X-Git-Url: https://git.saurik.com/redis.git/blobdiff_plain/e0ba14557e2b24d21e92be01afd2307f1cc57aac..3b5289a04c474ce66df2ef410e053795b1f0f1d2:/src/aof.c diff --git a/src/aof.c b/src/aof.c index 7b15acb8..0dcf905e 100644 --- a/src/aof.c +++ b/src/aof.c @@ -1,5 +1,6 @@ #include "redis.h" #include "bio.h" +#include "rio.h" #include #include @@ -11,10 +12,14 @@ void aofUpdateCurrentSize(void); +void aof_background_fsync(int fd) { + bioCreateBackgroundJob(REDIS_BIO_AOF_FSYNC,(void*)(long)fd,NULL,NULL); +} + /* Called when the user switches from "appendonly yes" to "appendonly no" * at runtime using the CONFIG command. */ void stopAppendOnly(void) { - flushAppendOnlyFile(); + flushAppendOnlyFile(1); aof_fsync(server.appendfd); close(server.appendfd); @@ -59,12 +64,51 @@ int startAppendOnly(void) { * and the only way the client socket can get a write is entering when the * the event loop, we accumulate all the AOF writes in a memory * buffer and write it on disk using this function just before entering - * the event loop again. */ -void flushAppendOnlyFile(void) { + * the event loop again. + * + * About the 'force' argument: + * + * When the fsync policy is set to 'everysec' we may delay the flush if there + * is still an fsync() going on in the background thread, since for instance + * on Linux write(2) will be blocked by the background fsync anyway. + * When this happens we remember that there is some aof buffer to be + * flushed ASAP, and will try to do that in the serverCron() function. + * + * However if force is set to 1 we'll write regardless of the background + * fsync. */ +void flushAppendOnlyFile(int force) { ssize_t nwritten; + int sync_in_progress = 0; if (sdslen(server.aofbuf) == 0) return; + if (server.appendfsync == APPENDFSYNC_EVERYSEC) + sync_in_progress = bioPendingJobsOfType(REDIS_BIO_AOF_FSYNC) != 0; + + if (server.appendfsync == APPENDFSYNC_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 postponinig, 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. */ + redisLog(REDIS_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."); + } + } + /* If you are following this code path, then we are going to write so + * set reset the postponed flush sentinel to zero. */ + server.aof_flush_postponed_start = 0; + /* We want to perform a single write. This should be guaranteed atomic * at least if the filesystem we are writing is a real physical one. * While this will save us against the server being killed I don't think @@ -100,14 +144,15 @@ void flushAppendOnlyFile(void) { return; /* Perform the fsync if needed. */ - if (server.appendfsync == APPENDFSYNC_ALWAYS || - (server.appendfsync == APPENDFSYNC_EVERYSEC && - server.unixtime > server.lastfsync)) - { + if (server.appendfsync == APPENDFSYNC_ALWAYS) { /* aof_fsync is defined as fdatasync() for Linux in order to avoid * flushing metadata. */ aof_fsync(server.appendfd); /* Let's try to get this data on the disk */ server.lastfsync = server.unixtime; + } else if ((server.appendfsync == APPENDFSYNC_EVERYSEC && + server.unixtime > server.lastfsync)) { + if (!sync_in_progress) aof_background_fsync(server.appendfd); + server.lastfsync = server.unixtime; } } @@ -282,6 +327,8 @@ int loadAppendOnlyFile(char *filename) { } if (buf[0] != '*') goto fmterr; argc = atoi(buf+1); + if (argc < 1) goto fmterr; + argv = zmalloc(sizeof(robj*)*argc); for (j = 0; j < argc; j++) { if (fgets(buf,sizeof(buf),fp) == NULL) goto readerr; @@ -340,11 +387,26 @@ fmterr: exit(1); } +/* Delegate writing an object to writing a bulk string or bulk long long. + * This is not placed in rio.c since that adds the redis.h dependency. */ +int rioWriteBulkObject(rio *r, robj *obj) { + /* Avoid using getDecodedObject to help copy-on-write (we are often + * in a child process when this function is called). */ + if (obj->encoding == REDIS_ENCODING_INT) { + return rioWriteBulkLongLong(r,(long)obj->ptr); + } else if (obj->encoding == REDIS_ENCODING_RAW) { + return rioWriteBulkString(r,obj->ptr,sdslen(obj->ptr)); + } else { + redisPanic("Unknown string encoding"); + } +} + /* Write a sequence of commands able to fully rebuild the dataset into * "filename". Used both by REWRITEAOF and BGREWRITEAOF. */ int rewriteAppendOnlyFile(char *filename) { dictIterator *di = NULL; dictEntry *de; + rio aof; FILE *fp; char tmpfile[256]; int j; @@ -358,6 +420,8 @@ int rewriteAppendOnlyFile(char *filename) { redisLog(REDIS_WARNING, "Failed rewriting the append only file: %s", strerror(errno)); return REDIS_ERR; } + + rioInitWithFile(&aof,fp); for (j = 0; j < server.dbnum; j++) { char selectcmd[] = "*2\r\n$6\r\nSELECT\r\n"; redisDb *db = server.db+j; @@ -370,8 +434,8 @@ int rewriteAppendOnlyFile(char *filename) { } /* SELECT the new DB */ - if (fwrite(selectcmd,sizeof(selectcmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkLongLong(fp,j) == 0) goto werr; + if (rioWrite(&aof,selectcmd,sizeof(selectcmd)-1) == 0) goto werr; + if (rioWriteBulkLongLong(&aof,j) == 0) goto werr; /* Iterate this DB writing every entry */ while((de = dictNext(di)) != NULL) { @@ -389,10 +453,10 @@ int rewriteAppendOnlyFile(char *filename) { if (o->type == REDIS_STRING) { /* Emit a SET command */ char cmd[]="*3\r\n$3\r\nSET\r\n"; - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; /* Key and value */ - if (fwriteBulkObject(fp,&key) == 0) goto werr; - if (fwriteBulkObject(fp,o) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; + if (rioWriteBulkObject(&aof,o) == 0) goto werr; } else if (o->type == REDIS_LIST) { /* Emit the RPUSHes needed to rebuild the list */ char cmd[]="*3\r\n$5\r\nRPUSH\r\n"; @@ -404,13 +468,13 @@ int rewriteAppendOnlyFile(char *filename) { long long vlong; while(ziplistGet(p,&vstr,&vlen,&vlong)) { - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkObject(fp,&key) == 0) goto werr; + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; if (vstr) { - if (fwriteBulkString(fp,(char*)vstr,vlen) == 0) + if (rioWriteBulkString(&aof,(char*)vstr,vlen) == 0) goto werr; } else { - if (fwriteBulkLongLong(fp,vlong) == 0) + if (rioWriteBulkLongLong(&aof,vlong) == 0) goto werr; } p = ziplistNext(zl,p); @@ -424,9 +488,9 @@ int rewriteAppendOnlyFile(char *filename) { while((ln = listNext(&li))) { robj *eleobj = listNodeValue(ln); - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkObject(fp,&key) == 0) goto werr; - if (fwriteBulkObject(fp,eleobj) == 0) goto werr; + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; + if (rioWriteBulkObject(&aof,eleobj) == 0) goto werr; } } else { redisPanic("Unknown list encoding"); @@ -439,18 +503,18 @@ int rewriteAppendOnlyFile(char *filename) { int ii = 0; int64_t llval; while(intsetGet(o->ptr,ii++,&llval)) { - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkObject(fp,&key) == 0) goto werr; - if (fwriteBulkLongLong(fp,llval) == 0) goto werr; + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; + if (rioWriteBulkLongLong(&aof,llval) == 0) goto werr; } } else if (o->encoding == REDIS_ENCODING_HT) { dictIterator *di = dictGetIterator(o->ptr); dictEntry *de; while((de = dictNext(di)) != NULL) { robj *eleobj = dictGetEntryKey(de); - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkObject(fp,&key) == 0) goto werr; - if (fwriteBulkObject(fp,eleobj) == 0) goto werr; + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; + if (rioWriteBulkObject(&aof,eleobj) == 0) goto werr; } dictReleaseIterator(di); } else { @@ -477,14 +541,14 @@ int rewriteAppendOnlyFile(char *filename) { redisAssert(ziplistGet(eptr,&vstr,&vlen,&vll)); score = zzlGetScore(sptr); - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkObject(fp,&key) == 0) goto werr; - if (fwriteBulkDouble(fp,score) == 0) goto werr; + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; + if (rioWriteBulkDouble(&aof,score) == 0) goto werr; if (vstr != NULL) { - if (fwriteBulkString(fp,(char*)vstr,vlen) == 0) + if (rioWriteBulkString(&aof,(char*)vstr,vlen) == 0) goto werr; } else { - if (fwriteBulkLongLong(fp,vll) == 0) + if (rioWriteBulkLongLong(&aof,vll) == 0) goto werr; } zzlNext(zl,&eptr,&sptr); @@ -498,10 +562,10 @@ int rewriteAppendOnlyFile(char *filename) { robj *eleobj = dictGetEntryKey(de); double *score = dictGetEntryVal(de); - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkObject(fp,&key) == 0) goto werr; - if (fwriteBulkDouble(fp,*score) == 0) goto werr; - if (fwriteBulkObject(fp,eleobj) == 0) goto werr; + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; + if (rioWriteBulkDouble(&aof,*score) == 0) goto werr; + if (rioWriteBulkObject(&aof,eleobj) == 0) goto werr; } dictReleaseIterator(di); } else { @@ -517,11 +581,11 @@ int rewriteAppendOnlyFile(char *filename) { unsigned int flen, vlen; while((p = zipmapNext(p,&field,&flen,&val,&vlen)) != NULL) { - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkObject(fp,&key) == 0) goto werr; - if (fwriteBulkString(fp,(char*)field,flen) == 0) + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; + if (rioWriteBulkString(&aof,(char*)field,flen) == 0) goto werr; - if (fwriteBulkString(fp,(char*)val,vlen) == 0) + if (rioWriteBulkString(&aof,(char*)val,vlen) == 0) goto werr; } } else { @@ -532,10 +596,10 @@ int rewriteAppendOnlyFile(char *filename) { robj *field = dictGetEntryKey(de); robj *val = dictGetEntryVal(de); - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkObject(fp,&key) == 0) goto werr; - if (fwriteBulkObject(fp,field) == 0) goto werr; - if (fwriteBulkObject(fp,val) == 0) goto werr; + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; + if (rioWriteBulkObject(&aof,field) == 0) goto werr; + if (rioWriteBulkObject(&aof,val) == 0) goto werr; } dictReleaseIterator(di); } @@ -547,9 +611,9 @@ int rewriteAppendOnlyFile(char *filename) { char cmd[]="*3\r\n$8\r\nEXPIREAT\r\n"; /* If this key is already expired skip it */ if (expiretime < now) continue; - if (fwrite(cmd,sizeof(cmd)-1,1,fp) == 0) goto werr; - if (fwriteBulkObject(fp,&key) == 0) goto werr; - if (fwriteBulkLongLong(fp,expiretime) == 0) goto werr; + if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr; + if (rioWriteBulkObject(&aof,&key) == 0) goto werr; + if (rioWriteBulkLongLong(&aof,expiretime) == 0) goto werr; } } dictReleaseIterator(di); @@ -762,16 +826,24 @@ void backgroundRewriteDoneHandler(int exitcode, int bysignal) { /* AOF enabled, replace the old fd with the new one. */ oldfd = server.appendfd; server.appendfd = newfd; - if (server.appendfsync != APPENDFSYNC_NO) aof_fsync(newfd); + if (server.appendfsync == APPENDFSYNC_ALWAYS) + aof_fsync(newfd); + else if (server.appendfsync == APPENDFSYNC_EVERYSEC) + aof_background_fsync(newfd); server.appendseldb = -1; /* Make sure SELECT is re-issued */ aofUpdateCurrentSize(); server.auto_aofrewrite_base_size = server.appendonly_current_size; + + /* Clear regular AOF buffer since its contents was just written to + * the new AOF from the background rewrite buffer. */ + sdsfree(server.aofbuf); + server.aofbuf = sdsempty(); } redisLog(REDIS_NOTICE, "Background AOF rewrite successful"); /* Asynchronously close the overwritten AOF. */ - if (oldfd != -1) bioCreateBackgroundJob(REDIS_BIO_CLOSE_FILE,(void*)(long)oldfd); + if (oldfd != -1) bioCreateBackgroundJob(REDIS_BIO_CLOSE_FILE,(void*)(long)oldfd,NULL,NULL); redisLog(REDIS_VERBOSE, "Background AOF rewrite signal handler took %lldus", ustime()-now);