X-Git-Url: https://git.saurik.com/redis.git/blobdiff_plain/2b2eca1f5643260ae7ca6bec70c263ff0db5a8e3..a1b1c1ea3adb864c8c2e3feb4eaed890a9b505cf:/src/replication.c
diff --git a/src/replication.c b/src/replication.c
index 9f8d9274..c1e46191 100644
--- a/src/replication.c
+++ b/src/replication.c
@@ -3,6 +3,7 @@
#include <sys/time.h>
#include <unistd.h>
#include <fcntl.h>
+#include <sys/socket.h>
#include <sys/stat.h>
/* ---------------------------------- MASTER -------------------------------- */
@@ -10,38 +11,8 @@
void replicationFeedSlaves(list *slaves, int dictid, robj **argv, int argc) {
listNode *ln;
listIter li;
- int outc = 0, j;
- robj **outv;
- /* We need 1+(ARGS*3) objects since commands are using the new protocol
- * and we one 1 object for the first "*<count>\r\n" multibulk count, then
- * for every additional object we have "$<count>\r\n" + object + "\r\n". */
- robj *static_outv[REDIS_STATIC_ARGS*3+1];
- robj *lenobj;
-
- if (argc <= REDIS_STATIC_ARGS) {
- outv = static_outv;
- } else {
- outv = zmalloc(sizeof(robj*)*(argc*3+1));
- }
+ int j;
- lenobj = createObject(REDIS_STRING,
- sdscatprintf(sdsempty(), "*%d\r\n", argc));
- lenobj->refcount = 0;
- outv[outc++] = lenobj;
- for (j = 0; j < argc; j++) {
- lenobj = createObject(REDIS_STRING,
- sdscatprintf(sdsempty(),"$%lu\r\n",
- (unsigned long) stringObjectLen(argv[j])));
- lenobj->refcount = 0;
- outv[outc++] = lenobj;
- outv[outc++] = argv[j];
- outv[outc++] = shared.crlf;
- }
-
- /* Increment all the refcounts at start and decrement at end in order to
- * be sure to free objects if there is no slave in a replication state
- * able to be feed with commands */
- for (j = 0; j < outc; j++) incrRefCount(outv[j]);
listRewind(slaves,&li);
while((ln = listNext(&li))) {
redisClient *slave = ln->value;
@@ -49,47 +20,45 @@ void replicationFeedSlaves(list *slaves, int dictid, robj **argv, int argc) {
/* Don't feed slaves that are still waiting for BGSAVE to start */
if (slave->replstate == REDIS_REPL_WAIT_BGSAVE_START) continue;
- /* Feed all the other slaves, MONITORs and so on */
+ /* 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. */
if (slave->slaveseldb != dictid) {
robj *selectcmd;
- switch(dictid) {
- case 0: selectcmd = shared.select0; break;
- case 1: selectcmd = shared.select1; break;
- case 2: selectcmd = shared.select2; break;
- case 3: selectcmd = shared.select3; break;
- case 4: selectcmd = shared.select4; break;
- case 5: selectcmd = shared.select5; break;
- case 6: selectcmd = shared.select6; break;
- case 7: selectcmd = shared.select7; break;
- case 8: selectcmd = shared.select8; break;
- case 9: selectcmd = shared.select9; break;
- default:
+ if (dictid >= 0 && dictid < REDIS_SHARED_SELECT_CMDS) {
+ selectcmd = shared.select[dictid];
+ incrRefCount(selectcmd);
+ } else {
selectcmd = createObject(REDIS_STRING,
sdscatprintf(sdsempty(),"select %d\r\n",dictid));
- selectcmd->refcount = 0;
- break;
}
addReply(slave,selectcmd);
+ decrRefCount(selectcmd);
slave->slaveseldb = dictid;
}
- for (j = 0; j < outc; j++) addReply(slave,outv[j]);
+ addReplyMultiBulkLen(slave,argc);
+ for (j = 0; j < argc; j++) addReplyBulk(slave,argv[j]);
}
- for (j = 0; j < outc; j++) decrRefCount(outv[j]);
- if (outv != static_outv) zfree(outv);
}
-void replicationFeedMonitors(list *monitors, int dictid, robj **argv, int argc) {
+void replicationFeedMonitors(redisClient *c, list *monitors, int dictid, robj **argv, int argc) {
listNode *ln;
listIter li;
- int j;
+ int j, port;
sds cmdrepr = sdsnew("+");
robj *cmdobj;
+ char ip[32];
struct timeval tv;
gettimeofday(&tv,NULL);
cmdrepr = sdscatprintf(cmdrepr,"%ld.%06ld ",(long)tv.tv_sec,(long)tv.tv_usec);
- if (dictid != 0) cmdrepr = sdscatprintf(cmdrepr,"(db %d) ", dictid);
+ 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);
+ }
for (j = 0; j < argc; j++) {
if (argv[j]->encoding == REDIS_ENCODING_INT) {
@@ -118,7 +87,7 @@ void syncCommand(redisClient *c) {
/* Refuse SYNC requests if we are a slave but the link with our master
* is not ok... */
- if (server.masterhost && server.replstate != REDIS_REPL_CONNECTED) {
+ if (server.masterhost && server.repl_state != REDIS_REPL_CONNECTED) {
addReplyError(c,"Can't SYNC while not connected with my master");
return;
}
@@ -135,7 +104,7 @@ void syncCommand(redisClient *c) {
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 */
- if (server.bgsavechildpid != -1) {
+ if (server.rdb_child_pid != -1) {
/* 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 */
@@ -151,8 +120,7 @@ void syncCommand(redisClient *c) {
if (ln) {
/* Perfect, the server is already registering differences for
* another slave. Set the right state, and copy the buffer. */
- listRelease(c->reply);
- c->reply = listDup(slave->reply);
+ copyClientOutputBuffer(c,slave);
c->replstate = REDIS_REPL_WAIT_BGSAVE_END;
redisLog(REDIS_NOTICE,"Waiting for end of BGSAVE for SYNC");
} else {
@@ -164,7 +132,7 @@ void syncCommand(redisClient *c) {
} else {
/* Ok we don't have a BGSAVE in progress, let's start one */
redisLog(REDIS_NOTICE,"Starting BGSAVE for SYNC");
- if (rdbSaveBackground(server.dbfilename) != REDIS_OK) {
+ if (rdbSaveBackground(server.rdb_filename) != REDIS_OK) {
redisLog(REDIS_NOTICE,"Replication failed, can't BGSAVE");
addReplyError(c,"Unable to perform background save");
return;
@@ -178,6 +146,46 @@ void syncCommand(redisClient *c) {
return;
}
+/* REPLCONF <option> <value> <option> <value> ...
+ * This command is used by a slave in order to configure the replication
+ * process before starting it with the SYNC command.
+ *
+ * Currently the only use of this command is to communicate to the master
+ * what is the listening port of the Slave redis instance, so that the
+ * master can accurately list slaves and their listening ports in
+ * the INFO output.
+ *
+ * In the future the same command can be used in order to configure
+ * the replication to initiate an incremental replication instead of a
+ * full resync. */
+void replconfCommand(redisClient *c) {
+ int j;
+
+ if ((c->argc % 2) == 0) {
+ /* Number of arguments must be odd to make sure that every
+ * option has a corresponding value. */
+ addReply(c,shared.syntaxerr);
+ return;
+ }
+
+ /* Process every option-value pair. */
+ for (j = 1; j < c->argc; j+=2) {
+ if (!strcasecmp(c->argv[j]->ptr,"listening-port")) {
+ long port;
+
+ if ((getLongFromObjectOrReply(c,c->argv[j+1],
+ &port,NULL) != REDIS_OK))
+ return;
+ c->slave_listening_port = port;
+ } else {
+ addReplyErrorFormat(c,"Unrecognized REPLCONF option: %s",
+ (char*)c->argv[j]->ptr);
+ return;
+ }
+ }
+ addReply(c,shared.ok);
+}
+
void sendBulkToSlave(aeEventLoop *el, int fd, void *privdata, int mask) {
redisClient *slave = privdata;
REDIS_NOTUSED(el);
@@ -227,7 +235,6 @@ void sendBulkToSlave(aeEventLoop *el, int fd, void *privdata, int mask) {
freeClient(slave);
return;
}
- addReplySds(slave,sdsempty());
redisLog(REDIS_NOTICE,"Synchronization with slave succeeded");
}
}
@@ -258,7 +265,7 @@ void updateSlavesWaitingBgsave(int bgsaveerr) {
redisLog(REDIS_WARNING,"SYNC failed. BGSAVE child returned an error");
continue;
}
- if ((slave->repldbfd = open(server.dbfilename,O_RDONLY)) == -1 ||
+ if ((slave->repldbfd = open(server.rdb_filename,O_RDONLY)) == -1 ||
redis_fstat(slave->repldbfd,&buf) == -1) {
freeClient(slave);
redisLog(REDIS_WARNING,"SYNC failed. Can't open/stat DB after BGSAVE: %s", strerror(errno));
@@ -275,7 +282,7 @@ void updateSlavesWaitingBgsave(int bgsaveerr) {
}
}
if (startbgsave) {
- if (rdbSaveBackground(server.dbfilename) != REDIS_OK) {
+ if (rdbSaveBackground(server.rdb_filename) != REDIS_OK) {
listIter li;
listRewind(server.slaves,&li);
@@ -294,55 +301,61 @@ void updateSlavesWaitingBgsave(int bgsaveerr) {
/* Abort the async download of the bulk dataset while SYNC-ing with master */
void replicationAbortSyncTransfer(void) {
- redisAssert(server.replstate == REDIS_REPL_TRANSFER);
+ redisAssert(server.repl_state == REDIS_REPL_TRANSFER);
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);
- server.replstate = REDIS_REPL_CONNECT;
+ server.repl_state = REDIS_REPL_CONNECT;
}
/* Asynchronously read the SYNC payload we receive from a master */
+#define REPL_MAX_WRITTEN_BEFORE_FSYNC (1024*1024*8) /* 8 MB */
void readSyncBulkPayload(aeEventLoop *el, int fd, void *privdata, int mask) {
char buf[4096];
ssize_t nread, readlen;
+ off_t left;
REDIS_NOTUSED(el);
REDIS_NOTUSED(privdata);
REDIS_NOTUSED(mask);
- /* If repl_transfer_left == -1 we still have to read the bulk length
+ /* If repl_transfer_size == -1 we still have to read the bulk length
* from the master reply. */
- if (server.repl_transfer_left == -1) {
- if (syncReadLine(fd,buf,1024,3600) == -1) {
+ if (server.repl_transfer_size == -1) {
+ if (syncReadLine(fd,buf,1024,server.repl_syncio_timeout*1000) == -1) {
redisLog(REDIS_WARNING,
"I/O error reading bulk count from MASTER: %s",
strerror(errno));
- replicationAbortSyncTransfer();
- return;
+ goto error;
}
+
if (buf[0] == '-') {
redisLog(REDIS_WARNING,
"MASTER aborted replication with an error: %s",
buf+1);
- replicationAbortSyncTransfer();
+ goto error;
+ } 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. */
+ server.repl_transfer_lastio = server.unixtime;
return;
} 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?");
- replicationAbortSyncTransfer();
- return;
+ goto error;
}
- server.repl_transfer_left = strtol(buf+1,NULL,10);
+ server.repl_transfer_size = strtol(buf+1,NULL,10);
redisLog(REDIS_NOTICE,
"MASTER <-> SLAVE sync: receiving %ld bytes from master",
- server.repl_transfer_left);
+ server.repl_transfer_size);
return;
}
/* Read bulk data */
- readlen = (server.repl_transfer_left < (signed)sizeof(buf)) ?
- server.repl_transfer_left : (signed)sizeof(buf);
+ left = server.repl_transfer_size - server.repl_transfer_read;
+ readlen = (left < (signed)sizeof(buf)) ? left : (signed)sizeof(buf);
nread = read(fd,buf,readlen);
if (nread <= 0) {
redisLog(REDIS_WARNING,"I/O error trying to sync with MASTER: %s",
@@ -350,16 +363,29 @@ void readSyncBulkPayload(aeEventLoop *el, int fd, void *privdata, int mask) {
replicationAbortSyncTransfer();
return;
}
- server.repl_transfer_lastio = time(NULL);
+ server.repl_transfer_lastio = server.unixtime;
if (write(server.repl_transfer_fd,buf,nread) != nread) {
- redisLog(REDIS_WARNING,"Write error or short write writing to the DB dump file needed for MASTER <-> SLAVE synchrnonization: %s", strerror(errno));
- replicationAbortSyncTransfer();
- return;
+ redisLog(REDIS_WARNING,"Write error or short write writing to the DB dump file needed for MASTER <-> SLAVE synchronization: %s", strerror(errno));
+ goto error;
+ }
+ server.repl_transfer_read += nread;
+
+ /* Sync data on disk from time to time, otherwise at the end of the transfer
+ * we may suffer a big delay as the memory buffers are copied into the
+ * actual disk. */
+ if (server.repl_transfer_read >=
+ server.repl_transfer_last_fsync_off + REPL_MAX_WRITTEN_BEFORE_FSYNC)
+ {
+ off_t sync_size = server.repl_transfer_read -
+ server.repl_transfer_last_fsync_off;
+ rdb_fsync_range(server.repl_transfer_fd,
+ server.repl_transfer_last_fsync_off, sync_size);
+ server.repl_transfer_last_fsync_off += sync_size;
}
- server.repl_transfer_left -= nread;
+
/* Check if the transfer is now complete */
- if (server.repl_transfer_left == 0) {
- if (rename(server.repl_transfer_tmpfile,server.dbfilename) == -1) {
+ if (server.repl_transfer_read == server.repl_transfer_size) {
+ if (rename(server.repl_transfer_tmpfile,server.rdb_filename) == -1) {
redisLog(REDIS_WARNING,"Failed trying to rename the temp DB into dump.rdb in MASTER <-> SLAVE synchronization: %s", strerror(errno));
replicationAbortSyncTransfer();
return;
@@ -371,7 +397,7 @@ void readSyncBulkPayload(aeEventLoop *el, int fd, void *privdata, int mask) {
* 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);
- if (rdbLoad(server.dbfilename) != REDIS_OK) {
+ if (rdbLoad(server.rdb_filename) != REDIS_OK) {
redisLog(REDIS_WARNING,"Failed trying to load the MASTER synchronization DB from disk");
replicationAbortSyncTransfer();
return;
@@ -382,84 +408,258 @@ void readSyncBulkPayload(aeEventLoop *el, int fd, void *privdata, int mask) {
server.master = createClient(server.repl_transfer_s);
server.master->flags |= REDIS_MASTER;
server.master->authenticated = 1;
- server.replstate = REDIS_REPL_CONNECTED;
+ server.repl_state = REDIS_REPL_CONNECTED;
redisLog(REDIS_NOTICE, "MASTER <-> SLAVE sync: Finished with success");
+ /* 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. */
+ if (server.aof_state != REDIS_AOF_OFF) {
+ 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);
+ }
+ }
}
+
+ return;
+
+error:
+ replicationAbortSyncTransfer();
+ return;
}
-int syncWithMaster(void) {
- char buf[1024], tmpfile[256], authcmd[1024];
- int fd = anetTcpConnect(NULL,server.masterhost,server.masterport);
+/* Send a synchronous command to the master. Used to send AUTH and
+ * REPLCONF commands before starting the replication with SYNC.
+ *
+ * On success NULL is returned.
+ * On error an sds string describing the error is returned.
+ */
+char *sendSynchronousCommand(int fd, ...) {
+ va_list ap;
+ sds cmd = sdsempty();
+ char *arg, buf[256];
+
+ /* Create the command to send to the master, we use simple inline
+ * protocol for simplicity as currently we only send simple strings. */
+ va_start(ap,fd);
+ while(1) {
+ arg = va_arg(ap, char*);
+ if (arg == NULL) break;
+
+ if (sdslen(cmd) != 0) cmd = sdscatlen(cmd," ",1);
+ cmd = sdscat(cmd,arg);
+ }
+ cmd = sdscatlen(cmd,"\r\n",2);
+
+ /* Transfer command to the server. */
+ if (syncWrite(fd,cmd,sdslen(cmd),server.repl_syncio_timeout*1000) == -1) {
+ sdsfree(cmd);
+ return sdscatprintf(sdsempty(),"Writing to master: %s",
+ strerror(errno));
+ }
+ sdsfree(cmd);
+
+ /* Read the reply from the server. */
+ if (syncReadLine(fd,buf,sizeof(buf),server.repl_syncio_timeout*1000) == -1)
+ {
+ return sdscatprintf(sdsempty(),"Reading from master: %s",
+ strerror(errno));
+ }
+
+ /* Check for errors from the server. */
+ if (buf[0] != '+') {
+ return sdscatprintf(sdsempty(),"Error from master: %s", buf);
+ }
+
+ return NULL; /* No errors. */
+}
+
+void syncWithMaster(aeEventLoop *el, int fd, void *privdata, int mask) {
+ char tmpfile[256], *err;
int dfd, maxtries = 5;
+ int sockerr = 0;
+ socklen_t errlen = sizeof(sockerr);
+ REDIS_NOTUSED(el);
+ REDIS_NOTUSED(privdata);
+ REDIS_NOTUSED(mask);
- if (fd == -1) {
- redisLog(REDIS_WARNING,"Unable to connect to MASTER: %s",
- strerror(errno));
- return REDIS_ERR;
+ /* If this event fired after the user turned the instance into a master
+ * with SLAVEOF NO ONE we must just return ASAP. */
+ if (server.repl_state == REDIS_REPL_NONE) {
+ close(fd);
+ return;
+ }
+
+ /* Check for errors in the socket. */
+ if (getsockopt(fd, SOL_SOCKET, SO_ERROR, &sockerr, &errlen) == -1)
+ sockerr = errno;
+ if (sockerr) {
+ aeDeleteFileEvent(server.el,fd,AE_READABLE|AE_WRITABLE);
+ redisLog(REDIS_WARNING,"Error condition on socket for SYNC: %s",
+ strerror(sockerr));
+ goto error;
+ }
+
+ /* If we were connecting, it's time to send a non blocking PING, we want to
+ * make sure the master is able to reply before going into the actual
+ * replication process where we have long timeouts in the order of
+ * seconds (in the meantime the slave would block). */
+ if (server.repl_state == REDIS_REPL_CONNECTING) {
+ redisLog(REDIS_NOTICE,"Non blocking connect for SYNC fired the event.");
+ /* Delete the writable event so that the readable event remains
+ * registered and we can wait for the PONG reply. */
+ aeDeleteFileEvent(server.el,fd,AE_WRITABLE);
+ server.repl_state = REDIS_REPL_RECEIVE_PONG;
+ /* Send the PING, don't check for errors at all, we have the timeout
+ * that will take care about this. */
+ syncWrite(fd,"PING\r\n",6,100);
+ return;
+ }
+
+ /* Receive the PONG command. */
+ if (server.repl_state == REDIS_REPL_RECEIVE_PONG) {
+ char buf[1024];
+
+ /* Delete the readable event, we no longer need it now that there is
+ * the PING reply to read. */
+ aeDeleteFileEvent(server.el,fd,AE_READABLE);
+
+ /* Read the reply with explicit timeout. */
+ buf[0] = '\0';
+ if (syncReadLine(fd,buf,sizeof(buf),
+ server.repl_syncio_timeout*1000) == -1)
+ {
+ redisLog(REDIS_WARNING,
+ "I/O error reading PING reply from master: %s",
+ strerror(errno));
+ goto error;
+ }
+
+ /* We don't care about the reply, it can be +PONG or an error since
+ * the server requires AUTH. As long as it replies correctly, it's
+ * fine from our point of view. */
+ if (buf[0] != '-' && buf[0] != '+') {
+ redisLog(REDIS_WARNING,"Unexpected reply to PING from master.");
+ goto error;
+ } else {
+ redisLog(REDIS_NOTICE,
+ "Master replied to PING, replication can continue...");
+ }
}
/* AUTH with the master if required. */
if(server.masterauth) {
- snprintf(authcmd, 1024, "AUTH %s\r\n", server.masterauth);
- if (syncWrite(fd, authcmd, strlen(server.masterauth)+7, 5) == -1) {
- close(fd);
- redisLog(REDIS_WARNING,"Unable to AUTH to MASTER: %s",
- strerror(errno));
- return REDIS_ERR;
- }
- /* Read the AUTH result. */
- if (syncReadLine(fd,buf,1024,3600) == -1) {
- close(fd);
- redisLog(REDIS_WARNING,"I/O error reading auth result from MASTER: %s",
- strerror(errno));
- return REDIS_ERR;
+ err = sendSynchronousCommand(fd,"AUTH",server.masterauth,NULL);
+ if (err) {
+ redisLog(REDIS_WARNING,"Unable to AUTH to MASTER: %s",err);
+ sdsfree(err);
+ goto error;
}
- if (buf[0] != '+') {
- close(fd);
- redisLog(REDIS_WARNING,"Cannot AUTH to MASTER, is the masterauth password correct?");
- return REDIS_ERR;
+ }
+
+ /* Set the slave port, so that Master's INFO command can list the
+ * slave listening port correctly. */
+ {
+ sds port = sdsfromlonglong(server.port);
+ err = sendSynchronousCommand(fd,"REPLCONF","listening-port",port,
+ NULL);
+ sdsfree(port);
+ /* Ignore the error if any, not all the Redis versions support
+ * REPLCONF listening-port. */
+ if (err) {
+ redisLog(REDIS_NOTICE,"(non critical): Master does not understand REPLCONF listening-port: %s", err);
+ sdsfree(err);
}
}
/* Issue the SYNC command */
- if (syncWrite(fd,"SYNC \r\n",7,5) == -1) {
- close(fd);
+ if (syncWrite(fd,"SYNC\r\n",6,server.repl_syncio_timeout*1000) == -1) {
redisLog(REDIS_WARNING,"I/O error writing to MASTER: %s",
strerror(errno));
- return REDIS_ERR;
+ goto error;
}
/* Prepare a suitable temp file for bulk transfer */
while(maxtries--) {
snprintf(tmpfile,256,
- "temp-%d.%ld.rdb",(int)time(NULL),(long int)getpid());
+ "temp-%d.%ld.rdb",(int)server.unixtime,(long int)getpid());
dfd = open(tmpfile,O_CREAT|O_WRONLY|O_EXCL,0644);
if (dfd != -1) break;
sleep(1);
}
if (dfd == -1) {
- close(fd);
redisLog(REDIS_WARNING,"Opening the temp file needed for MASTER <-> SLAVE synchronization: %s",strerror(errno));
- return REDIS_ERR;
+ goto error;
}
/* Setup the non blocking download of the bulk file. */
- if (aeCreateFileEvent(server.el, fd, AE_READABLE, readSyncBulkPayload, NULL)
+ if (aeCreateFileEvent(server.el,fd, AE_READABLE,readSyncBulkPayload,NULL)
== AE_ERR)
+ {
+ redisLog(REDIS_WARNING,"Can't create readable event for SYNC");
+ goto error;
+ }
+
+ server.repl_state = REDIS_REPL_TRANSFER;
+ server.repl_transfer_size = -1;
+ server.repl_transfer_read = 0;
+ server.repl_transfer_last_fsync_off = 0;
+ server.repl_transfer_fd = dfd;
+ server.repl_transfer_lastio = server.unixtime;
+ server.repl_transfer_tmpfile = zstrdup(tmpfile);
+ return;
+
+error:
+ close(fd);
+ server.repl_transfer_s = -1;
+ server.repl_state = REDIS_REPL_CONNECT;
+ 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;
+ }
+
+ if (aeCreateFileEvent(server.el,fd,AE_READABLE|AE_WRITABLE,syncWithMaster,NULL) ==
+ AE_ERR)
{
close(fd);
redisLog(REDIS_WARNING,"Can't create readable event for SYNC");
return REDIS_ERR;
}
- server.replstate = REDIS_REPL_TRANSFER;
- server.repl_transfer_left = -1;
+
+ server.repl_transfer_lastio = server.unixtime;
server.repl_transfer_s = fd;
- server.repl_transfer_fd = dfd;
- server.repl_transfer_lastio = time(NULL);
- server.repl_transfer_tmpfile = zstrdup(tmpfile);
+ server.repl_state = REDIS_REPL_CONNECTING;
return REDIS_OK;
}
+/* 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;
+
+ redisAssert(server.repl_state == REDIS_REPL_CONNECTING ||
+ server.repl_state == REDIS_REPL_RECEIVE_PONG);
+ aeDeleteFileEvent(server.el,fd,AE_READABLE|AE_WRITABLE);
+ close(fd);
+ server.repl_transfer_s = -1;
+ server.repl_state = REDIS_REPL_CONNECT;
+}
+
void slaveofCommand(redisClient *c) {
if (!strcasecmp(c->argv[1]->ptr,"no") &&
!strcasecmp(c->argv[2]->ptr,"one")) {
@@ -467,19 +667,37 @@ void slaveofCommand(redisClient *c) {
sdsfree(server.masterhost);
server.masterhost = NULL;
if (server.master) freeClient(server.master);
- if (server.replstate == REDIS_REPL_TRANSFER)
+ if (server.repl_state == REDIS_REPL_TRANSFER)
replicationAbortSyncTransfer();
- server.replstate = REDIS_REPL_NONE;
+ else if (server.repl_state == REDIS_REPL_CONNECTING ||
+ server.repl_state == REDIS_REPL_RECEIVE_PONG)
+ undoConnectWithMaster();
+ server.repl_state = REDIS_REPL_NONE;
redisLog(REDIS_NOTICE,"MASTER MODE enabled (user request)");
}
} else {
+ 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. */
sdsfree(server.masterhost);
server.masterhost = sdsdup(c->argv[1]->ptr);
- server.masterport = atoi(c->argv[2]->ptr);
+ server.masterport = port;
if (server.master) freeClient(server.master);
- if (server.replstate == REDIS_REPL_TRANSFER)
+ disconnectSlaves(); /* Force our slaves to resync with us as well. */
+ if (server.repl_state == REDIS_REPL_TRANSFER)
replicationAbortSyncTransfer();
- server.replstate = REDIS_REPL_CONNECT;
+ server.repl_state = REDIS_REPL_CONNECT;
redisLog(REDIS_NOTICE,"SLAVE OF %s:%d enabled (user request)",
server.masterhost, server.masterport);
}
@@ -488,23 +706,69 @@ void slaveofCommand(redisClient *c) {
/* --------------------------- REPLICATION CRON ---------------------------- */
-#define REDIS_REPL_TRANSFER_TIMEOUT 60
-
void replicationCron(void) {
+ /* Non blocking connection timeout? */
+ if (server.masterhost &&
+ (server.repl_state == REDIS_REPL_CONNECTING ||
+ server.repl_state == REDIS_REPL_RECEIVE_PONG) &&
+ (time(NULL)-server.repl_transfer_lastio) > server.repl_timeout)
+ {
+ redisLog(REDIS_WARNING,"Timeout connecting to the MASTER...");
+ undoConnectWithMaster();
+ }
+
/* Bulk transfer I/O timeout? */
- if (server.masterhost && server.replstate == REDIS_REPL_TRANSFER &&
- (time(NULL)-server.repl_transfer_lastio) > REDIS_REPL_TRANSFER_TIMEOUT)
+ if (server.masterhost && server.repl_state == REDIS_REPL_TRANSFER &&
+ (time(NULL)-server.repl_transfer_lastio) > server.repl_timeout)
{
- redisLog(REDIS_WARNING,"Timeout receiving bulk data from MASTER...");
+ redisLog(REDIS_WARNING,"Timeout receiving bulk data from MASTER... If the problem persists try to set the 'repl-timeout' parameter in redis.conf to a larger value.");
replicationAbortSyncTransfer();
}
+ /* Timed out master when we are an already connected slave? */
+ if (server.masterhost && server.repl_state == REDIS_REPL_CONNECTED &&
+ (time(NULL)-server.master->lastinteraction) > server.repl_timeout)
+ {
+ redisLog(REDIS_WARNING,"MASTER time out: no data nor PING received...");
+ freeClient(server.master);
+ }
+
/* Check if we should connect to a MASTER */
- if (server.replstate == REDIS_REPL_CONNECT) {
+ if (server.repl_state == REDIS_REPL_CONNECT) {
redisLog(REDIS_NOTICE,"Connecting to MASTER...");
- if (syncWithMaster() == REDIS_OK) {
- redisLog(REDIS_NOTICE,"MASTER <-> SLAVE sync started: SYNC sent");
- if (server.appendonly) rewriteAppendOnlyFileBackground();
+ if (connectWithMaster() == REDIS_OK) {
+ redisLog(REDIS_NOTICE,"MASTER <-> SLAVE sync started");
+ }
+ }
+
+ /* 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. */
+ if (!(server.cronloops % (server.repl_ping_slave_period * REDIS_HZ))) {
+ 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 */
+ addReplySds(slave,sdsnew("*1\r\n$4\r\nPING\r\n"));
+ } 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. */
+ if (write(slave->fd, "\n", 1) == -1) {
+ /* Don't worry, it's just a ping. */
+ }
+ }
}
}
}