replication.c 22.7 KB
Newer Older
1 2 3 4 5 6 7
#include "redis.h"

#include <sys/time.h>
#include <unistd.h>
#include <fcntl.h>
#include <sys/stat.h>

8 9
/* ---------------------------------- MASTER -------------------------------- */

10 11 12
void replicationFeedSlaves(list *slaves, int dictid, robj **argv, int argc) {
    listNode *ln;
    listIter li;
13
    int j;
14 15 16 17 18 19 20 21

    listRewind(slaves,&li);
    while((ln = listNext(&li))) {
        redisClient *slave = ln->value;

        /* Don't feed slaves that are still waiting for BGSAVE to start */
        if (slave->replstate == REDIS_REPL_WAIT_BGSAVE_START) continue;

22 23 24
        /* Feed slaves that are waiting for the initial SYNC (so these commands
         * are queued in the output buffer until the intial SYNC completes),
         * or are already in sync with the master. */
25 26 27
        if (slave->slaveseldb != dictid) {
            robj *selectcmd;

28 29
            if (dictid >= 0 && dictid < REDIS_SHARED_SELECT_CMDS) {
                selectcmd = shared.select[dictid];
A
antirez 已提交
30
                incrRefCount(selectcmd);
31
            } else {
32 33 34 35
                selectcmd = createObject(REDIS_STRING,
                    sdscatprintf(sdsempty(),"select %d\r\n",dictid));
            }
            addReply(slave,selectcmd);
36
            decrRefCount(selectcmd);
37 38
            slave->slaveseldb = dictid;
        }
39 40
        addReplyMultiBulkLen(slave,argc);
        for (j = 0; j < argc; j++) addReplyBulk(slave,argv[j]);
41 42 43
    }
}

44
void replicationFeedMonitors(redisClient *c, list *monitors, int dictid, robj **argv, int argc) {
45 46
    listNode *ln;
    listIter li;
47
    int j, port;
48 49
    sds cmdrepr = sdsnew("+");
    robj *cmdobj;
50
    char ip[32];
51 52 53
    struct timeval tv;

    gettimeofday(&tv,NULL);
54
    cmdrepr = sdscatprintf(cmdrepr,"%ld.%06ld ",(long)tv.tv_sec,(long)tv.tv_usec);
55 56 57 58 59 60
    if (c->flags & REDIS_LUA_CLIENT) {
        cmdrepr = sdscatprintf(cmdrepr,"[%d lua] ", dictid);
    } else {
        anetPeerToString(c->fd,ip,&port);
        cmdrepr = sdscatprintf(cmdrepr,"[%d %s:%d] ", dictid,ip,port);
    }
61 62 63

    for (j = 0; j < argc; j++) {
        if (argv[j]->encoding == REDIS_ENCODING_INT) {
64
            cmdrepr = sdscatprintf(cmdrepr, "\"%ld\"", (long)argv[j]->ptr);
65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
        } else {
            cmdrepr = sdscatrepr(cmdrepr,(char*)argv[j]->ptr,
                        sdslen(argv[j]->ptr));
        }
        if (j != argc-1)
            cmdrepr = sdscatlen(cmdrepr," ",1);
    }
    cmdrepr = sdscatlen(cmdrepr,"\r\n",2);
    cmdobj = createObject(REDIS_STRING,cmdrepr);

    listRewind(monitors,&li);
    while((ln = listNext(&li))) {
        redisClient *monitor = ln->value;
        addReply(monitor,cmdobj);
    }
    decrRefCount(cmdobj);
}

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

87 88
    /* Refuse SYNC requests if we are a slave but the link with our master
     * is not ok... */
A
antirez 已提交
89
    if (server.masterhost && server.repl_state != REDIS_REPL_CONNECTED) {
90
        addReplyError(c,"Can't SYNC while not connected with my master");
91 92 93
        return;
    }

94 95 96 97 98
    /* SYNC can't be issued when the server has pending data to send to
     * the client about already issued commands. We need a fresh reply
     * buffer registering the differences between the BGSAVE and the current
     * dataset, so that we can copy to other slaves if needed. */
    if (listLength(c->reply) != 0) {
99
        addReplyError(c,"SYNC is invalid with pending input");
100 101 102 103 104 105
        return;
    }

    redisLog(REDIS_NOTICE,"Slave ask for synchronization");
    /* Here we need to check if there is a background saving operation
     * in progress, or if it is required to start one */
A
antirez 已提交
106
    if (server.rdb_child_pid != -1) {
107 108 109 110 111 112 113 114 115 116 117 118 119 120 121
        /* Ok a background save is in progress. Let's check if it is a good
         * one for replication, i.e. if there is another slave that is
         * registering differences since the server forked to save */
        redisClient *slave;
        listNode *ln;
        listIter li;

        listRewind(server.slaves,&li);
        while((ln = listNext(&li))) {
            slave = ln->value;
            if (slave->replstate == REDIS_REPL_WAIT_BGSAVE_END) break;
        }
        if (ln) {
            /* Perfect, the server is already registering differences for
             * another slave. Set the right state, and copy the buffer. */
122
            copyClientOutputBuffer(c,slave);
123 124 125 126 127 128 129 130 131 132 133
            c->replstate = REDIS_REPL_WAIT_BGSAVE_END;
            redisLog(REDIS_NOTICE,"Waiting for end of BGSAVE for SYNC");
        } else {
            /* No way, we need to wait for the next BGSAVE in order to
             * register differences */
            c->replstate = REDIS_REPL_WAIT_BGSAVE_START;
            redisLog(REDIS_NOTICE,"Waiting for next BGSAVE for SYNC");
        }
    } else {
        /* Ok we don't have a BGSAVE in progress, let's start one */
        redisLog(REDIS_NOTICE,"Starting BGSAVE for SYNC");
A
antirez 已提交
134
        if (rdbSaveBackground(server.rdb_filename) != REDIS_OK) {
135
            redisLog(REDIS_NOTICE,"Replication failed, can't BGSAVE");
136
            addReplyError(c,"Unable to perform background save");
137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226
            return;
        }
        c->replstate = REDIS_REPL_WAIT_BGSAVE_END;
    }
    c->repldbfd = -1;
    c->flags |= REDIS_SLAVE;
    c->slaveseldb = 0;
    listAddNodeTail(server.slaves,c);
    return;
}

void sendBulkToSlave(aeEventLoop *el, int fd, void *privdata, int mask) {
    redisClient *slave = privdata;
    REDIS_NOTUSED(el);
    REDIS_NOTUSED(mask);
    char buf[REDIS_IOBUF_LEN];
    ssize_t nwritten, buflen;

    if (slave->repldboff == 0) {
        /* Write the bulk write count before to transfer the DB. In theory here
         * we don't know how much room there is in the output buffer of the
         * socket, but in pratice SO_SNDLOWAT (the minimum count for output
         * operations) will never be smaller than the few bytes we need. */
        sds bulkcount;

        bulkcount = sdscatprintf(sdsempty(),"$%lld\r\n",(unsigned long long)
            slave->repldbsize);
        if (write(fd,bulkcount,sdslen(bulkcount)) != (signed)sdslen(bulkcount))
        {
            sdsfree(bulkcount);
            freeClient(slave);
            return;
        }
        sdsfree(bulkcount);
    }
    lseek(slave->repldbfd,slave->repldboff,SEEK_SET);
    buflen = read(slave->repldbfd,buf,REDIS_IOBUF_LEN);
    if (buflen <= 0) {
        redisLog(REDIS_WARNING,"Read error sending DB to slave: %s",
            (buflen == 0) ? "premature EOF" : strerror(errno));
        freeClient(slave);
        return;
    }
    if ((nwritten = write(fd,buf,buflen)) == -1) {
        redisLog(REDIS_VERBOSE,"Write error sending DB to slave: %s",
            strerror(errno));
        freeClient(slave);
        return;
    }
    slave->repldboff += nwritten;
    if (slave->repldboff == slave->repldbsize) {
        close(slave->repldbfd);
        slave->repldbfd = -1;
        aeDeleteFileEvent(server.el,slave->fd,AE_WRITABLE);
        slave->replstate = REDIS_REPL_ONLINE;
        if (aeCreateFileEvent(server.el, slave->fd, AE_WRITABLE,
            sendReplyToClient, slave) == AE_ERR) {
            freeClient(slave);
            return;
        }
        redisLog(REDIS_NOTICE,"Synchronization with slave succeeded");
    }
}

/* This function is called at the end of every backgrond saving.
 * The argument bgsaveerr is REDIS_OK if the background saving succeeded
 * otherwise REDIS_ERR is passed to the function.
 *
 * The goal of this function is to handle slaves waiting for a successful
 * background saving in order to perform non-blocking synchronization. */
void updateSlavesWaitingBgsave(int bgsaveerr) {
    listNode *ln;
    int startbgsave = 0;
    listIter li;

    listRewind(server.slaves,&li);
    while((ln = listNext(&li))) {
        redisClient *slave = ln->value;

        if (slave->replstate == REDIS_REPL_WAIT_BGSAVE_START) {
            startbgsave = 1;
            slave->replstate = REDIS_REPL_WAIT_BGSAVE_END;
        } else if (slave->replstate == REDIS_REPL_WAIT_BGSAVE_END) {
            struct redis_stat buf;

            if (bgsaveerr != REDIS_OK) {
                freeClient(slave);
                redisLog(REDIS_WARNING,"SYNC failed. BGSAVE child returned an error");
                continue;
            }
A
antirez 已提交
227
            if ((slave->repldbfd = open(server.rdb_filename,O_RDONLY)) == -1 ||
228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243
                redis_fstat(slave->repldbfd,&buf) == -1) {
                freeClient(slave);
                redisLog(REDIS_WARNING,"SYNC failed. Can't open/stat DB after BGSAVE: %s", strerror(errno));
                continue;
            }
            slave->repldboff = 0;
            slave->repldbsize = buf.st_size;
            slave->replstate = REDIS_REPL_SEND_BULK;
            aeDeleteFileEvent(server.el,slave->fd,AE_WRITABLE);
            if (aeCreateFileEvent(server.el, slave->fd, AE_WRITABLE, sendBulkToSlave, slave) == AE_ERR) {
                freeClient(slave);
                continue;
            }
        }
    }
    if (startbgsave) {
A
antirez 已提交
244
        if (rdbSaveBackground(server.rdb_filename) != REDIS_OK) {
245 246 247 248 249 250 251 252 253 254 255 256 257 258
            listIter li;

            listRewind(server.slaves,&li);
            redisLog(REDIS_WARNING,"SYNC failed. BGSAVE failed");
            while((ln = listNext(&li))) {
                redisClient *slave = ln->value;

                if (slave->replstate == REDIS_REPL_WAIT_BGSAVE_START)
                    freeClient(slave);
            }
        }
    }
}

259 260 261 262
/* ----------------------------------- SLAVE -------------------------------- */

/* Abort the async download of the bulk dataset while SYNC-ing with master */
void replicationAbortSyncTransfer(void) {
A
antirez 已提交
263
    redisAssert(server.repl_state == REDIS_REPL_TRANSFER);
264 265 266 267 268 269

    aeDeleteFileEvent(server.el,server.repl_transfer_s,AE_READABLE);
    close(server.repl_transfer_s);
    close(server.repl_transfer_fd);
    unlink(server.repl_transfer_tmpfile);
    zfree(server.repl_transfer_tmpfile);
A
antirez 已提交
270
    server.repl_state = REDIS_REPL_CONNECT;
271 272 273 274
}

/* Asynchronously read the SYNC payload we receive from a master */
void readSyncBulkPayload(aeEventLoop *el, int fd, void *privdata, int mask) {
275
    char buf[4096];
276 277 278 279
    ssize_t nread, readlen;
    REDIS_NOTUSED(el);
    REDIS_NOTUSED(privdata);
    REDIS_NOTUSED(mask);
280

281 282 283
    /* If repl_transfer_left == -1 we still have to read the bulk length
     * from the master reply. */
    if (server.repl_transfer_left == -1) {
284
        if (syncReadLine(fd,buf,1024,server.repl_syncio_timeout*1000) == -1) {
285 286 287
            redisLog(REDIS_WARNING,
                "I/O error reading bulk count from MASTER: %s",
                strerror(errno));
288
            goto error;
289
        }
290

291 292 293 294
        if (buf[0] == '-') {
            redisLog(REDIS_WARNING,
                "MASTER aborted replication with an error: %s",
                buf+1);
295
            goto error;
296 297 298 299
        } else if (buf[0] == '\0') {
            /* At this stage just a newline works as a PING in order to take
             * the connection live. So we refresh our last interaction
             * timestamp. */
300
            server.repl_transfer_lastio = server.unixtime;
301
            return;
302 303
        } else if (buf[0] != '$') {
            redisLog(REDIS_WARNING,"Bad protocol from MASTER, the first byte is not '$', are you sure the host and port are right?");
304
            goto error;
305 306 307 308 309 310 311 312 313
        }
        server.repl_transfer_left = strtol(buf+1,NULL,10);
        redisLog(REDIS_NOTICE,
            "MASTER <-> SLAVE sync: receiving %ld bytes from master",
            server.repl_transfer_left);
        return;
    }

    /* Read bulk data */
314 315
    readlen = (server.repl_transfer_left < (signed)sizeof(buf)) ?
        server.repl_transfer_left : (signed)sizeof(buf);
316 317 318 319 320 321 322
    nread = read(fd,buf,readlen);
    if (nread <= 0) {
        redisLog(REDIS_WARNING,"I/O error trying to sync with MASTER: %s",
            (nread == -1) ? strerror(errno) : "connection lost");
        replicationAbortSyncTransfer();
        return;
    }
323
    server.repl_transfer_lastio = server.unixtime;
324
    if (write(server.repl_transfer_fd,buf,nread) != nread) {
325
        redisLog(REDIS_WARNING,"Write error or short write writing to the DB dump file needed for MASTER <-> SLAVE synchronization: %s", strerror(errno));
326
        goto error;
327 328 329 330
    }
    server.repl_transfer_left -= nread;
    /* Check if the transfer is now complete */
    if (server.repl_transfer_left == 0) {
A
antirez 已提交
331
        if (rename(server.repl_transfer_tmpfile,server.rdb_filename) == -1) {
332 333 334 335
            redisLog(REDIS_WARNING,"Failed trying to rename the temp DB into dump.rdb in MASTER <-> SLAVE synchronization: %s", strerror(errno));
            replicationAbortSyncTransfer();
            return;
        }
A
antirez 已提交
336
        redisLog(REDIS_NOTICE, "MASTER <-> SLAVE sync: Loading DB in memory");
337
        emptyDb();
338 339 340 341 342
        /* Before loading the DB into memory we need to delete the readable
         * handler, otherwise it will get called recursively since
         * rdbLoad() will call the event loop to process events from time to
         * time for non blocking loading. */
        aeDeleteFileEvent(server.el,server.repl_transfer_s,AE_READABLE);
A
antirez 已提交
343
        if (rdbLoad(server.rdb_filename) != REDIS_OK) {
344 345 346 347 348 349 350 351 352 353
            redisLog(REDIS_WARNING,"Failed trying to load the MASTER synchronization DB from disk");
            replicationAbortSyncTransfer();
            return;
        }
        /* Final setup of the connected slave <- master link */
        zfree(server.repl_transfer_tmpfile);
        close(server.repl_transfer_fd);
        server.master = createClient(server.repl_transfer_s);
        server.master->flags |= REDIS_MASTER;
        server.master->authenticated = 1;
A
antirez 已提交
354
        server.repl_state = REDIS_REPL_CONNECTED;
355
        redisLog(REDIS_NOTICE, "MASTER <-> SLAVE sync: Finished with success");
356 357 358
        /* Restart the AOF subsystem now that we finished the sync. This
         * will trigger an AOF rewrite, and when done will start appending
         * to the new file. */
359
        if (server.aof_state != REDIS_AOF_OFF) {
360 361 362 363 364 365 366 367 368 369 370 371
            int retry = 10;

            stopAppendOnly();
            while (retry-- && startAppendOnly() == REDIS_ERR) {
                redisLog(REDIS_WARNING,"Failed enabling the AOF after successful master synchrnization! Trying it again in one second.");
                sleep(1);
            }
            if (!retry) {
                redisLog(REDIS_WARNING,"FATAL: this slave instance finished the synchronization with its master, but the AOF can't be turned on. Exiting now.");
                exit(1);
            }
        }
372
    }
373 374 375 376 377 378

    return;

error:
    replicationAbortSyncTransfer();
    return;
379 380
}

381 382
void syncWithMaster(aeEventLoop *el, int fd, void *privdata, int mask) {
    char buf[1024], tmpfile[256];
383
    int dfd, maxtries = 5;
384 385 386
    REDIS_NOTUSED(el);
    REDIS_NOTUSED(privdata);
    REDIS_NOTUSED(mask);
387

388 389
    /* If this event fired after the user turned the instance into a master
     * with SLAVEOF NO ONE we must just return ASAP. */
A
antirez 已提交
390
    if (server.repl_state == REDIS_REPL_NONE) {
391 392 393 394
        close(fd);
        return;
    }

395
    redisLog(REDIS_NOTICE,"Non blocking connect for SYNC fired the event.");
396 397 398
    /* This event should only be triggered once since it is used to have a
     * non-blocking connect(2) to the master. It has been triggered when this
     * function is called, so we can delete it. */
399
    aeDeleteFileEvent(server.el,fd,AE_READABLE|AE_WRITABLE);
400 401 402

    /* AUTH with the master if required. */
    if(server.masterauth) {
403 404 405 406
        char authcmd[1024];
        size_t authlen;

        authlen = snprintf(authcmd,sizeof(authcmd),"AUTH %s\r\n",server.masterauth);
407
        if (syncWrite(fd,authcmd,authlen,server.repl_syncio_timeout*1000) == -1) {
408 409
            redisLog(REDIS_WARNING,"Unable to AUTH to MASTER: %s",
                strerror(errno));
410 411
            goto error;
        }
412
        /* Read the AUTH result.  */
413
        if (syncReadLine(fd,buf,1024,server.repl_syncio_timeout*1000) == -1) {
414 415
            redisLog(REDIS_WARNING,"I/O error reading auth result from MASTER: %s",
                strerror(errno));
416
            goto error;
417 418 419
        }
        if (buf[0] != '+') {
            redisLog(REDIS_WARNING,"Cannot AUTH to MASTER, is the masterauth password correct?");
420
            goto error;
421 422 423 424
        }
    }

    /* Issue the SYNC command */
425
    if (syncWrite(fd,"SYNC\r\n",6,server.repl_syncio_timeout*1000) == -1) {
426 427
        redisLog(REDIS_WARNING,"I/O error writing to MASTER: %s",
            strerror(errno));
428
        goto error;
429
    }
430 431

    /* Prepare a suitable temp file for bulk transfer */
432 433
    while(maxtries--) {
        snprintf(tmpfile,256,
434
            "temp-%d.%ld.rdb",(int)server.unixtime,(long int)getpid());
435 436 437 438 439 440
        dfd = open(tmpfile,O_CREAT|O_WRONLY|O_EXCL,0644);
        if (dfd != -1) break;
        sleep(1);
    }
    if (dfd == -1) {
        redisLog(REDIS_WARNING,"Opening the temp file needed for MASTER <-> SLAVE synchronization: %s",strerror(errno));
441
        goto error;
442 443
    }

444
    /* Setup the non blocking download of the bulk file. */
445
    if (aeCreateFileEvent(server.el,fd, AE_READABLE,readSyncBulkPayload,NULL)
446
            == AE_ERR)
447 448
    {
        redisLog(REDIS_WARNING,"Can't create readable event for SYNC");
449
        goto error;
450
    }
451

A
antirez 已提交
452
    server.repl_state = REDIS_REPL_TRANSFER;
453
    server.repl_transfer_left = -1;
454
    server.repl_transfer_fd = dfd;
455
    server.repl_transfer_lastio = server.unixtime;
456
    server.repl_transfer_tmpfile = zstrdup(tmpfile);
457 458 459
    return;

error:
A
antirez 已提交
460
    server.repl_state = REDIS_REPL_CONNECT;
461 462 463 464 465 466 467 468 469 470 471 472 473 474
    close(fd);
    return;
}

int connectWithMaster(void) {
    int fd;

    fd = anetTcpNonBlockConnect(NULL,server.masterhost,server.masterport);
    if (fd == -1) {
        redisLog(REDIS_WARNING,"Unable to connect to MASTER: %s",
            strerror(errno));
        return REDIS_ERR;
    }

475
    if (aeCreateFileEvent(server.el,fd,AE_READABLE|AE_WRITABLE,syncWithMaster,NULL) ==
476
            AE_ERR)
477 478 479 480 481 482
    {
        close(fd);
        redisLog(REDIS_WARNING,"Can't create readable event for SYNC");
        return REDIS_ERR;
    }

483
    server.repl_transfer_lastio = server.unixtime;
484
    server.repl_transfer_s = fd;
A
antirez 已提交
485
    server.repl_state = REDIS_REPL_CONNECTING;
486 487 488
    return REDIS_OK;
}

489 490 491 492 493
/* This function can be called when a non blocking connection is currently
 * in progress to undo it. */
void undoConnectWithMaster(void) {
    int fd = server.repl_transfer_s;

A
antirez 已提交
494
    redisAssert(server.repl_state == REDIS_REPL_CONNECTING);
495 496 497
    aeDeleteFileEvent(server.el,fd,AE_READABLE|AE_WRITABLE);
    close(fd);
    server.repl_transfer_s = -1;
A
antirez 已提交
498
    server.repl_state = REDIS_REPL_CONNECT;
499 500
}

501 502 503 504 505 506 507
void slaveofCommand(redisClient *c) {
    if (!strcasecmp(c->argv[1]->ptr,"no") &&
        !strcasecmp(c->argv[2]->ptr,"one")) {
        if (server.masterhost) {
            sdsfree(server.masterhost);
            server.masterhost = NULL;
            if (server.master) freeClient(server.master);
A
antirez 已提交
508
            if (server.repl_state == REDIS_REPL_TRANSFER)
509
                replicationAbortSyncTransfer();
A
antirez 已提交
510
            else if (server.repl_state == REDIS_REPL_CONNECTING)
511
                undoConnectWithMaster();
A
antirez 已提交
512
            server.repl_state = REDIS_REPL_NONE;
513 514 515
            redisLog(REDIS_NOTICE,"MASTER MODE enabled (user request)");
        }
    } else {
516 517 518 519 520 521 522 523 524 525 526 527 528 529
        long port;

        if ((getLongFromObjectOrReply(c, c->argv[2], &port, NULL) != REDIS_OK))
            return;

        /* Check if we are already attached to the specified slave */
        if (server.masterhost && !strcasecmp(server.masterhost,c->argv[1]->ptr)
            && server.masterport == port) {
            redisLog(REDIS_NOTICE,"SLAVE OF would result into synchronization with the master we are already connected with. No operation performed.");
            addReplySds(c,sdsnew("+OK Already connected to specified master\r\n"));
            return;
        }
        /* There was no previous master or the user specified a different one,
         * we can continue. */
530 531
        sdsfree(server.masterhost);
        server.masterhost = sdsdup(c->argv[1]->ptr);
532
        server.masterport = port;
533
        if (server.master) freeClient(server.master);
534
        disconnectSlaves(); /* Force our slaves to resync with us as well. */
A
antirez 已提交
535
        if (server.repl_state == REDIS_REPL_TRANSFER)
536
            replicationAbortSyncTransfer();
A
antirez 已提交
537
        server.repl_state = REDIS_REPL_CONNECT;
538 539 540 541 542
        redisLog(REDIS_NOTICE,"SLAVE OF %s:%d enabled (user request)",
            server.masterhost, server.masterport);
    }
    addReply(c,shared.ok);
}
543 544 545 546

/* --------------------------- REPLICATION CRON  ---------------------------- */

void replicationCron(void) {
547
    /* Non blocking connection timeout? */
A
antirez 已提交
548
    if (server.masterhost && server.repl_state == REDIS_REPL_CONNECTING &&
549 550 551 552 553 554
        (time(NULL)-server.repl_transfer_lastio) > server.repl_timeout)
    {
        redisLog(REDIS_WARNING,"Timeout connecting to the MASTER...");
        undoConnectWithMaster();
    }

555
    /* Bulk transfer I/O timeout? */
A
antirez 已提交
556
    if (server.masterhost && server.repl_state == REDIS_REPL_TRANSFER &&
A
7c6da73  
antirez 已提交
557
        (time(NULL)-server.repl_transfer_lastio) > server.repl_timeout)
558 559 560 561 562
    {
        redisLog(REDIS_WARNING,"Timeout receiving bulk data from MASTER...");
        replicationAbortSyncTransfer();
    }

563
    /* Timed out master when we are an already connected slave? */
A
antirez 已提交
564
    if (server.masterhost && server.repl_state == REDIS_REPL_CONNECTED &&
A
7c6da73  
antirez 已提交
565
        (time(NULL)-server.master->lastinteraction) > server.repl_timeout)
566 567 568 569 570
    {
        redisLog(REDIS_WARNING,"MASTER time out: no data nor PING received...");
        freeClient(server.master);
    }

571
    /* Check if we should connect to a MASTER */
A
antirez 已提交
572
    if (server.repl_state == REDIS_REPL_CONNECT) {
573
        redisLog(REDIS_NOTICE,"Connecting to MASTER...");
574 575
        if (connectWithMaster() == REDIS_OK) {
            redisLog(REDIS_NOTICE,"MASTER <-> SLAVE sync started");
576 577
        }
    }
578 579 580 581 582
    
    /* If we have attached slaves, PING them from time to time.
     * So slaves can implement an explicit timeout to masters, and will
     * be able to detect a link disconnection even if the TCP connection
     * will not actually go down. */
A
7c6da73  
antirez 已提交
583
    if (!(server.cronloops % (server.repl_ping_slave_period*10))) {
584 585 586 587 588 589 590 591 592 593 594 595
        listIter li;
        listNode *ln;

        listRewind(server.slaves,&li);
        while((ln = listNext(&li))) {
            redisClient *slave = ln->value;

            /* Don't ping slaves that are in the middle of a bulk transfer
             * with the master for first synchronization. */
            if (slave->replstate == REDIS_REPL_SEND_BULK) continue;
            if (slave->replstate == REDIS_REPL_ONLINE) {
                /* If the slave is online send a normal ping */
596
                addReplySds(slave,sdsnew("*1\r\n$4\r\nPING\r\n"));
597 598 599 600 601 602
            } else {
                /* Otherwise we are in the pre-synchronization stage.
                 * Just a newline will do the work of refreshing the
                 * connection last interaction time, and at the same time
                 * we'll be sure that being a single char there are no
                 * short-write problems. */
A
antirez 已提交
603 604 605
                if (write(slave->fd, "\n", 1) == -1) {
                    /* Don't worry, it's just a ping. */
                }
606 607 608
            }
        }
    }
609
}