redis.c 72.7 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
/*
 * Copyright (c) 2009-2010, 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 "redis.h"
31
#include "slowlog.h"
32
#include "bio.h"
33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

#ifdef HAVE_BACKTRACE
#include <execinfo.h>
#include <ucontext.h>
#endif /* HAVE_BACKTRACE */

#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 <limits.h>
#include <float.h>
#include <math.h>
55
#include <sys/resource.h>
56 57 58 59 60

/* Our shared "common" objects */

struct sharedObjectsStruct shared;

61
/* Global vars that are actually used as constants. The following double
62 63 64 65 66 67 68 69 70 71
 * 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 */
struct redisCommand *commandTable;
72 73 74 75 76 77 78 79 80 81 82 83

/* Our command table. 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.
84 85 86 87 88
 * 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. */
89
struct redisCommand redisCommandTable[] = {
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
    {"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,2,2,1,0,0},
    {"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,"w",0,NULL,1,1,1,0,0},
    {"brpoplpush",brpoplpushCommand,4,"wm",0,NULL,1,2,1,0,0},
    {"blpop",blpopCommand,-3,"w",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},
128
    {"spop",spopCommand,2,"wRs",0,NULL,1,1,1,0,0},
129
    {"srandmember",srandmemberCommand,2,"rR",0,NULL,1,1,1,0,0},
130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169
    {"sinter",sinterCommand,-2,"r",0,NULL,1,-1,1,0,0},
    {"sinterstore",sinterstoreCommand,-3,"wm",0,NULL,2,-1,1,0,0},
    {"sunion",sunionCommand,-2,"r",0,NULL,1,-1,1,0,0},
    {"sunionstore",sunionstoreCommand,-3,"wm",0,NULL,2,-1,1,0,0},
    {"sdiff",sdiffCommand,-2,"r",0,NULL,1,-1,1,0,0},
    {"sdiffstore",sdiffstoreCommand,-3,"wm",0,NULL,2,-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},
    {"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},
    {"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},
170
    {"randomkey",randomkeyCommand,1,"rR",0,NULL,0,0,0,0,0},
171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187
    {"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},
    {"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,"r",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,"ar",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},
188 189 190 191
    {"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},
192 193 194 195
    {"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},
196
    {"monitor",monitorCommand,1,"ars",0,NULL,0,0,0,0,0},
197 198
    {"ttl",ttlCommand,2,"r",0,NULL,1,1,1,0,0},
    {"persist",persistCommand,2,"w",0,NULL,1,1,1,0,0},
199
    {"slaveof",slaveofCommand,3,"aws",0,NULL,0,0,0,0,0},
200 201
    {"debug",debugCommand,-2,"aw",0,NULL,0,0,0,0,0},
    {"config",configCommand,-2,"ar",0,NULL,0,0,0,0,0},
202 203 204 205
    {"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},
206
    {"publish",publishCommand,3,"rpf",0,NULL,0,0,0,0,0},
207 208
    {"watch",watchCommand,-2,"rs",0,noPreloadGetKeys,1,-1,1,0,0},
    {"unwatch",unwatchCommand,1,"rs",0,NULL,0,0,0,0,0},
209
    {"cluster",clusterCommand,-2,"ar",0,NULL,0,0,0,0,0},
210
    {"restore",restoreCommand,4,"awm",0,NULL,1,1,1,0,0},
211
    {"migrate",migrateCommand,6,"aw",0,NULL,0,0,0,0,0},
212
    {"asking",askingCommand,1,"r",0,NULL,0,0,0,0,0},
213 214 215
    {"dump",dumpCommand,2,"ar",0,NULL,0,0,0,0,0},
    {"object",objectCommand,-2,"r",0,NULL,0,0,0,0,0},
    {"client",clientCommand,-2,"ar",0,NULL,0,0,0,0,0},
216 217
    {"eval",evalCommand,-3,"wms",0,zunionInterGetKeys,0,0,0,0,0},
    {"evalsha",evalShaCommand,-3,"wms",0,zunionInterGetKeys,0,0,0,0,0},
218 219
    {"slowlog",slowlogCommand,-2,"r",0,NULL,0,0,0,0,0},
    {"script",scriptCommand,-2,"ras",0,NULL,0,0,0,0,0}
220 221 222 223
};

/*============================ Utility functions ============================ */

224 225 226
/* Low level logging. To use only for very big messages, otherwise
 * redisLog() is to prefer. */
void redisLogRaw(int level, const char *msg) {
J
Jonah H. Harris 已提交
227 228 229
    const int syslogLevelMap[] = { LOG_DEBUG, LOG_INFO, LOG_NOTICE, LOG_WARNING };
    const char *c = ".-*#";
    time_t now = time(NULL);
230
    FILE *fp;
231
    char buf[64];
A
antirez 已提交
232
    int rawmode = (level & REDIS_LOG_RAW);
233

A
antirez 已提交
234
    level &= 0xff; /* clear flags */
235
    if (level < server.verbosity) return;
236 237 238 239

    fp = (server.logfile == NULL) ? stdout : fopen(server.logfile,"a");
    if (!fp) return;

A
antirez 已提交
240 241 242 243 244 245
    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);
    }
J
Jonah H. Harris 已提交
246 247
    fflush(fp);

248
    if (server.logfile) fclose(fp);
J
Jonah H. Harris 已提交
249 250

    if (server.syslog_enabled) syslog(syslogLevelMap[level], "%s", msg);
251 252
}

253 254 255 256 257 258 259
/* 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];

A
antirez 已提交
260
    if ((level&0xff) < server.verbosity) return;
261 262 263 264 265 266 267 268

    va_start(ap, fmt);
    vsnprintf(msg, sizeof(msg), fmt, ap);
    va_end(ap);

    redisLogRaw(level,msg);
}

269 270 271 272 273 274 275 276 277 278 279
/* Redis generally does not try to recover from out of memory conditions
 * when allocating objects or strings, it is not clear if it will be possible
 * to report this condition to the client since the networking layer itself
 * is based on heap allocation for send buffers, so we simply abort.
 * At least the code will be simpler to read... */
void oom(const char *msg) {
    redisLog(REDIS_WARNING, "%s: Out of memory\n",msg);
    sleep(1);
    abort();
}

A
antirez 已提交
280 281 282 283 284 285 286 287 288 289 290
/* 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;
}

291 292 293 294 295
/* Return the UNIX time in milliseconds */
long long mstime(void) {
    return ustime()/1000;
}

296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325
/*====================== Hash table type implementation  ==================== */

/* This is an hash table type that uses the SDS dynamic strings libary as
 * keys and radis 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;
}

326 327 328 329 330 331 332 333 334
/* 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;
}

335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365
void dictRedisObjectDestructor(void *privdata, void *val)
{
    DICT_NOTUSED(privdata);

    if (val == NULL) return; /* Values of swapped out keys as set 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);
}

unsigned int dictObjHash(const void *key) {
    const robj *o = key;
    return dictGenHashFunction(o->ptr, sdslen((sds)o->ptr));
}

unsigned int dictSdsHash(const void *key) {
    return dictGenHashFunction((unsigned char*)key, sdslen((char*)key));
}

366 367 368 369
unsigned int dictSdsCaseHash(const void *key) {
    return dictGenCaseHashFunction((unsigned char*)key, sdslen((char*)key));
}

370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410
int dictEncObjKeyCompare(void *privdata, const void *key1,
        const void *key2)
{
    robj *o1 = (robj*) key1, *o2 = (robj*) key2;
    int cmp;

    if (o1->encoding == REDIS_ENCODING_INT &&
        o2->encoding == REDIS_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;
}

unsigned int dictEncObjHash(const void *key) {
    robj *o = (robj*) key;

    if (o->encoding == REDIS_ENCODING_RAW) {
        return dictGenHashFunction(o->ptr, sdslen((sds)o->ptr));
    } else {
        if (o->encoding == REDIS_ENCODING_INT) {
            char buf[32];
            int len;

            len = ll2string(buf,32,(long)o->ptr);
            return dictGenHashFunction((unsigned char*)buf, len);
        } else {
            unsigned int hash;

            o = getDecodedObject(o);
            hash = dictGenHashFunction(o->ptr, sdslen((sds)o->ptr));
            decrRefCount(o);
            return hash;
        }
    }
}

411
/* Sets type hash table */
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427
dictType setDictType = {
    dictEncObjHash,            /* hash function */
    NULL,                      /* key dup */
    NULL,                      /* val dup */
    dictEncObjKeyCompare,      /* key compare */
    dictRedisObjectDestructor, /* key destructor */
    NULL                       /* val destructor */
};

/* Sorted sets hash (note: a skiplist is used in addition to the hash table) */
dictType zsetDictType = {
    dictEncObjHash,            /* hash function */
    NULL,                      /* key dup */
    NULL,                      /* val dup */
    dictEncObjKeyCompare,      /* key compare */
    dictRedisObjectDestructor, /* key destructor */
428
    NULL                       /* val destructor */
429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450
};

/* 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 */
    dictRedisObjectDestructor   /* val destructor */
};

/* Db->expires */
dictType keyptrDictType = {
    dictSdsHash,               /* hash function */
    NULL,                      /* key dup */
    NULL,                      /* val dup */
    dictSdsKeyCompare,         /* key compare */
    NULL,                      /* key destructor */
    NULL                       /* val destructor */
};

451 452 453 454 455 456 457 458 459 460
/* 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 */
};

461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482
/* Hash type hash table (note that small hashes are represented with zimpaps) */
dictType hashDictType = {
    dictEncObjHash,             /* hash function */
    NULL,                       /* key dup */
    NULL,                       /* val dup */
    dictEncObjKeyCompare,       /* key compare */
    dictRedisObjectDestructor,  /* key destructor */
    dictRedisObjectDestructor   /* 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 */
    dictRedisObjectDestructor,  /* key destructor */
    dictListDestructor          /* val destructor */
};

A
antirez 已提交
483 484 485 486 487 488 489 490 491 492 493
/* 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 */
};

494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545
int htNeedsResize(dict *dict) {
    long long size, used;

    size = dictSlots(dict);
    used = dictSize(dict);
    return (size && used && size > DICT_HT_INITIAL_SIZE &&
            (used*100/size < REDIS_HT_MINFILL));
}

/* If the percentage of used slots in the HT reaches REDIS_HT_MINFILL
 * we resize the hash table to save memory */
void tryResizeHashTables(void) {
    int j;

    for (j = 0; j < server.dbnum; j++) {
        if (htNeedsResize(server.db[j].dict))
            dictResize(server.db[j].dict);
        if (htNeedsResize(server.db[j].expires))
            dictResize(server.db[j].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 serverCron() loop in order to rehash some key. */
void incrementallyRehash(void) {
    int j;

    for (j = 0; j < server.dbnum; j++) {
        if (dictIsRehashing(server.db[j].dict)) {
            dictRehashMilliseconds(server.db[j].dict,1);
            break; /* already used our millisecond for this loop... */
        }
    }
}

/* 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 (server.bgsavechildpid == -1 && server.bgrewritechildpid == -1)
        dictEnableResize();
    else
        dictDisableResize();
}

/* ======================= Cron: called every 100 ms ======================== */

546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570
/* 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);
            time_t now = time(NULL);

            expired = 0;
            if (num > REDIS_EXPIRELOOKUPS_PER_CRON)
                num = REDIS_EXPIRELOOKUPS_PER_CRON;
            while (num--) {
                dictEntry *de;
                time_t t;

                if ((de = dictGetRandomKey(db->expires)) == NULL) break;
571
                t = (time_t) dictGetVal(de);
572
                if (now > t) {
573
                    sds key = dictGetKey(de);
574 575 576 577 578 579 580 581 582 583 584 585 586
                    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);
    }
}

587 588 589 590
void updateLRUClock(void) {
    server.lruclock = (time(NULL)/REDIS_LRU_CLOCK_RESOLUTION) &
                                                REDIS_LRU_CLOCK_MAX;
}
591

592
int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) {
593
    int j, loops = server.cronloops;
594 595 596 597 598 599 600 601 602
    REDIS_NOTUSED(eventLoop);
    REDIS_NOTUSED(id);
    REDIS_NOTUSED(clientData);

    /* 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 faster than calling time(NULL) */
    server.unixtime = time(NULL);
A
antirez 已提交
603

604
    /* We have just 22 bits per object for LRU information.
605 606
     * 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.
607
     *
608
     * Note that even if this will wrap after 1.5 years it's not a problem,
609
     * everything will still work but just some object will appear younger
610 611 612 613 614
     * to Redis. But for this to happen a given object should never be touched
     * for 1.5 years.
     *
     * Note that you can change the resolution altering the
     * REDIS_LRU_CLOCK_RESOLUTION define.
615
     */
616
    updateLRUClock();
617

618 619 620 621
    /* 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();

622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657
    /* 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);
        redisLog(REDIS_WARNING,"SIGTERM received but errors trying to shut down the server, check the logs for more information");
    }

    /* Show some info about non-empty databases */
    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 (!(loops % 50) && (used || vkeys)) {
            redisLog(REDIS_VERBOSE,"DB %d: %lld keys (%lld volatile) in %lld slots HT.",j,used,vkeys,size);
            /* dictPrintStats(server.dict); */
        }
    }

    /* We don't want to resize the hash tables while a bacground saving
     * is in progress: the saving child is created using fork() that is
     * implemented with a copy-on-write semantic in most modern systems, so
     * 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 (!(loops % 10)) tryResizeHashTables();
        if (server.activerehashing) incrementallyRehash();
    }

    /* Show information about connected clients */
    if (!(loops % 50)) {
        redisLog(REDIS_VERBOSE,"%d clients connected (%d slaves), %zu bytes in use",
            listLength(server.clients)-listLength(server.slaves),
            listLength(server.slaves),
658
            zmalloc_used_memory());
659 660 661
    }

    /* Close connections of timedout clients */
662
    if ((server.maxidletime && !(loops % 100)) || server.bpop_blocked_clients)
663 664
        closeTimedoutClients();

665 666 667 668 669 670 671 672
    /* Start a scheduled AOF rewrite if this was requested by the user while
     * a BGSAVE was in progress. */
    if (server.bgsavechildpid == -1 && server.bgrewritechildpid == -1 &&
        server.aofrewrite_scheduled)
    {
        rewriteAppendOnlyFileBackground();
    }

A
antirez 已提交
673
    /* Check if a background saving or AOF rewrite in progress terminated. */
674 675 676 677 678
    if (server.bgsavechildpid != -1 || server.bgrewritechildpid != -1) {
        int statloc;
        pid_t pid;

        if ((pid = wait3(&statloc,WNOHANG,NULL)) != 0) {
679 680 681 682 683
            int exitcode = WEXITSTATUS(statloc);
            int bysignal = 0;
            
            if (WIFSIGNALED(statloc)) bysignal = WTERMSIG(statloc);

684
            if (pid == server.bgsavechildpid) {
685
                backgroundSaveDoneHandler(exitcode,bysignal);
686
            } else {
687
                backgroundRewriteDoneHandler(exitcode,bysignal);
688 689 690
            }
            updateDictResizePolicy();
        }
A
antirez 已提交
691
    } else {
692
         time_t now = time(NULL);
693 694 695

        /* If there is not a background saving/rewrite in progress check if
         * we have to save/rewrite now */
696 697 698 699 700 701 702 703 704 705 706
         for (j = 0; j < server.saveparamslen; j++) {
            struct saveparam *sp = server.saveparams+j;

            if (server.dirty >= sp->changes &&
                now-server.lastsave > sp->seconds) {
                redisLog(REDIS_NOTICE,"%d changes in %d seconds. Saving...",
                    sp->changes, sp->seconds);
                rdbSaveBackground(server.dbfilename);
                break;
            }
         }
707 708

         /* Trigger an AOF rewrite if needed */
709 710 711
         if (server.bgsavechildpid == -1 &&
             server.bgrewritechildpid == -1 &&
             server.auto_aofrewrite_perc &&
712 713
             server.appendonly_current_size > server.auto_aofrewrite_min_size)
         {
714
            long long base = server.auto_aofrewrite_base_size ?
715
                            server.auto_aofrewrite_base_size : 1;
716
            long long growth = (server.appendonly_current_size*100/base) - 100;
717
            if (growth >= server.auto_aofrewrite_perc) {
718
                redisLog(REDIS_NOTICE,"Starting automatic rewriting of AOF on %lld%% growth",growth);
719 720 721
                rewriteAppendOnlyFileBackground();
            }
         }
722 723
    }

724 725 726 727 728

    /* 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);

729 730 731 732
    /* 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();
733

734 735
    /* Replication cron function -- used to reconnect to master and
     * to detect transfer failures. */
736
    if (!(loops % 10)) replicationCron();
737

A
antirez 已提交
738 739 740
    /* Run other sub-systems specific cron jobs */
    if (server.cluster_enabled && !(loops % 10)) clusterCron();

741
    server.cronloops++;
742 743 744 745 746 747 748 749
    return 100;
}

/* 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) {
    REDIS_NOTUSED(eventLoop);
750 751
    listNode *ln;
    redisClient *c;
752

753 754 755 756 757 758
    /* 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);
759
        c->flags &= ~REDIS_UNBLOCKED;
760 761 762 763 764 765

        /* Process remaining data in the input buffer. */
        if (c->querybuf && sdslen(c->querybuf) > 0)
            processInputBuffer(c);
    }

766
    /* Write the AOF buffer on disk */
767
    flushAppendOnlyFile(0);
768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796
}

/* =========================== Server initialization ======================== */

void createSharedObjects(void) {
    int j;

    shared.crlf = createObject(REDIS_STRING,sdsnew("\r\n"));
    shared.ok = createObject(REDIS_STRING,sdsnew("+OK\r\n"));
    shared.err = createObject(REDIS_STRING,sdsnew("-ERR\r\n"));
    shared.emptybulk = createObject(REDIS_STRING,sdsnew("$0\r\n\r\n"));
    shared.czero = createObject(REDIS_STRING,sdsnew(":0\r\n"));
    shared.cone = createObject(REDIS_STRING,sdsnew(":1\r\n"));
    shared.cnegone = createObject(REDIS_STRING,sdsnew(":-1\r\n"));
    shared.nullbulk = createObject(REDIS_STRING,sdsnew("$-1\r\n"));
    shared.nullmultibulk = createObject(REDIS_STRING,sdsnew("*-1\r\n"));
    shared.emptymultibulk = createObject(REDIS_STRING,sdsnew("*0\r\n"));
    shared.pong = createObject(REDIS_STRING,sdsnew("+PONG\r\n"));
    shared.queued = createObject(REDIS_STRING,sdsnew("+QUEUED\r\n"));
    shared.wrongtypeerr = createObject(REDIS_STRING,sdsnew(
        "-ERR Operation against a key holding the wrong kind of value\r\n"));
    shared.nokeyerr = createObject(REDIS_STRING,sdsnew(
        "-ERR no such key\r\n"));
    shared.syntaxerr = createObject(REDIS_STRING,sdsnew(
        "-ERR syntax error\r\n"));
    shared.sameobjecterr = createObject(REDIS_STRING,sdsnew(
        "-ERR source and destination objects are the same\r\n"));
    shared.outofrangeerr = createObject(REDIS_STRING,sdsnew(
        "-ERR index out of range\r\n"));
A
antirez 已提交
797 798
    shared.noscripterr = createObject(REDIS_STRING,sdsnew(
        "-NOSCRIPT No matching script. Please use EVAL.\r\n"));
799 800
    shared.loadingerr = createObject(REDIS_STRING,sdsnew(
        "-LOADING Redis is loading the dataset in memory\r\n"));
801
    shared.slowscripterr = createObject(REDIS_STRING,sdsnew(
802
        "-BUSY Redis is busy running a script. Please wait or stop the server with SHUTDOWN.\r\n"));
803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831
    shared.space = createObject(REDIS_STRING,sdsnew(" "));
    shared.colon = createObject(REDIS_STRING,sdsnew(":"));
    shared.plus = createObject(REDIS_STRING,sdsnew("+"));
    shared.select0 = createStringObject("select 0\r\n",10);
    shared.select1 = createStringObject("select 1\r\n",10);
    shared.select2 = createStringObject("select 2\r\n",10);
    shared.select3 = createStringObject("select 3\r\n",10);
    shared.select4 = createStringObject("select 4\r\n",10);
    shared.select5 = createStringObject("select 5\r\n",10);
    shared.select6 = createStringObject("select 6\r\n",10);
    shared.select7 = createStringObject("select 7\r\n",10);
    shared.select8 = createStringObject("select 8\r\n",10);
    shared.select9 = createStringObject("select 9\r\n",10);
    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.mbulk3 = createStringObject("*3\r\n",4);
    shared.mbulk4 = createStringObject("*4\r\n",4);
    for (j = 0; j < REDIS_SHARED_INTEGERS; j++) {
        shared.integers[j] = createObject(REDIS_STRING,(void*)(long)j);
        shared.integers[j]->encoding = REDIS_ENCODING_INT;
    }
}

void initServerConfig() {
    server.port = REDIS_SERVERPORT;
832
    server.bindaddr = NULL;
833
    server.unixsocket = NULL;
834
    server.unixsocketperm = 0;
835 836 837
    server.ipfd = -1;
    server.sofd = -1;
    server.dbnum = REDIS_DEFAULT_DBNUM;
838 839 840
    server.verbosity = REDIS_VERBOSE;
    server.maxidletime = REDIS_MAXIDLETIME;
    server.saveparams = NULL;
841
    server.loading = 0;
842
    server.logfile = NULL; /* NULL = log on standard output */
J
Jonah H. Harris 已提交
843 844 845
    server.syslog_enabled = 0;
    server.syslog_ident = zstrdup("redis");
    server.syslog_facility = LOG_LOCAL0;
846 847 848 849
    server.daemonize = 0;
    server.appendonly = 0;
    server.appendfsync = APPENDFSYNC_EVERYSEC;
    server.no_appendfsync_on_rewrite = 0;
850 851 852 853
    server.auto_aofrewrite_perc = REDIS_AUTO_AOFREWRITE_PERC;
    server.auto_aofrewrite_min_size = REDIS_AUTO_AOFREWRITE_MIN_SIZE;
    server.auto_aofrewrite_base_size = 0;
    server.aofrewrite_scheduled = 0;
854 855 856
    server.lastfsync = time(NULL);
    server.appendfd = -1;
    server.appendseldb = -1; /* Make sure the first time will not match */
857
    server.aof_flush_postponed_start = 0;
858 859 860 861 862 863
    server.pidfile = zstrdup("/var/run/redis.pid");
    server.dbfilename = zstrdup("dump.rdb");
    server.appendfilename = zstrdup("appendonly.aof");
    server.requirepass = NULL;
    server.rdbcompression = 1;
    server.activerehashing = 1;
864
    server.maxclients = REDIS_MAX_CLIENTS;
865
    server.bpop_blocked_clients = 0;
866
    server.maxmemory = 0;
867 868
    server.maxmemory_policy = REDIS_MAXMEMORY_VOLATILE_LRU;
    server.maxmemory_samples = 3;
869 870 871 872
    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;
873
    server.set_max_intset_entries = REDIS_SET_MAX_INTSET_ENTRIES;
874 875
    server.zset_max_ziplist_entries = REDIS_ZSET_MAX_ZIPLIST_ENTRIES;
    server.zset_max_ziplist_value = REDIS_ZSET_MAX_ZIPLIST_VALUE;
876
    server.shutdown_asap = 0;
A
1fe4cd5  
antirez 已提交
877 878
    server.repl_ping_slave_period = REDIS_REPL_PING_SLAVE_PERIOD;
    server.repl_timeout = REDIS_REPL_TIMEOUT;
A
antirez 已提交
879
    server.cluster_enabled = 0;
880
    server.cluster.configfile = zstrdup("nodes.conf");
A
antirez 已提交
881
    server.lua_time_limit = REDIS_LUA_TIME_LIMIT;
882
    server.lua_client = NULL;
883
    server.lua_timedout = 0;
884

A
antirez 已提交
885
    updateLRUClock();
886 887 888 889 890 891 892 893 894 895 896 897
    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;
898
    server.repl_syncio_timeout = REDIS_REPL_SYNCIO_TIMEOUT;
899
    server.repl_serve_stale_data = 1;
900
    server.repl_down_since = -1;
901 902 903 904 905 906

    /* 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;
907 908 909 910 911 912 913 914

    /* 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");
915 916 917 918
    
    /* Slow log */
    server.slowlog_log_slower_than = REDIS_SLOWLOG_LOG_SLOWER_THAN;
    server.slowlog_max_len = REDIS_SLOWLOG_MAX_LEN;
919 920 921 922 923 924 925
}

void initServer() {
    int j;

    signal(SIGHUP, SIG_IGN);
    signal(SIGPIPE, SIG_IGN);
926
    setupSignalHandlers();
927

J
Jonah H. Harris 已提交
928 929 930 931 932
    if (server.syslog_enabled) {
        openlog(server.syslog_ident, LOG_PID | LOG_NDELAY | LOG_NOWAIT,
            server.syslog_facility);
    }

933 934 935
    server.clients = listCreate();
    server.slaves = listCreate();
    server.monitors = listCreate();
936
    server.unblocked_clients = listCreate();
937

938 939 940
    createSharedObjects();
    server.el = aeCreateEventLoop();
    server.db = zmalloc(sizeof(redisDb)*server.dbnum);
941

942
    if (server.port != 0) {
943
        server.ipfd = anetTcpServer(server.neterr,server.port,server.bindaddr);
944
        if (server.ipfd == ANET_ERR) {
945 946
            redisLog(REDIS_WARNING, "Opening port %d: %s",
                server.port, server.neterr);
947 948
            exit(1);
        }
949
    }
950 951
    if (server.unixsocket != NULL) {
        unlink(server.unixsocket); /* don't care if this fails */
952
        server.sofd = anetUnixServer(server.neterr,server.unixsocket,server.unixsocketperm);
953 954 955 956
        if (server.sofd == ANET_ERR) {
            redisLog(REDIS_WARNING, "Opening socket: %s", server.neterr);
            exit(1);
        }
957
    }
958 959
    if (server.ipfd < 0 && server.sofd < 0) {
        redisLog(REDIS_WARNING, "Configured to not listen anywhere, exiting.");
960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982
        exit(1);
    }
    for (j = 0; j < server.dbnum; j++) {
        server.db[j].dict = dictCreate(&dbDictType,NULL);
        server.db[j].expires = dictCreate(&keyptrDictType,NULL);
        server.db[j].blocking_keys = dictCreate(&keylistDictType,NULL);
        server.db[j].watched_keys = dictCreate(&keylistDictType,NULL);
        server.db[j].id = j;
    }
    server.pubsub_channels = dictCreate(&keylistDictType,NULL);
    server.pubsub_patterns = listCreate();
    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.lastsave = time(NULL);
    server.dirty = 0;
    server.stat_numcommands = 0;
    server.stat_numconnections = 0;
    server.stat_expiredkeys = 0;
983
    server.stat_evictedkeys = 0;
984
    server.stat_starttime = time(NULL);
985 986
    server.stat_keyspace_misses = 0;
    server.stat_keyspace_hits = 0;
987
    server.stat_peak_memory = 0;
988
    server.stat_fork_time = 0;
989 990
    server.unixtime = time(NULL);
    aeCreateTimeEvent(server.el, 1, serverCron, NULL, NULL);
991
    if (server.ipfd > 0 && aeCreateFileEvent(server.el,server.ipfd,AE_READABLE,
992
        acceptTcpHandler,NULL) == AE_ERR) oom("creating file event");
993
    if (server.sofd > 0 && aeCreateFileEvent(server.el,server.sofd,AE_READABLE,
994
        acceptUnixHandler,NULL) == AE_ERR) oom("creating file event");
995 996 997 998 999 1000 1001 1002 1003 1004

    if (server.appendonly) {
        server.appendfd = open(server.appendfilename,O_WRONLY|O_APPEND|O_CREAT,0644);
        if (server.appendfd == -1) {
            redisLog(REDIS_WARNING, "Can't open the append-only file: %s",
                strerror(errno));
            exit(1);
        }
    }

A
antirez 已提交
1005
    if (server.cluster_enabled) clusterInit();
1006
    scriptingInit();
1007
    slowlogInit();
1008
    bioInit();
A
antirez 已提交
1009
    srand(time(NULL)^getpid());
1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042

    /* Try to raise the max number of open files accordingly to the
     * configured max number of clients. Also account for 32 additional
     * file descriptors as we need a few more for persistence, listening
     * sockets, log files and so forth. */
    {
        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);
                }
            }
        }
    }
1043 1044
}

1045 1046 1047 1048
/* Populates the Redis Command Table starting from the hard coded list
 * we have on top of redis.c file. */
void populateCommandTable(void) {
    int j;
1049
    int numcommands = sizeof(redisCommandTable)/sizeof(struct redisCommand);
1050 1051

    for (j = 0; j < numcommands; j++) {
1052
        struct redisCommand *c = redisCommandTable+j;
1053
        char *f = c->sflags;
1054
        int retval;
1055

1056 1057 1058 1059 1060 1061 1062 1063
        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;
1064 1065
            case 's': c->flags |= REDIS_CMD_NOSCRIPT; break;
            case 'R': c->flags |= REDIS_CMD_RANDOM; break;
1066 1067 1068 1069 1070
            default: redisPanic("Unsupported command flag"); break;
            }
            f++;
        }

1071 1072 1073
        retval = dictAdd(server.commands, sdsnew(c->name), c);
        assert(retval == DICT_OK);
    }
1074 1075
}

1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087
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;
    }
}

1088 1089
/* ====================== Commands lookup and execution ===================== */

1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100
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;
1101 1102 1103
}

/* Call() is the core of Redis execution of a command */
1104
void call(redisClient *c) {
1105
    long long dirty, start = ustime(), duration;
1106 1107

    dirty = server.dirty;
1108
    c->cmd->proc(c);
1109
    dirty = server.dirty-dirty;
1110
    duration = ustime()-start;
1111
    c->cmd->microseconds += duration;
1112
    slowlogPushEntryIfNeeded(c->argv,c->argc,duration);
1113
    c->cmd->calls++;
1114

A
antirez 已提交
1115
    if (server.appendonly && dirty > 0)
1116
        feedAppendOnlyFile(c->cmd,c->db->id,c->argv,c->argc);
A
antirez 已提交
1117
    if ((dirty > 0 || c->cmd->flags & REDIS_CMD_FORCE_REPLICATION) &&
1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133
        listLength(server.slaves))
        replicationFeedSlaves(server.slaves,c->db->id,c->argv,c->argc);
    if (listLength(server.monitors))
        replicationFeedMonitors(server.monitors,c->db->id,c->argv,c->argc);
    server.stat_numcommands++;
}

/* If this function gets called we already read a whole
 * command, argments are in the client argv/argc fields.
 * processCommand() execute the command or prepare the
 * server for a bulk read from the client.
 *
 * If 1 is returned the client is still alive and valid and
 * 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) {
P
Pieter Noordhuis 已提交
1134 1135 1136 1137
    /* 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. */
1138
    if (!strcasecmp(c->argv[0]->ptr,"quit")) {
P
Pieter Noordhuis 已提交
1139
        addReply(c,shared.ok);
1140
        c->flags |= REDIS_CLOSE_AFTER_REPLY;
1141
        return REDIS_ERR;
1142 1143 1144
    }

    /* Now lookup the command and check ASAP about trivial error conditions
1145 1146 1147
     * such as wrong arity, bad command name and so forth. */
    c->cmd = lookupCommand(c->argv[0]->ptr);
    if (!c->cmd) {
1148 1149
        addReplyErrorFormat(c,"unknown command '%s'",
            (char*)c->argv[0]->ptr);
1150
        return REDIS_OK;
1151 1152
    } else if ((c->cmd->arity > 0 && c->cmd->arity != c->argc) ||
               (c->argc < -c->cmd->arity)) {
1153
        addReplyErrorFormat(c,"wrong number of arguments for '%s' command",
1154
            c->cmd->name);
1155
        return REDIS_OK;
1156 1157 1158
    }

    /* Check if the user is authenticated */
1159 1160
    if (server.requirepass && !c->authenticated && c->cmd->proc != authCommand)
    {
1161
        addReplyError(c,"operation not permitted");
1162
        return REDIS_OK;
1163 1164
    }

A
antirez 已提交
1165 1166
    /* If cluster is enabled, redirect here */
    if (server.cluster_enabled &&
1167
                !(c->cmd->getkeys_proc == NULL && c->cmd->firstkey == 0)) {
A
antirez 已提交
1168 1169 1170 1171 1172 1173
        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 {
1174
            int ask;
1175
            clusterNode *n = getNodeByQuery(c,c->cmd,c->argv,c->argc,&hashslot,&ask);
A
antirez 已提交
1176
            if (n == NULL) {
1177
                addReplyError(c,"Multi keys request invalid in cluster");
A
antirez 已提交
1178 1179 1180
                return REDIS_OK;
            } else if (n != server.cluster.myself) {
                addReplySds(c,sdscatprintf(sdsempty(),
1181 1182
                    "-%s %d %s:%d\r\n", ask ? "ASK" : "MOVED",
                    hashslot,n->ip,n->port));
A
antirez 已提交
1183 1184 1185 1186 1187
                return REDIS_OK;
            }
        }
    }

1188 1189 1190 1191 1192 1193
    /* 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();
1194
    if (server.maxmemory && (c->cmd->flags & REDIS_CMD_DENYOOM) &&
1195
        zmalloc_used_memory() > server.maxmemory)
1196
    {
1197
        addReplyError(c,"command not allowed when used memory > 'maxmemory'");
1198
        return REDIS_OK;
1199 1200 1201 1202 1203
    }

    /* Only allow SUBSCRIBE and UNSUBSCRIBE in the context of Pub/Sub */
    if ((dictSize(c->pubsub_channels) > 0 || listLength(c->pubsub_patterns) > 0)
        &&
1204 1205 1206 1207
        c->cmd->proc != subscribeCommand &&
        c->cmd->proc != unsubscribeCommand &&
        c->cmd->proc != psubscribeCommand &&
        c->cmd->proc != punsubscribeCommand) {
1208
        addReplyError(c,"only (P)SUBSCRIBE / (P)UNSUBSCRIBE / QUIT allowed in this context");
1209
        return REDIS_OK;
1210 1211
    }

1212 1213 1214 1215
    /* 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.replstate != REDIS_REPL_CONNECTED &&
        server.repl_serve_stale_data == 0 &&
1216
        c->cmd->proc != infoCommand && c->cmd->proc != slaveofCommand)
1217 1218 1219 1220 1221 1222
    {
        addReplyError(c,
            "link with MASTER is down and slave-serve-stale-data is set to no");
        return REDIS_OK;
    }

1223
    /* Loading DB? Return an error if the command is not INFO */
1224
    if (server.loading && c->cmd->proc != infoCommand) {
1225 1226 1227 1228
        addReply(c, shared.loadingerr);
        return REDIS_OK;
    }

1229 1230 1231 1232 1233 1234
    /* Lua script too slow? */
    if (server.lua_timedout && c->cmd->proc != shutdownCommand) {
        addReply(c, shared.slowscripterr);
        return REDIS_OK;
    }

1235 1236
    /* Exec the command */
    if (c->flags & REDIS_MULTI &&
1237 1238
        c->cmd->proc != execCommand && c->cmd->proc != discardCommand &&
        c->cmd->proc != multiCommand && c->cmd->proc != watchCommand)
1239
    {
1240
        queueMultiCommand(c);
1241 1242
        addReply(c,shared.queued);
    } else {
1243
        call(c);
1244
    }
1245
    return REDIS_OK;
1246 1247 1248 1249 1250
}

/*================================== Shutdown =============================== */

int prepareForShutdown() {
1251
    redisLog(REDIS_WARNING,"User requested shutdown...");
1252 1253 1254 1255
    /* 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) {
1256
        redisLog(REDIS_WARNING,"There is a child saving an .rdb. Killing it!");
1257 1258 1259
        kill(server.bgsavechildpid,SIGKILL);
        rdbRemoveTempFile(server.bgsavechildpid);
    }
A
antirez 已提交
1260
    if (server.appendonly) {
1261 1262 1263 1264 1265 1266 1267
        /* Kill the AOF saving child as the AOF we already have may be longer
         * but contains the full dataset anyway. */
        if (server.bgrewritechildpid != -1) {
            redisLog(REDIS_WARNING,
                "There is a child rewriting the AOF. Killing it!");
            kill(server.bgrewritechildpid,SIGKILL);
        }
1268
        /* Append only file: fsync() the AOF and exit */
1269
        redisLog(REDIS_NOTICE,"Calling fsync() on the AOF file.");
1270
        aof_fsync(server.appendfd);
1271 1272 1273
    }
    if (server.saveparamslen > 0) {
        redisLog(REDIS_NOTICE,"Saving the final RDB snapshot before exiting.");
1274
        /* Snapshotting. Perform a SYNC SAVE and exit */
1275
        if (rdbSave(server.dbfilename) != REDIS_OK) {
1276 1277 1278 1279 1280
            /* 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... */
1281
            redisLog(REDIS_WARNING,"Error trying to save the DB, can't exit.");
1282 1283 1284
            return REDIS_ERR;
        }
    }
1285 1286 1287 1288
    if (server.daemonize) {
        redisLog(REDIS_NOTICE,"Removing the pid file.");
        unlink(server.pidfile);
    }
1289 1290 1291
    /* Close the listening sockets. Apparently this allows faster restarts. */
    if (server.ipfd != -1) close(server.ipfd);
    if (server.sofd != -1) close(server.sofd);
1292 1293 1294 1295
    if (server.unixsocket) {
        redisLog(REDIS_NOTICE,"Removing the unix socket file.");
        unlink(server.unixsocket); /* don't care if this fails */
    }
1296

1297
    redisLog(REDIS_WARNING,"Redis is now ready to exit, bye bye...");
1298 1299 1300 1301 1302 1303
    return REDIS_OK;
}

/*================================== Commands =============================== */

void authCommand(redisClient *c) {
1304 1305 1306
    if (!server.requirepass) {
        addReplyError(c,"Client sent AUTH, but no password is set");
    } else if (!strcmp(c->argv[1]->ptr, server.requirepass)) {
1307 1308 1309 1310
      c->authenticated = 1;
      addReply(c,shared.ok);
    } else {
      c->authenticated = 0;
1311
      addReplyError(c,"invalid password");
1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346
    }
}

void pingCommand(redisClient *c) {
    addReply(c,shared.pong);
}

void echoCommand(redisClient *c) {
    addReplyBulk(c,c->argv[1]);
}

/* Convert an amount of bytes into a human readable string in the form
 * of 100B, 2G, 100M, 4K, and so forth. */
void bytesToHuman(char *s, unsigned long long n) {
    double d;

    if (n < 1024) {
        /* Bytes */
        sprintf(s,"%lluB",n);
        return;
    } else if (n < (1024*1024)) {
        d = (double)n/(1024);
        sprintf(s,"%.2fK",d);
    } else if (n < (1024LL*1024*1024)) {
        d = (double)n/(1024*1024);
        sprintf(s,"%.2fM",d);
    } else if (n < (1024LL*1024*1024*1024)) {
        d = (double)n/(1024LL*1024*1024);
        sprintf(s,"%.2fG",d);
    }
}

/* 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. */
1347 1348
sds genRedisInfoString(char *section) {
    sds info = sdsempty();
1349
    time_t uptime = time(NULL)-server.stat_starttime;
A
antirez 已提交
1350
    int j, numcommands;
1351
    struct rusage self_ru, c_ru;
1352
    unsigned long lol, bib;
1353 1354 1355 1356 1357
    int allsections = 0, defsections = 0;
    int sections = 0;
    
    if (section) {
        allsections = strcasecmp(section,"all") == 0;
1358
        defsections = strcasecmp(section,"default") == 0;
1359
    }
1360 1361 1362

    getrusage(RUSAGE_SELF, &self_ru);
    getrusage(RUSAGE_CHILDREN, &c_ru);
1363
    getClientsMaxBuffers(&lol,&bib);
1364 1365 1366 1367

    /* Server */
    if (allsections || defsections || !strcasecmp(section,"server")) {
        if (sections++) info = sdscat(info,"\r\n");
1368
        info = sdscatprintf(info,
1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407
            "# 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"
            "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(),
            (long) getpid(),
            server.port,
            uptime,
            uptime/(3600*24),
            (unsigned long) server.lruclock);
    }

    /* Clients */
    if (allsections || defsections || !strcasecmp(section,"clients")) {
        if (sections++) info = sdscat(info,"\r\n");
        info = sdscatprintf(info,
            "# 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);
    }

    /* Memory */
    if (allsections || defsections || !strcasecmp(section,"memory")) {
1408 1409 1410 1411 1412
        char hmem[64];
        char peak_hmem[64];

        bytesToHuman(hmem,zmalloc_used_memory());
        bytesToHuman(peak_hmem,server.stat_peak_memory);
1413 1414 1415 1416 1417 1418
        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"
1419 1420
            "used_memory_peak:%zu\r\n"
            "used_memory_peak_human:%s\r\n"
A
antirez 已提交
1421
            "used_memory_lua:%lld\r\n"
1422
            "mem_fragmentation_ratio:%.2f\r\n"
1423
            "mem_allocator:%s\r\n",
1424 1425 1426
            zmalloc_used_memory(),
            hmem,
            zmalloc_get_rss(),
1427 1428
            server.stat_peak_memory,
            peak_hmem,
A
antirez 已提交
1429
            ((long long)lua_gc(server.lua,LUA_GCCOUNT,0))*1024LL,
1430
            zmalloc_get_fragmentation_ratio(),
1431
            ZMALLOC_LIB
1432
            );
1433 1434
    }

1435 1436 1437
    /* Persistence */
    if (allsections || defsections || !strcasecmp(section,"persistence")) {
        if (sections++) info = sdscat(info,"\r\n");
1438
        info = sdscatprintf(info,
1439 1440 1441 1442 1443 1444 1445 1446 1447 1448
            "# 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.appendonly,
            server.dirty,
A
antirez 已提交
1449
            server.bgsavechildpid != -1,
1450 1451 1452
            server.lastsave,
            server.bgrewritechildpid != -1);

1453 1454 1455 1456 1457 1458 1459 1460 1461 1462
        if (server.appendonly) {
            info = sdscatprintf(info,
                "aof_current_size:%lld\r\n"
                "aof_base_size:%lld\r\n"
                "aof_pending_rewrite:%d\r\n",
                (long long) server.appendonly_current_size,
                (long long) server.auto_aofrewrite_base_size,
                server.aofrewrite_scheduled);
        }

1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492
        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
            );
        }
1493
    }
1494 1495 1496 1497

    /* Stats */
    if (allsections || defsections || !strcasecmp(section,"stats")) {
        if (sections++) info = sdscat(info,"\r\n");
1498
        info = sdscatprintf(info,
1499 1500 1501 1502 1503 1504 1505 1506
            "# Stats\r\n"
            "total_connections_received:%lld\r\n"
            "total_commands_processed:%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"
1507 1508
            "pubsub_patterns:%u\r\n"
            "latest_fork_usec:%lld\r\n",
1509 1510 1511 1512 1513 1514 1515
            server.stat_numconnections,
            server.stat_numcommands,
            server.stat_expiredkeys,
            server.stat_evictedkeys,
            server.stat_keyspace_hits,
            server.stat_keyspace_misses,
            dictSize(server.pubsub_channels),
1516 1517
            listLength(server.pubsub_patterns),
            server.stat_fork_time);
1518
    }
A
antirez 已提交
1519

1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550
    /* 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.replstate == REDIS_REPL_CONNECTED) ?
                    "up" : "down",
                server.master ?
                ((int)(time(NULL)-server.master->lastinteraction)) : -1,
                server.replstate == REDIS_REPL_TRANSFER
            );

            if (server.replstate == 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)
                );
            }
1551 1552 1553 1554 1555 1556

            if (server.replstate != REDIS_REPL_CONNECTED) {
                info = sdscatprintf(info,
                    "master_link_down_since_seconds:%ld\r\n",
                    (long)time(NULL)-server.repl_down_since);
            }
A
antirez 已提交
1557
        }
1558 1559 1560
        info = sdscatprintf(info,
            "connected_slaves:%d\r\n",
            listLength(server.slaves));
A
antirez 已提交
1561 1562
    }

1563 1564
    /* CPU */
    if (allsections || defsections || !strcasecmp(section,"cpu")) {
1565 1566
        if (sections++) info = sdscat(info,"\r\n");
        info = sdscatprintf(info,
1567
        "# CPU\r\n"
1568 1569
        "used_cpu_sys:%.2f\r\n"
        "used_cpu_user:%.2f\r\n"
1570 1571
        "used_cpu_sys_children:%.2f\r\n"
        "used_cpu_user_children:%.2f\r\n",
1572
        (float)self_ru.ru_stime.tv_sec+(float)self_ru.ru_stime.tv_usec/1000000,
1573 1574 1575
        (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);
1576
    }
1577

1578 1579 1580 1581
    /* cmdtime */
    if (allsections || !strcasecmp(section,"commandstats")) {
        if (sections++) info = sdscat(info,"\r\n");
        info = sdscatprintf(info, "# Commandstats\r\n");
1582
        numcommands = sizeof(redisCommandTable)/sizeof(struct redisCommand);
1583
        for (j = 0; j < numcommands; j++) {
1584
            struct redisCommand *c = redisCommandTable+j;
1585

1586 1587 1588 1589 1590
            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));
1591
        }
A
antirez 已提交
1592 1593
    }

1594 1595 1596 1597 1598 1599 1600 1601 1602
    /* 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);
    }

1603 1604 1605 1606 1607 1608
    /* 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;
1609

1610 1611 1612 1613 1614 1615
            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);
            }
1616 1617 1618 1619 1620 1621
        }
    }
    return info;
}

void infoCommand(redisClient *c) {
1622 1623 1624 1625 1626 1627 1628
    char *section = c->argc == 2 ? c->argv[1]->ptr : "default";

    if (c->argc > 2) {
        addReply(c,shared.syntaxerr);
        return;
    }
    sds info = genRedisInfoString(section);
1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658
    addReplySds(c,sdscatprintf(sdsempty(),"$%lu\r\n",
        (unsigned long)sdslen(info)));
    addReplySds(c,info);
    addReply(c,shared.crlf);
}

void monitorCommand(redisClient *c) {
    /* ignore MONITOR if aleady slave or in monitor mode */
    if (c->flags & REDIS_SLAVE) return;

    c->flags |= (REDIS_SLAVE|REDIS_MONITOR);
    c->slaveseldb = 0;
    listAddNodeTail(server.monitors,c);
    addReply(c,shared.ok);
}

/* ============================ Maxmemory directive  ======================== */

/* 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:
 *
 * - Free objects from the free list
 * - Try to remove keys with an EXPIRE set
 *
 * It is not possible to free enough memory to reach used-memory < maxmemory
 * the server will start refusing commands that will enlarge even more the
 * memory usage.
 */
void freeMemoryIfNeeded(void) {
1659 1660
    /* Remove keys accordingly to the active policy as long as we are
     * over the memory limit. */
1661 1662
    if (server.maxmemory_policy == REDIS_MAXMEMORY_NO_EVICTION) return;

1663
    while (server.maxmemory && zmalloc_used_memory() > server.maxmemory) {
1664 1665
        int j, k, freed = 0;

1666
        for (j = 0; j < server.dbnum; j++) {
1667
            long bestval = 0; /* just to prevent warning */
1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686
            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);
1687
                bestkey = dictGetKey(de);
1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699
            }

            /* 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);
1700
                    thiskey = dictGetKey(de);
1701 1702 1703 1704
                    /* 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);
1705
                    o = dictGetVal(de);
1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722
                    thisval = estimateObjectIdleTime(o);

                    /* Higher idle time is better candidate for deletion */
                    if (bestkey == NULL || thisval > bestval) {
                        bestkey = thiskey;
                        bestval = thisval;
                    }
                }
            }

            /* 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);
1723 1724
                    thiskey = dictGetKey(de);
                    thisval = (long) dictGetVal(de);
1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737

                    /* Expire sooner (minor expire unix timestamp) is better
                     * candidate for deletion */
                    if (bestkey == NULL || thisval < bestval) {
                        bestkey = thiskey;
                        bestval = thisval;
                    }
                }
            }

            /* Finally remove the selected key. */
            if (bestkey) {
                robj *keyobj = createStringObject(bestkey,sdslen(bestkey));
1738
                propagateExpire(db,keyobj);
1739
                dbDelete(db,keyobj);
1740
                server.stat_evictedkeys++;
1741 1742 1743 1744 1745 1746
                decrRefCount(keyobj);
                freed++;
            }
        }
        if (!freed) return; /* nothing to free... */
    }
1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772
}

/* =================================== Main! ================================ */

#ifdef __linux__
int linuxOvercommitMemoryValue(void) {
    FILE *fp = fopen("/proc/sys/vm/overcommit_memory","r");
    char buf[64];

    if (!fp) return -1;
    if (fgets(buf,64,fp) == NULL) {
        fclose(fp);
        return -1;
    }
    fclose(fp);

    return atoi(buf);
}

void linuxOvercommitMemoryWarning(void) {
    if (linuxOvercommitMemoryValue() == 0) {
        redisLog(REDIS_WARNING,"WARNING overcommit_memory is set to 0! Background save may fail under low memory condition. To fix this issue add 'vm.overcommit_memory = 1' to /etc/sysctl.conf and then reboot or run the command 'sysctl vm.overcommit_memory=1' for this to take effect.");
    }
}
#endif /* __linux__ */

1773 1774 1775 1776
void createPidFile(void) {
    /* Try to write the pid file in a best-effort way. */
    FILE *fp = fopen(server.pidfile,"w");
    if (fp) {
1777
        fprintf(fp,"%d\n",(int)getpid());
1778 1779 1780 1781
        fclose(fp);
    }
}

1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810
void daemonize(void) {
    int fd;

    if (fork() != 0) exit(0); /* parent exits */
    setsid(); /* create a new session */

    /* Every output goes to /dev/null. If Redis is daemonized but
     * the 'logfile' is set to 'stdout' in the configuration file
     * it will not log at all. */
    if ((fd = open("/dev/null", O_RDWR, 0)) != -1) {
        dup2(fd, STDIN_FILENO);
        dup2(fd, STDOUT_FILENO);
        dup2(fd, STDERR_FILENO);
        if (fd > STDERR_FILENO) close(fd);
    }
}

void version() {
    printf("Redis server version %s (%s:%d)\n", REDIS_VERSION,
        redisGitSHA1(), atoi(redisGitDirty()) > 0);
    exit(0);
}

void usage() {
    fprintf(stderr,"Usage: ./redis-server [/path/to/redis.conf]\n");
    fprintf(stderr,"       ./redis-server - (read config from stdin)\n");
    exit(1);
}

A
antirez 已提交
1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827
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);
}

1828
int main(int argc, char **argv) {
1829
    long long start;
1830

1831
    zmalloc_enable_thread_safeness();
1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845
    initServerConfig();
    if (argc == 2) {
        if (strcmp(argv[1], "-v") == 0 ||
            strcmp(argv[1], "--version") == 0) version();
        if (strcmp(argv[1], "--help") == 0) usage();
        resetServerSaveParams();
        loadServerConfig(argv[1]);
    } else if ((argc > 2)) {
        usage();
    } 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();
1846
    if (server.daemonize) createPidFile();
A
antirez 已提交
1847
    redisAsciiArt();
1848 1849 1850 1851
    redisLog(REDIS_NOTICE,"Server started, Redis version " REDIS_VERSION);
#ifdef __linux__
    linuxOvercommitMemoryWarning();
#endif
1852
    start = ustime();
A
antirez 已提交
1853
    if (server.appendonly) {
1854
        if (loadAppendOnlyFile(server.appendfilename) == REDIS_OK)
1855
            redisLog(REDIS_NOTICE,"DB loaded from append only file: %.3f seconds",(float)(ustime()-start)/1000000);
1856
    } else {
1857 1858 1859 1860 1861 1862 1863
        if (rdbLoad(server.dbfilename) == 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);
        }
1864
    }
1865 1866 1867
    if (server.ipfd > 0)
        redisLog(REDIS_NOTICE,"The server is now ready to accept connections on port %d", server.port);
    if (server.sofd > 0)
1868
        redisLog(REDIS_NOTICE,"The server is now ready to accept connections at %s", server.unixsocket);
1869 1870 1871 1872 1873 1874 1875
    aeSetBeforeSleepProc(server.el,beforeSleep);
    aeMain(server.el);
    aeDeleteEventLoop(server.el);
    return 0;
}

#ifdef HAVE_BACKTRACE
1876
static void *getMcontextEip(ucontext_t *uc) {
1877 1878 1879 1880 1881 1882 1883
#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;
1884
  #elif __i386__
1885
    return (void*) uc->uc_mcontext->__ss.__eip;
1886 1887
  #else
    return (void*) uc->uc_mcontext->__ss.__srr0;
1888 1889 1890 1891 1892 1893 1894
  #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
1895 1896 1897 1898
#elif defined(__i386__)
    return (void*) uc->uc_mcontext.gregs[14]; /* Linux 32 */
#elif defined(__X86_64__) || defined(__x86_64__)
    return (void*) uc->uc_mcontext.gregs[16]; /* Linux 64 */
1899 1900 1901 1902 1903 1904 1905
#elif defined(__ia64__) /* Linux IA64 */
    return (void*) uc->uc_mcontext.sc_ip;
#else
    return NULL;
#endif
}

1906
static void sigsegvHandler(int sig, siginfo_t *info, void *secret) {
1907 1908 1909 1910 1911
    void *trace[100];
    char **messages = NULL;
    int i, trace_size = 0;
    ucontext_t *uc = (ucontext_t*) secret;
    sds infostring;
1912
    struct sigaction act;
1913 1914 1915 1916
    REDIS_NOTUSED(info);

    redisLog(REDIS_WARNING,
        "======= Ooops! Redis %s got signal: -%d- =======", REDIS_VERSION, sig);
1917
    infostring = genRedisInfoString("all");
1918
    redisLogRaw(REDIS_WARNING, infostring);
1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932
    /* 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<trace_size; ++i)
        redisLog(REDIS_WARNING,"%s", messages[i]);

    /* free(messages); Don't call free() with possibly corrupted memory. */
1933
    if (server.daemonize) unlink(server.pidfile);
1934 1935 1936 1937 1938 1939 1940 1941 1942 1943

    /* Make sure we exit with the right signal at the end. So for instance
     * the core will be dumped if enabled. */
    sigemptyset (&act.sa_mask);
    /* When the SA_SIGINFO flag is set in sa_flags then sa_sigaction
     * is used. Otherwise, sa_handler is used */
    act.sa_flags = SA_NODEFER | SA_ONSTACK | SA_RESETHAND;
    act.sa_handler = SIG_DFL;
    sigaction (sig, &act, NULL);
    kill(getpid(),sig);
1944
}
1945
#endif /* HAVE_BACKTRACE */
1946

1947
static void sigtermHandler(int sig) {
1948 1949
    REDIS_NOTUSED(sig);

1950
    redisLog(REDIS_WARNING,"Received SIGTERM, scheduling shutdown...");
1951 1952 1953
    server.shutdown_asap = 1;
}

1954
void setupSignalHandlers(void) {
1955 1956
    struct sigaction act;

1957 1958 1959
    /* When the SA_SIGINFO flag is set in sa_flags then sa_sigaction is used.
     * Otherwise, sa_handler is used. */
    sigemptyset(&act.sa_mask);
1960 1961
    act.sa_flags = SA_NODEFER | SA_ONSTACK | SA_RESETHAND;
    act.sa_handler = sigtermHandler;
1962
    sigaction(SIGTERM, &act, NULL);
1963

1964 1965 1966 1967 1968 1969 1970 1971 1972 1973
#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;
1974 1975 1976
}

/* The End */