - return (void*) uc->uc_mcontext.sc_ip; -#else - return NULL; -#endif -} - -void segvHandler(int sig, siginfo_t *info, void *secret) { - void *trace[100]; - char **messages = NULL; - int i, trace_size = 0; - ucontext_t *uc = (ucontext_t*) secret; - sds infostring; - REDIS_NOTUSED(info); - - redisLog(REDIS_WARNING, - "======= Ooops! Redis %s got signal: -%d- =======", REDIS_VERSION, sig); - infostring = genRedisInfoString(); - redisLog(REDIS_WARNING, "%s",infostring); - /* It's not safe to sdsfree() the returned string under memory - * corruption conditions. Let it leak as we are going to abort */ - - trace_size = backtrace(trace, 100); - /* overwrite sigaction with caller's address */ - if (getMcontextEip(uc) != NULL) { - trace[1] = getMcontextEip(uc); - } - messages = backtrace_symbols(trace, trace_size); - - for (i=1; i