AOF设置成为always时，数据绝对不会丢失吗？

上次在群里讨论了aof设置成为appendfsync always到底会不会丢失数据，那我们就从源码层面去看看redis的实现，看看到底是不是绝对不会丢失数据(redis-4.0.10)。

//filename ae.c    
typedef struct aeEventLoop {
    int maxfd;   /* highest file descriptor currently registered */
    int setsize; /* max number of file descriptors tracked */
    long long timeEventNextId;
    time_t lastTime;     /* Used to detect system clock skew */
    aeFileEvent *events; /* Registered events */
    aeFiredEvent *fired; /* Fired events */
    aeTimeEvent *timeEventHead;
    int stop;
    void *apidata; /* This is used for polling API specific data */
    aeBeforeSleepProc *beforesleep;
    aeBeforeSleepProc *aftersleep;
} aeEventLoop;

typedef void aeBeforeSleepProc(struct aeEventLoop *eventLoop);

//ae网络事件主程序，一个event loop
void aeMain(aeEventLoop *eventLoop) {
    eventLoop->stop = 0;
     //如果eventLoop中的stop标志位不为1，就循环处理
    while (!eventLoop->stop) {
        if (eventLoop->beforesleep != NULL)
            //处理事件时候的前置操作，处理回调函数
            //那么这个回调函数是在什么时候设置的呢？
            //后面会有
            eventLoop->beforesleep(eventLoop);
        aeProcessEvents(eventLoop, AE_ALL_EVENTS|AE_CALL_AFTER_SLEEP);
    }
}

//server.c

void beforeSleep(struct aeEventLoop *eventLoop) {
    UNUSED(eventLoop);

    if (server.cluster_enabled) clusterBeforeSleep();


    if (server.active_expire_enabled && server.masterhost == NULL)
        activeExpireCycle(ACTIVE_EXPIRE_CYCLE_FAST);

    if (server.get_ack_from_slaves) {
        robj *argv[3];

        argv[0] = createStringObject("REPLCONF",8);
        argv[1] = createStringObject("GETACK",6);
        argv[2] = createStringObject("*",1); /* Not used argument. */
        replicationFeedSlaves(server.slaves, server.slaveseldb, argv, 3);
        decrRefCount(argv[0]);
        decrRefCount(argv[1]);
        decrRefCount(argv[2]);
        server.get_ack_from_slaves = 0;
    }
    if (listLength(server.clients_waiting_acks))
        processClientsWaitingReplicas();
        
    moduleHandleBlockedClients();

    /* Try to process pending commands for clients that were just unblocked. */
    if (listLength(server.unblocked_clients))
        processUnblockedClients();
    //前置操作就会有调用flushAppendOnlyFile函数   
    flushAppendOnlyFile(0);

    /* Handle writes with pending output buffers. */
    handleClientsWithPendingWrites();

    /* Before we are going to sleep, let the threads access the dataset by
     * releasing the GIL. Redis main thread will not touch anything at this
     * time. */
    if (moduleCount()) moduleReleaseGIL();
}

int main(){
    .....
    //设置了事件的前置操作函数
    eSetBeforeSleepProc(server.el,beforeSleep);
    aeSetAfterSleepProc(server.el,afterSleep);
}


//filename aof.c
//进行aof的append
void flushAppendOnlyFile(int force) {
    ssize_t nwritten;
    int sync_in_progress = 0;
    mstime_t latency;
    //aof_buf区间为0的话，不进行增加操作
    if (sdslen(server.aof_buf) == 0) return;
     //start 以下是aof配置为 EVERYSEC的逻辑我们可以暂时不用看  
    if (server.aof_fsync == AOF_FSYNC_EVERYSEC)
        sync_in_progress = bioPendingJobsOfType(BIO_AOF_FSYNC) != 0;

    if (server.aof_fsync == AOF_FSYNC_EVERYSEC && !force) {
        /* With this append fsync policy we do background fsyncing.
         * If the fsync is still in progress we can try to delay
         * the write for a couple of seconds. */
        if (sync_in_progress) {
            if (server.aof_flush_postponed_start == 0) {
                /* No previous write postponing, remember that we are
                 * postponing the flush and return. */
                server.aof_flush_postponed_start = server.unixtime;
                return;
            } else if (server.unixtime - server.aof_flush_postponed_start < 2) {
                /* We were already waiting for fsync to finish, but for less
                 * than two seconds this is still ok. Postpone again. */
                return;
            }
            /* Otherwise fall trough, and go write since we can't wait
             * over two seconds. */
            server.aof_delayed_fsync++;
            serverLog(LL_NOTICE,"Asynchronous AOF fsync is taking too long (disk is busy?). Writing the AOF buffer without waiting for fsync to complete, this may slow down Redis.");
        }
    }
    //end 


    latencyStartMonitor(latency);
    //将aof_buf中的数据写入到aof中
    nwritten = aofWrite(server.aof_fd,server.aof_buf,sdslen(server.aof_buf));
    latencyEndMonitor(latency);
    if (sync_in_progress) {
        latencyAddSampleIfNeeded("aof-write-pending-fsync",latency);
    } else if (server.aof_child_pid != -1 || server.rdb_child_pid != -1) {
        latencyAddSampleIfNeeded("aof-write-active-child",latency);
    } else {
        latencyAddSampleIfNeeded("aof-write-alone",latency);
    }
    latencyAddSampleIfNeeded("aof-write",latency);
    //上面latency开头的函数都是对一些延时操作进行统计
    //类似与性能统计吧，主要是会对一些IO操作或者fork操作进行统计
    //有兴趣的同学可以仔细研究用处哈，这里不做过多介绍
   
    server.aof_flush_postponed_start = 0;
    //一些错位的处理
    if (nwritten != (ssize_t)sdslen(server.aof_buf)) {
        static time_t last_write_error_log = 0;
        int can_log = 0;

        /* Limit logging rate to 1 line per AOF_WRITE_LOG_ERROR_RATE seconds. */
        if ((server.unixtime - last_write_error_log) > AOF_WRITE_LOG_ERROR_RATE) {
            can_log = 1;
            last_write_error_log = server.unixtime;
        }

        /* Log the AOF write error and record the error code. */
        if (nwritten == -1) {
            if (can_log) {
                serverLog(LL_WARNING,"Error writing to the AOF file: %s",
                    strerror(errno));
                server.aof_last_write_errno = errno;
            }
        } else {
            if (can_log) {
                serverLog(LL_WARNING,"Short write while writing to "
                                       "the AOF file: (nwritten=%lld, "
                                       "expected=%lld)",
                                       (long long)nwritten,
                                       (long long)sdslen(server.aof_buf));
            }

            if (ftruncate(server.aof_fd, server.aof_current_size) == -1) {
                if (can_log) {
                    serverLog(LL_WARNING, "Could not remove short write "
                             "from the append-only file.  Redis may refuse "
                             "to load the AOF the next time it starts.  "
                             "ftruncate: %s", strerror(errno));
                }
            } else {
                /* If the ftruncate() succeeded we can set nwritten to
                 * -1 since there is no longer partial data into the AOF. */
                nwritten = -1;
            }
            server.aof_last_write_errno = ENOSPC;
        }

        /* Handle the AOF write error. */
        if (server.aof_fsync == AOF_FSYNC_ALWAYS) {
            /* We can't recover when the fsync policy is ALWAYS since the
             * reply for the client is already in the output buffers, and we
             * have the contract with the user that on acknowledged write data
             * is synced on disk. */
            serverLog(LL_WARNING,"Can't recover from AOF write error when the AOF fsync policy is 'always'. Exiting...");
            exit(1);
        } else {
            /* Recover from failed write leaving data into the buffer. However
             * set an error to stop accepting writes as long as the error
             * condition is not cleared. */
            server.aof_last_write_status = C_ERR;

            /* Trim the sds buffer if there was a partial write, and there
             * was no way to undo it with ftruncate(2). */
            if (nwritten > 0) {
                server.aof_current_size += nwritten;
                sdsrange(server.aof_buf,nwritten,-1);
            }
            return; /* We'll try again on the next call... */
        }
    } else {
        /* Successful write(2). If AOF was in error state, restore the
         * OK state and log the event. */
        if (server.aof_last_write_status == C_ERR) {
            serverLog(LL_WARNING,
                "AOF write error looks solved, Redis can write again.");
            server.aof_last_write_status = C_OK;
        }
    }
    server.aof_current_size += nwritten;
    
    //清空aofbuf，如果aofbuf较小时，复用之，大了的话free掉，重新申请
       //这样对性能有好处吗？？我暂时没有想明白为什么
    if ((sdslen(server.aof_buf)+sdsavail(server.aof_buf)) < 4000) {
        sdsclear(server.aof_buf);
    } else {
        sdsfree(server.aof_buf);
        server.aof_buf = sdsempty();
    }

      //aof正在进行rewrite或者说有rdb的快照同步的话不进行fsync操作
    if (server.aof_no_fsync_on_rewrite &&
        (server.aof_child_pid != -1 || server.rdb_child_pid != -1))
            return;
    //进行fsync操作，上面只是写入缓冲区了，并没有真正意义上的flush到disk
    if (server.aof_fsync == AOF_FSYNC_ALWAYS) {
        /* aof_fsync is defined as fdatasync() for Linux in order to avoid
         * flushing metadata. */
        latencyStartMonitor(latency);
        aof_fsync(server.aof_fd); /* Let's try to get this data on the disk */
        latencyEndMonitor(latency);
        latencyAddSampleIfNeeded("aof-fsync-always",latency);
        server.aof_last_fsync = server.unixtime;
    } else if ((server.aof_fsync == AOF_FSYNC_EVERYSEC &&
                server.unixtime > server.aof_last_fsync)) {
        if (!sync_in_progress) aof_background_fsync(server.aof_fd);
        server.aof_last_fsync = server.unixtime;
    }
}

以上我们进行了初步的一些源码分析，细心的朋友可能会发现我们是在什么时候写入到aof_buf内存的呢，不要紧，我们继续上源码。

//filename server.c
//执行客户端命令的过程
void call(client *c, int flags) {
    long long dirty, start, duration;
    int client_old_flags = c->flags;
    if (listLength(server.monitors) &&
        !server.loading &&
        !(c->cmd->flags & (CMD_SKIP_MONITOR|CMD_ADMIN)))
    {
        replicationFeedMonitors(c,server.monitors,c->db->id,c->argv,c->argc);
    }
    c->flags &= ~(CLIENT_FORCE_AOF|CLIENT_FORCE_REPL|CLIENT_PREVENT_PROP);
    redisOpArray prev_also_propagate = server.also_propagate;
    redisOpArrayInit(&server.also_propagate);

    /* Call the command. */
    dirty = server.dirty;
    start = ustime();
    c->cmd->proc(c);
    duration = ustime()-start;
    dirty = server.dirty-dirty;
    if (dirty < 0) dirty = 0;

    if (server.loading && c->flags & CLIENT_LUA)
        flags &= ~(CMD_CALL_SLOWLOG | CMD_CALL_STATS);

    if (c->flags & CLIENT_LUA && server.lua_caller) {
        if (c->flags & CLIENT_FORCE_REPL)
            server.lua_caller->flags |= CLIENT_FORCE_REPL;
        if (c->flags & CLIENT_FORCE_AOF)
            server.lua_caller->flags |= CLIENT_FORCE_AOF;
    }

    if (flags & CMD_CALL_SLOWLOG && c->cmd->proc != execCommand) {
        char *latency_event = (c->cmd->flags & CMD_FAST) ?
                              "fast-command" : "command";
        latencyAddSampleIfNeeded(latency_event,duration/1000);
        slowlogPushEntryIfNeeded(c,c->argv,c->argc,duration);
    }
    if (flags & CMD_CALL_STATS) {
        c->lastcmd->microseconds += duration;
        c->lastcmd->calls++;
    }

    if (flags & CMD_CALL_PROPAGATE &&
        (c->flags & CLIENT_PREVENT_PROP) != CLIENT_PREVENT_PROP)
    {
        int propagate_flags = PROPAGATE_NONE;

        /* Check if the command operated changes in the data set. If so
         * set for replication / AOF propagation. */
        if (dirty) propagate_flags |= (PROPAGATE_AOF|PROPAGATE_REPL);

        /* If the client forced AOF / replication of the command, set
         * the flags regardless of the command effects on the data set. */
        if (c->flags & CLIENT_FORCE_REPL) propagate_flags |= PROPAGATE_REPL;
        if (c->flags & CLIENT_FORCE_AOF) propagate_flags |= PROPAGATE_AOF;

        /* However prevent AOF / replication propagation if the command
         * implementatino called preventCommandPropagation() or similar,
         * or if we don't have the call() flags to do so. */
        if (c->flags & CLIENT_PREVENT_REPL_PROP ||
            !(flags & CMD_CALL_PROPAGATE_REPL))
                propagate_flags &= ~PROPAGATE_REPL;
        if (c->flags & CLIENT_PREVENT_AOF_PROP ||
            !(flags & CMD_CALL_PROPAGATE_AOF))
                propagate_flags &= ~PROPAGATE_AOF;
        //这里是重点哦，如果设置了aof就会调用propagate
        if (propagate_flags != PROPAGATE_NONE && !(c->cmd->flags & CMD_MODULE))
            propagate(c->cmd,c->db->id,c->argv,c->argc,propagate_flags);
    }

    /* Restore the old replication flags, since call() can be executed
     * recursively. */
    c->flags &= ~(CLIENT_FORCE_AOF|CLIENT_FORCE_REPL|CLIENT_PREVENT_PROP);
    c->flags |= client_old_flags &
        (CLIENT_FORCE_AOF|CLIENT_FORCE_REPL|CLIENT_PREVENT_PROP);

    /* Handle the alsoPropagate() API to handle commands that want to propagate
     * multiple separated commands. Note that alsoPropagate() is not affected
     * by CLIENT_PREVENT_PROP flag. */
    if (server.also_propagate.numops) {
        int j;
        redisOp *rop;

        if (flags & CMD_CALL_PROPAGATE) {
            for (j = 0; j < server.also_propagate.numops; j++) {
                rop = &server.also_propagate.ops[j];
                int target = rop->target;
                /* Whatever the command wish is, we honor the call() flags. */
                if (!(flags&CMD_CALL_PROPAGATE_AOF)) target &= ~PROPAGATE_AOF;
                if (!(flags&CMD_CALL_PROPAGATE_REPL)) target &= ~PROPAGATE_REPL;
                if (target)
                    propagate(rop->cmd,rop->dbid,rop->argv,rop->argc,target);
            }
        }
        redisOpArrayFree(&server.also_propagate);
    }
    server.also_propagate = prev_also_propagate;
    server.stat_numcommands++;
}

void propagate(struct redisCommand *cmd, int dbid, robj **argv, int argc,
               int flags)
{
    //aof开启调用feedAppendOnlyFile函数
    if (server.aof_state != AOF_OFF && flags & PROPAGATE_AOF)
        //重点
        feedAppendOnlyFile(cmd,dbid,argv,argc);
    if (flags & PROPAGATE_REPL)
        replicationFeedSlaves(server.slaves,dbid,argv,argc);
}


void feedAppendOnlyFile(struct redisCommand *cmd, int dictid, robj **argv, int argc) {
    sds buf = sdsempty();
    robj *tmpargv[3];

    /* The DB this command was targeting is not the same as the last command
     * we appended. To issue a SELECT command is needed. */
    if (dictid != server.aof_selected_db) {
        char seldb[64];

        snprintf(seldb,sizeof(seldb),"%d",dictid);
        buf = sdscatprintf(buf,"*2\r\n$6\r\nSELECT\r\n$%lu\r\n%s\r\n",
            (unsigned long)strlen(seldb),seldb);
        server.aof_selected_db = dictid;
    }

    if (cmd->proc == expireCommand || cmd->proc == pexpireCommand ||
        cmd->proc == expireatCommand) {
        buf = catAppendOnlyExpireAtCommand(buf,cmd,argv[1],argv[2]);
    } else if (cmd->proc == setexCommand || cmd->proc == psetexCommand) {
        /* Translate SETEX/PSETEX to SET and PEXPIREAT */
        tmpargv[0] = createStringObject("SET",3);
        tmpargv[1] = argv[1];
        tmpargv[2] = argv[3];
        buf = catAppendOnlyGenericCommand(buf,3,tmpargv);
        decrRefCount(tmpargv[0]);
        buf = catAppendOnlyExpireAtCommand(buf,cmd,argv[1],argv[2]);
    } else if (cmd->proc == setCommand && argc > 3) {
        int i;
        robj *exarg = NULL, *pxarg = NULL;
        /* Translate SET [EX seconds][PX milliseconds] to SET and PEXPIREAT */
        buf = catAppendOnlyGenericCommand(buf,3,argv);
        for (i = 3; i < argc; i ++) {
            if (!strcasecmp(argv[i]->ptr, "ex")) exarg = argv[i+1];
            if (!strcasecmp(argv[i]->ptr, "px")) pxarg = argv[i+1];
        }
        serverAssert(!(exarg && pxarg));
        if (exarg)
            buf = catAppendOnlyExpireAtCommand(buf,server.expireCommand,argv[1],
                                               exarg);
        if (pxarg)
            buf = catAppendOnlyExpireAtCommand(buf,server.pexpireCommand,argv[1],
                                               pxarg);
    } else {

        buf = catAppendOnlyGenericCommand(buf,argc,argv);
    }

    //aof打开写入aof当中
    if (server.aof_state == AOF_ON)
        //sdscatlen本质就是memcpy，具体可见filename sds.c
        server.aof_buf = sdscatlen(server.aof_buf,buf,sdslen(buf));


    if (server.aof_child_pid != -1)
        aofRewriteBufferAppend((unsigned char*)buf,sdslen(buf));

    sdsfree(buf);
}

那么从上面的分析得出，其实我们每次执行客户端命令的时候操作并没有写到aof文件中，只是写到了aof_buf内存当中，当进行下一个事件循环的时候执行beforeSleep之时，才会去fsync到disk中，从redis的这种策略上我们也可以看出，redis和mysql在数据持久化之间的区别，redis的数据持久化仅仅就是一个附带功能，并不是其主要功能，但是mysql就不一样，mysql的在数据坚持久化方面就是刚需，最终目的就是让数据成功落盘，不会有任何的丢失和数据一致性问题，因此mysql（innodb存储引擎）在做事务提交的时候会有两阶段提交、日志立即刷盘（当然这和sync_binlog和innodb_flush_log_at_trx_commit配置有关）来确保数据的成功提交，不会丢失。下次如果面试被问到持久化了，就可以这么和面试官愉快的吹水了，分析就到此结束啦。