/* Write a sequence of commands able to fully rebuild the dataset into
* "filename". Used both by REWRITEAOF and BGREWRITEAOF.
*
* In order to minimize the number of commands needed in the rewritten
* log Redis uses variadic commands when possible, such as RPUSH, SADD
* and ZADD. However at max REDIS_AOF_REWRITE_ITEMS_PER_CMD items per time
* are inserted using a single command. */
int rewriteAppendOnlyFile(char *filename) {
dictIterator *di = NULL;
dictEntry *de;
rio aof;
FILE *fp;
char tmpfile[256];
int j;
long long now = mstime();
/* Note that we have to use a different temp name here compared to the
* one used by rewriteAppendOnlyFileBackground() function. */
snprintf(tmpfile,256,"temp-rewriteaof-%d.aof", (int) getpid());
fp = fopen(tmpfile,"w");
if (!fp) {
redisLog(REDIS_WARNING, "Opening the temp file for AOF rewrite in rewriteAppendOnlyFile(): %s", strerror(errno));
return REDIS_ERR;
}
rioInitWithFile(&aof,fp);
if (server.aof_rewrite_incremental_fsync)
rioSetAutoSync(&aof,REDIS_AOF_AUTOSYNC_BYTES);
for (j = 0; j < server.dbnum; j++) {
char selectcmd[] = "*2\r\n$6\r\nSELECT\r\n";
redisDb *db = server.db+j;
dict *d = db->dict;
if (dictSize(d) == 0) continue;
di = dictGetSafeIterator(d);
if (!di) {
fclose(fp);
return REDIS_ERR;
}
/* SELECT the new DB */
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) {
sds keystr;
robj key, *o;
long long expiretime;
keystr = dictGetKey(de);
o = dictGetVal(de);
initStaticStringObject(key,keystr);
expiretime = getExpire(db,&key);
/* If this key is already expired skip it */
if (expiretime != -1 && expiretime < now) continue;
/* Save the key and associated value */
if (o->type == REDIS_STRING) {
/* Emit a SET command */
char cmd[]="*3\r\n$3\r\nSET\r\n";
if (rioWrite(&aof,cmd,sizeof(cmd)-1) == 0) goto werr;
/* Key and value */
if (rioWriteBulkObject(&aof,&key) == 0) goto werr;
if (rioWriteBulkObject(&aof,o) == 0) goto werr;
} else if (o->type == REDIS_LIST) {
if (rewriteListObject(&aof,&key,o) == 0) goto werr;
} else if (o->type == REDIS_SET) {
if (rewriteSetObject(&aof,&key,o) == 0) goto werr;
} else if (o->type == REDIS_ZSET) {
if (rewriteSortedSetObject(&aof,&key,o) == 0) goto werr;
} else if (o->type == REDIS_HASH) {
if (rewriteHashObject(&aof,&key,o) == 0) goto werr;
} else {
redisPanic("Unknown object type");
}
/* Save the expire time */
if (expiretime != -1) {
char cmd[]="*3\r\n$9\r\nPEXPIREAT\r\n";
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);
}
/* Make sure data will not remain on the OS's output buffers */
fflush(fp);
aof_fsync(fileno(fp));
fclose(fp);
/* Use RENAME to make sure the DB file is changed atomically only
* if the generate DB file is ok. */
if (rename(tmpfile,filename) == -1) {
redisLog(REDIS_WARNING,"Error moving temp append only file on the final destination: %s", strerror(errno));
unlink(tmpfile);
return REDIS_ERR;
}
redisLog(REDIS_NOTICE,"SYNC append only file rewrite performed");
return REDIS_OK;
werr:
fclose(fp);
unlink(tmpfile);
redisLog(REDIS_WARNING,"Write error writing append only file on disk: %s", strerror(errno));
if (di) dictReleaseIterator(di);
return REDIS_ERR;
}
/* 根据延时Sample的结果，创建阅读性比较好的分析报告 */
sds createLatencyReport(void) {
sds report = sdsempty();
int advise_better_vm = 0; /* Better virtual machines. */
int advise_slowlog_enabled = 0; /* Enable slowlog. */
int advise_slowlog_tuning = 0; /* Reconfigure slowlog. */
int advise_slowlog_inspect = 0; /* Check your slowlog. */
int advise_disk_contention = 0; /* Try to lower disk contention. */
int advise_scheduler = 0; /* Intrinsic latency. */
int advise_data_writeback = 0; /* data=writeback. */
int advise_no_appendfsync = 0; /* don't fsync during rewrites. */
int advise_local_disk = 0; /* Avoid remote disks. */
int advise_ssd = 0; /* Use an SSD drive. */
int advise_write_load_info = 0; /* Print info about AOF and write load. */
int advise_hz = 0; /* Use higher HZ. */
int advise_large_objects = 0; /* Deletion of large objects. */
int advise_relax_fsync_policy = 0; /* appendfsync always is slow. */
int advices = 0;
/* Return ASAP if the latency engine is disabled and it looks like it
* was never enabled so far. */
if (dictSize(server.latency_events) == 0 &&
server.latency_monitor_threshold == 0)
{
report = sdscat(report,"I'm sorry, Dave, I can't do that. Latency monitoring is disabled in this Redis instance. You may use \"CONFIG SET latency-monitor-threshold <milliseconds>.\" in order to enable it. If we weren't in a deep space mission I'd suggest to take a look at http://redis.io/topics/latency-monitor.\n");
return report;
}
/* Show all the events stats and add for each event some event-related
* comment depending on the values. */
dictIterator *di;
dictEntry *de;
int eventnum = 0;
di = dictGetIterator(server.latency_events);
while((de = dictNext(di)) != NULL) {
char *event = dictGetKey(de);
struct latencyTimeSeries *ts = dictGetVal(de);
struct latencyStats ls;
if (ts == NULL) continue;
eventnum++;
if (eventnum == 1) {
report = sdscat(report,"Dave, I have observed latency spikes in this Redis instance. You don't mind talking about it, do you Dave?\n\n");
}
analyzeLatencyForEvent(event,&ls);
//根据结果构造报告字符串
report = sdscatprintf(report,
"%d. %s: %d latency spikes (average %lums, mean deviation %lums, period %.2f sec). Worst all time event %lums.",
eventnum, event,
ls.samples,
(unsigned long) ls.avg,
(unsigned long) ls.mad,
(double) ls.period/ls.samples,
(unsigned long) ts->max);
/* Fork */
if (!strcasecmp(event,"fork")) {
char *fork_quality;
if (server.stat_fork_rate < 10) {
fork_quality = "terrible";
advise_better_vm = 1;
advices++;
} else if (server.stat_fork_rate < 25) {
fork_quality = "poor";
advise_better_vm = 1;
advices++;
} else if (server.stat_fork_rate < 100) {
fork_quality = "good";
} else {
fork_quality = "excellent";
}
report = sdscatprintf(report,
" Fork rate is %.2f GB/sec (%s).", server.stat_fork_rate,
fork_quality);
}
/* Potentially commands. */
if (!strcasecmp(event,"command")) {
if (server.slowlog_log_slower_than == 0) {
advise_slowlog_enabled = 1;
advices++;
} else if (server.slowlog_log_slower_than/1000 >
server.latency_monitor_threshold)
{
advise_slowlog_tuning = 1;
advices++;
}
advise_slowlog_inspect = 1;
advise_large_objects = 1;
advices += 2;
}
/* fast-command. */
if (!strcasecmp(event,"fast-command")) {
advise_scheduler = 1;
advices++;
}
/* AOF and I/O. */
if (!strcasecmp(event,"aof-write-pending-fsync")) {
advise_local_disk = 1;
advise_disk_contention = 1;
advise_ssd = 1;
advise_data_writeback = 1;
advices += 4;
}
if (!strcasecmp(event,"aof-write-active-child")) {
advise_no_appendfsync = 1;
advise_data_writeback = 1;
advise_ssd = 1;
advices += 3;
}
if (!strcasecmp(event,"aof-write-alone")) {
advise_local_disk = 1;
advise_data_writeback = 1;
advise_ssd = 1;
advices += 3;
}
if (!strcasecmp(event,"aof-fsync-always")) {
advise_relax_fsync_policy = 1;
advices++;
}
if (!strcasecmp(event,"aof-fstat") ||
!strcasecmp(event,"rdb-unlik-temp-file")) {
advise_disk_contention = 1;
advise_local_disk = 1;
advices += 2;
}
if (!strcasecmp(event,"aof-rewrite-diff-write") ||
!strcasecmp(event,"aof-rename")) {
advise_write_load_info = 1;
advise_data_writeback = 1;
advise_ssd = 1;
advise_local_disk = 1;
advices += 4;
}
/* Expire cycle. */
if (!strcasecmp(event,"expire-cycle")) {
advise_hz = 1;
advise_large_objects = 1;
advices += 2;
}
/* Eviction cycle. */
if (!strcasecmp(event,"eviction-cycle")) {
advise_large_objects = 1;
advices++;
}
report = sdscatlen(report,"\n",1);
}
dictReleaseIterator(di);
if (eventnum == 0) {
report = sdscat(report,"Dave, no latency spike was observed during the lifetime of this Redis instance, not in the slightest bit. I honestly think you ought to sit down calmly, take a stress pill, and think things over.\n");
} else if (advices == 0) {
report = sdscat(report,"\nWhile there are latency events logged, I'm not able to suggest any easy fix. Please use the Redis community to get some help, providing this report in your help request.\n");
} else {
/* Add all the suggestions accumulated so far. */
/* Better VM. */
report = sdscat(report,"\nI have a few advices for you:\n\n");
if (advise_better_vm) {
report = sdscat(report,"- If you are using a virtual machine, consider upgrading it with a faster one using an hypervisior that provides less latency during fork() calls. Xen is known to have poor fork() performance. Even in the context of the same VM provider, certain kinds of instances can execute fork faster than others.\n");
}
/* Slow log. */
if (advise_slowlog_enabled) {
report = sdscatprintf(report,"- There are latency issues with potentially slow commands you are using. Try to enable the Slow Log Redis feature using the command 'CONFIG SET slowlog-log-slower-than %llu'. If the Slow log is disabled Redis is not able to log slow commands execution for you.\n", (unsigned long long)server.latency_monitor_threshold*1000);
}
if (advise_slowlog_tuning) {
report = sdscatprintf(report,"- Your current Slow Log configuration only logs events that are slower than your configured latency monitor threshold. Please use 'CONFIG SET slowlog-log-slower-than %llu'.\n", (unsigned long long)server.latency_monitor_threshold*1000);
}
if (advise_slowlog_inspect) {
report = sdscat(report,"- Check your Slow Log to understand what are the commands you are running which are too slow to execute. Please check http://redis.io/commands/slowlog for more information.\n");
}
/* Intrinsic latency. */
if (advise_scheduler) {
report = sdscat(report,"- The system is slow to execute Redis code paths not containing system calls. This usually means the system does not provide Redis CPU time to run for long periods. You should try to:\n"
" 1) Lower the system load.\n"
" 2) Use a computer / VM just for Redis if you are running other softawre in the same system.\n"
" 3) Check if you have a \"noisy neighbour\" problem.\n"
" 4) Check with 'redis-cli --intrinsic-latency 100' what is the intrinsic latency in your system.\n"
" 5) Check if the problem is allocator-related by recompiling Redis with MALLOC=libc, if you are using Jemalloc. However this may create fragmentation problems.\n");
}
/* AOF / Disk latency. */
if (advise_local_disk) {
report = sdscat(report,"- It is strongly advised to use local disks for persistence, especially if you are using AOF. Remote disks provided by platform-as-a-service providers are known to be slow.\n");
}
if (advise_ssd) {
report = sdscat(report,"- SSD disks are able to reduce fsync latency, and total time needed for snapshotting and AOF log rewriting (resulting in smaller memory usage and smaller final AOF rewrite buffer flushes). With extremely high write load SSD disks can be a good option. However Redis should perform reasonably with high load using normal disks. Use this advice as a last resort.\n");
}
if (advise_data_writeback) {
report = sdscat(report,"- Mounting ext3/4 filesystems with data=writeback can provide a performance boost compared to data=ordered, however this mode of operation provides less guarantees, and sometimes it can happen that after a hard crash the AOF file will have an half-written command at the end and will require to be repaired before Redis restarts.\n");
}
if (advise_disk_contention) {
report = sdscat(report,"- Try to lower the disk contention. This is often caused by other disk intensive processes running in the same computer (including other Redis instances).\n");
}
if (advise_no_appendfsync) {
report = sdscat(report,"- Assuming from the point of view of data safety this is viable in your environment, you could try to enable the 'no-appendfsync-on-rewrite' option, so that fsync will not be performed while there is a child rewriting the AOF file or producing an RDB file (the moment where there is high disk contention).\n");
}
if (advise_relax_fsync_policy && server.aof_fsync == AOF_FSYNC_ALWAYS) {
report = sdscat(report,"- Your fsync policy is set to 'always'. It is very hard to get good performances with such a setup, if possible try to relax the fsync policy to 'onesec'.\n");
}
if (advise_write_load_info) {
report = sdscat(report,"- Latency during the AOF atomic rename operation or when the final difference is flushed to the AOF file at the end of the rewrite, sometimes is caused by very high write load, causing the AOF buffer to get very large. If possible try to send less commands to accomplish the same work, or use Lua scripts to group multiple operations into a single EVALSHA call.\n");
}
if (advise_hz && server.hz < 100) {
report = sdscat(report,"- In order to make the Redis keys expiring process more incremental, try to set the 'hz' configuration parameter to 100 using 'CONFIG SET hz 100'.\n");
}
if (advise_large_objects) {
report = sdscat(report,"- Deleting, expiring or evicting (because of maxmemory policy) large objects is a blocking operation. If you have very large objects that are often deleted, expired, or evicted, try to fragment those objects into multiple smaller objects.\n");
}
}
return report;
}
/* Emit the commands needed to rebuild a sorted set object.
* The function returns 0 on error, 1 on success. */
int rewriteSortedSetObject(rio *r, robj *key, robj *o) {
long long count = 0, items = zsetLength(o);
if (o->encoding == REDIS_ENCODING_ZIPLIST) {
unsigned char *zl = o->ptr;
unsigned char *eptr, *sptr;
unsigned char *vstr;
unsigned int vlen;
long long vll;
double score;
eptr = ziplistIndex(zl,0);
redisAssert(eptr != NULL);
sptr = ziplistNext(zl,eptr);
redisAssert(sptr != NULL);
while (eptr != NULL) {
redisAssert(ziplistGet(eptr,&vstr,&vlen,&vll));
score = zzlGetScore(sptr);
if (count == 0) {
int cmd_items = (items > REDIS_AOF_REWRITE_ITEMS_PER_CMD) ?
REDIS_AOF_REWRITE_ITEMS_PER_CMD : items;
if (rioWriteBulkCount(r,'*',2+cmd_items*2) == 0) return 0;
if (rioWriteBulkString(r,"ZADD",4) == 0) return 0;
if (rioWriteBulkObject(r,key) == 0) return 0;
}
if (rioWriteBulkDouble(r,score) == 0) return 0;
if (vstr != NULL) {
if (rioWriteBulkString(r,(char*)vstr,vlen) == 0) return 0;
} else {
if (rioWriteBulkLongLong(r,vll) == 0) return 0;
}
zzlNext(zl,&eptr,&sptr);
if (++count == REDIS_AOF_REWRITE_ITEMS_PER_CMD) count = 0;
items--;
}
} else if (o->encoding == REDIS_ENCODING_SKIPLIST) {
zset *zs = o->ptr;
dictIterator *di = dictGetIterator(zs->dict);
dictEntry *de;
while((de = dictNext(di)) != NULL) {
robj *eleobj = dictGetKey(de);
double *score = dictGetVal(de);
if (count == 0) {
int cmd_items = (items > REDIS_AOF_REWRITE_ITEMS_PER_CMD) ?
REDIS_AOF_REWRITE_ITEMS_PER_CMD : items;
if (rioWriteBulkCount(r,'*',2+cmd_items*2) == 0) return 0;
if (rioWriteBulkString(r,"ZADD",4) == 0) return 0;
if (rioWriteBulkObject(r,key) == 0) return 0;
}
if (rioWriteBulkDouble(r,*score) == 0) return 0;
if (rioWriteBulkObject(r,eleobj) == 0) return 0;
if (++count == REDIS_AOF_REWRITE_ITEMS_PER_CMD) count = 0;
items--;
}
dictReleaseIterator(di);
} else {
redisPanic("Unknown sorted zset encoding");
}
return 1;
}
