| import{_ as o,c as d,b as l,d as t,t as s,e as a,o as i}from"./app-D83tFSpd.js";const r={};function u(n,e){return i(),d("div",null,[e[41]||(e[41]=l('<p>当前用户可以使用多种手段对正在运行的IoTDB进程进行系统监控，包括使用 Java 的 Jconsole 工具对正在运行的 IoTDB 进程进行系统状态监控，使用 IoTDB 为用户开发的接口查看数据统计量，使用监控框架进行 IoTDB 的运行状态监控</p><h1 id="_1-监控框架" tabindex="-1"><a class="header-anchor" href="#_1-监控框架"><span>1. 监控框架</span></a></h1><p>在IoTDB运行过程中，我们希望对IoTDB的状态进行观测，以便于排查系统问题或者及时发现系统潜在的风险。能<strong>反映系统运行状态的一系列指标</strong>就是系统监控指标。</p><h2 id="_1-1-什么场景下会使用到监控框架" tabindex="-1"><a class="header-anchor" href="#_1-1-什么场景下会使用到监控框架"><span>1.1. 什么场景下会使用到监控框架?</span></a></h2><p>那么什么时候会用到监控框架呢？下面列举一些常见的场景。</p><ol><li><p>系统变慢了</p><p>系统变慢几乎是最常见也最头疼的问题，这时候我们需要尽可能多的信息来帮助我们找到系统变慢的原因，比如：</p><ul><li>JVM信息：是不是有FGC？GC耗时多少？GC后内存有没有恢复？是不是有大量的线程？</li><li>系统信息：CPU使用率是不是太高了？磁盘IO是不是很频繁？</li><li>连接数：当前连接是不是太多？</li><li>接口：当前TPS是多少？各个接口耗时有没有变化？</li><li>线程池：系统中各种任务是否有积压？</li><li>缓存命中率</li></ul></li><li><p>磁盘快满了</p><p>这时候我们迫切想知道最近一段时间数据文件的增长情况，看看是不是某种文件有突增。</p></li><li><p>系统运行是否正常</p><p>此时我们可能需要通过错误日志的数量、集群节点的状态等指标来判断系统是否在正常运行。</p></li></ol><h2 id="_1-2-什么人需要使用监控框架" tabindex="-1"><a class="header-anchor" href="#_1-2-什么人需要使用监控框架"><span>1.2. 什么人需要使用监控框架?</span></a></h2><p>所有关注系统状态的人员都可以使用，包括但不限于研发、测试、运维、DBA等等</p><h2 id="_1-3-iotdb都有哪些监控指标" tabindex="-1"><a class="header-anchor" href="#_1-3-iotdb都有哪些监控指标"><span>1.3. IoTDB都有哪些监控指标?</span></a></h2><p>目前，IoTDB对外提供一些主要模块的metrics，并且随着新功能的开发以及系统优化或者重构，metrics也会同步添加和更新。</p><h3 id="_1-3-1-名词解释" tabindex="-1"><a class="header-anchor" href="#_1-3-1-名词解释"><span>1.3.1. 名词解释</span></a></h3><p>在进一步了解这些指标之前，我们先来看几个名词解释：</p><ul><li><p>Metric Name</p><p>指标名称，比如logback_events_total表示日志事件发生的总次数。</p></li><li><p>Tag</p><p>每个指标下面可以有0到多个分类，比如logback_events_total下有一个<code>level</code>的分类，用来表示特定级别下的日志数量。</p></li></ul><h3 id="_1-3-2-数据格式" tabindex="-1"><a class="header-anchor" href="#_1-3-2-数据格式"><span>1.3.2. 数据格式</span></a></h3><p>IoTDB对外提供JMX和Prometheus格式的监控指标，对于JMX，可以通过<code>org.apache.iotdb.metrics</code>获取系统监控指标指标。</p><p>接下来我们以Prometheus格式为例对目前已有监控项进行说明。</p><h3 id="_1-3-3-iotdb-默认指标" tabindex="-1"><a class="header-anchor" href="#_1-3-3-iotdb-默认指标"><span>1.3.3. IoTDB 默认指标</span></a></h3><h4 id="_1-3-3-1-接入层" tabindex="-1"><a class="header-anchor" href="#_1-3-3-1-接入层"><span>1.3.3.1. 接入层</span></a></h4>',18)),t("table",null,[e[17]||(e[17]=t("thead",null,[t("tr",null,[t("th",null,"Metric"),t("th",null,"Tag"),t("th",null,"level"),t("th",null,"说明"),t("th",null,"示例")])],-1)),t("tbody",null,[t("tr",null,[e[0]||(e[0]=t("td",null,"entry_seconds_count",-1)),t("td",null,'name="'+s(n.interface)+'"',1),e[1]||(e[1]=t("td",null,"important",-1)),e[2]||(e[2]=t("td",null,"接口累计访问次数",-1)),e[3]||(e[3]=t("td",null,'entry_seconds_count{name="openSession",} 1.0',-1))]),t("tr",null,[e[4]||(e[4]=t("td",null,"entry_seconds_sum",-1)),t("td",null,'name="'+s(n.interface)+'"',1),e[5]||(e[5]=t("td",null,"important",-1)),e[6]||(e[6]=t("td",null,"接口累计耗时(s)",-1)),e[7]||(e[7]=t("td",null,'entry_seconds_sum{name="openSession",} 0.024',-1))]),t("tr",null,[e[8]||(e[8]=t("td",null,"entry_seconds_max",-1)),t("td",null,'name="'+s(n.interface)+'"',1),e[9]||(e[9]=t("td",null,"important",-1)),e[10]||(e[10]=t("td",null,"接口最大耗时(s)",-1)),e[11]||(e[11]=t("td",null,'entry_seconds_max{name="openSession",} 0.024',-1))]),e[16]||(e[16]=t("tr",null,[t("td",null,"quantity_total"),t("td",null,'name="pointsIn"'),t("td",null,"important"),t("td",null,"系统累计写入点数"),t("td",null,'quantity_total{name="pointsIn",} 1.0')],-1)),t("tr",null,[e[12]||(e[12]=t("td",null,"thrift_connections",-1)),t("td",null,'name="'+s(n.thriftService)+'"',1),e[13]||(e[13]=t("td",null,"core",-1)),e[14]||(e[14]=t("td",null,"thrift当前连接数",-1)),e[15]||(e[15]=t("td",null,'thrift_connections{name="RPC",} 1.0',-1))])])]),e[42]||(e[42]=l('<h4 id="_1-3-3-2-task" tabindex="-1"><a class="header-anchor" href="#_1-3-3-2-task"><span>1.3.3.2. Task</span></a></h4><table><thead><tr><th>Metric</th><th>Tag</th><th>level</th><th>说明</th><th>示例</th></tr></thead><tbody><tr><td>queue</td><td>name="compaction_inner/compaction_cross/flush",<br>status="running/waiting"</td><td>important</td><td>当前时间任务数</td><td>queue{name="flush",status="waiting",} 0.0<br>queue{name="compaction/flush",status="running",} 0.0</td></tr><tr><td>cost_task_seconds_count</td><td>name="compaction/flush"</td><td>important</td><td>任务累计发生次数</td><td>cost_task_seconds_count{name="flush",} 1.0</td></tr><tr><td>cost_task_seconds_max</td><td>name="compaction/flush"</td><td>important</td><td>到目前为止任务耗时(s)最大的一次</td><td>cost_task_seconds_max{name="flush",} 0.363</td></tr><tr><td>cost_task_seconds_sum</td><td>name="compaction/flush"</td><td>important</td><td>任务累计耗时(s)</td><td>cost_task_seconds_sum{name="flush",} 0.363</td></tr><tr><td>data_written_total</td><td>name="compaction", <br>type="aligned/not-aligned/total"</td><td>important</td><td>合并文件时写入量</td><td>data_written{name="compaction",type="total",} 10240</td></tr><tr><td>data_read_total</td><td>name="compaction"</td><td>important</td><td>合并文件时的读取量</td><td>data_read={name="compaction",} 10240</td></tr></tbody></table><h4 id="_1-3-3-3-内存占用" tabindex="-1"><a class="header-anchor" href="#_1-3-3-3-内存占用"><span>1.3.3.3. 内存占用</span></a></h4><table><thead><tr><th>Metric</th><th>Tag</th><th>level</th><th>说明</th><th>示例</th></tr></thead><tbody><tr><td>mem</td><td>name="chunkMetaData/storageGroup/schemaUsage/schemaRemaining"</td><td>important</td><td>chunkMetaData 占用/storageGroup 占用/schema 占用/schema 剩余的内存（byte）</td><td>mem{name="chunkMetaData",} 2050.0</td></tr></tbody></table><h4 id="_1-3-3-4-缓存" tabindex="-1"><a class="header-anchor" href="#_1-3-3-4-缓存"><span>1.3.3.4. 缓存</span></a></h4><table><thead><tr><th>Metric</th><th>Tag</th><th>level</th><th>说明</th><th>示例</th></tr></thead><tbody><tr><td>cache_hit</td><td>name="chunk/timeSeriesMeta/bloomFilter"</td><td>important</td><td>chunk/timeSeriesMeta缓存命中率,bloomFilter拦截率</td><td>cache_hit{name="chunk",} 80</td></tr></tbody></table><h4 id="_1-3-3-5-业务数据" tabindex="-1"><a class="header-anchor" href="#_1-3-3-5-业务数据"><span>1.3.3.5. 业务数据</span></a></h4>',7)),t("table",null,[e[23]||(e[23]=t("thead",null,[t("tr",null,[t("th",null,"Metric"),t("th",null,"Tag"),t("th",null,"level"),t("th",null,"说明"),t("th",null,"示例")])],-1)),t("tbody",null,[e[22]||(e[22]=t("tr",null,[t("td",null,"quantity"),t("td",null,'name="timeSeries/storageGroup/device/deviceUsingTemplate", type="total/normal/template/template"'),t("td",null,"important"),t("td",null,"当前时间timeSeries/storageGroup/device/激活了模板的device的数量"),t("td",null,'quantity{name="timeSeries",type="normal"} 1.0')],-1)),t("tr",null,[e[18]||(e[18]=t("td",null,"points",-1)),t("td",null,'sg="'+s(n.storageGroup)+'", type="flush"',1),e[19]||(e[19]=t("td",null,"core",-1)),e[20]||(e[20]=t("td",null,"最新一个刷盘的memtale的点数",-1)),e[21]||(e[21]=t("td",null,'quantity{name="memtable",type="flush"} 1.0',-1))])])]),e[43]||(e[43]=t("h4",{id:"_1-3-3-6-集群",tabindex:"-1"},[t("a",{class:"header-anchor",href:"#_1-3-3-6-集群"},[t("span",null,"1.3.3.6. 集群")])],-1)),t("table",null,[e[40]||(e[40]=t("thead",null,[t("tr",null,[t("th",null,"Metric"),t("th",null,"Tag"),t("th",null,"level"),t("th",null,"说明"),t("th",null,"示例")])],-1)),t("tbody",null,[t("tr",null,[e[24]||(e[24]=t("td",null,"cluster_node_leader_count",-1)),t("td",null,'name="'+s(n.ip)+'"',1),e[25]||(e[25]=t("td",null,"important",-1)),e[26]||(e[26]=t("td",null,[a("节点上"),t("code",null,"dataGroupLeader"),a("的数量，用来观察leader是否分布均匀")],-1)),e[27]||(e[27]=t("td",null,'cluster_node_leader_count{name="127.0.0.1",} 2.0',-1))]),t("tr",null,[e[28]||(e[28]=t("td",null,"cluster_uncommitted_log",-1)),t("td",null,'name="'+s(n.ip_datagroupHeader)+'"',1),e[29]||(e[29]=t("td",null,"important",-1)),e[30]||(e[30]=t("td",null,[a("节点"),t("code",null,"uncommitted_log"),a("的数量")],-1)),e[31]||(e[31]=t("td",null,'cluster_uncommitted_log{name="127.0.0.1_Data-127.0.0.1-40010-raftId-0",} 0.0',-1))]),t("tr",null,[e[32]||(e[32]=t("td",null,"cluster_node_status",-1)),t("td",null,'name="'+s(n.ip)+'"',1),e[33]||(e[33]=t("td",null,"important",-1)),e[34]||(e[34]=t("td",null,"节点状态，1=online 2=offline",-1)),e[35]||(e[35]=t("td",null,'cluster_node_status{name="127.0.0.1",} 1.0',-1))]),t("tr",null,[e[36]||(e[36]=t("td",null,"cluster_elect_total",-1)),t("td",null,'name="'+s(n.ip)+'",status="fail/win"',1),e[37]||(e[37]=t("td",null,"important",-1)),e[38]||(e[38]=t("td",null,"节点参与选举的次数及结果",-1)),e[39]||(e[39]=t("td",null,'cluster_elect_total{name="127.0.0.1",status="win",} 1.0',-1))])])]),e[44]||(e[44]=l('<h3 id="_1-3-4-iotdb-预定义指标集" tabindex="-1"><a class="header-anchor" href="#_1-3-4-iotdb-预定义指标集"><span>1.3.4. IoTDB 预定义指标集</span></a></h3><p>用户可以在<code>iotdb-metric.yml</code>文件中，修改<code>predefinedMetrics</code>的值来启用预定义指标集，目前有<code>JVM</code>、<code>LOGBACK</code>、<code>FILE</code>、<code>PROCESS</code>、<code>SYSYTEM</code>这五种。</p><h4 id="_1-3-4-1-jvm" tabindex="-1"><a class="header-anchor" href="#_1-3-4-1-jvm"><span>1.3.4.1. JVM</span></a></h4><h5 id="_1-3-4-1-1-线程" tabindex="-1"><a class="header-anchor" href="#_1-3-4-1-1-线程"><span>1.3.4.1.1. 线程</span></a></h5><table><thead><tr><th>Metric</th><th>Tag</th><th>level</th><th>说明</th><th>示例</th></tr></thead><tbody><tr><td>jvm_threads_live_threads</td><td>无</td><td>important</td><td>当前线程数</td><td>jvm_threads_live_threads 25.0</td></tr><tr><td>jvm_threads_daemon_threads</td><td>无</td><td>important</td><td>当前daemon线程数</td><td>jvm_threads_daemon_threads 12.0</td></tr><tr><td>jvm_threads_peak_threads</td><td>无</td><td>important</td><td>峰值线程数</td><td>jvm_threads_peak_threads 28.0</td></tr><tr><td>jvm_threads_states_threads</td><td>state="runnable/blocked/waiting/timed-waiting/new/terminated"</td><td>important</td><td>当前处于各种状态的线程数</td><td>jvm_threads_states_threads{state="runnable",} 10.0</td></tr></tbody></table><h5 id="_1-3-4-1-2-垃圾回收" tabindex="-1"><a class="header-anchor" href="#_1-3-4-1-2-垃圾回收"><span>1.3.4.1.2. 垃圾回收</span></a></h5><table><thead><tr><th>Metric</th><th>Tag</th><th>level</th><th>说明</th><th>示例</th></tr></thead><tbody><tr><td>jvm_gc_pause_seconds_count</td><td>action="end of major GC/end of minor GC",cause="xxxx"</td><td>important</td><td>YGC/FGC发生次数及其原因</td><td>jvm_gc_pause_seconds_count{action="end of major GC",cause="Metadata GC Threshold",} 1.0</td></tr><tr><td>jvm_gc_pause_seconds_sum</td><td>action="end of major GC/end of minor GC",cause="xxxx"</td><td>important</td><td>YGC/FGC累计耗时及其原因</td><td>jvm_gc_pause_seconds_sum{action="end of major GC",cause="Metadata GC Threshold",} 0.03</td></tr><tr><td>jvm_gc_pause_seconds_max</td><td>action="end of major GC",cause="Metadata GC Threshold"</td><td>important</td><td>YGC/FGC最大耗时及其原因</td><td>jvm_gc_pause_seconds_max{action="end of major GC",cause="Metadata GC Threshold",} 0.0</td></tr><tr><td>jvm_gc_memory_promoted_bytes_total</td><td>无</td><td>important</td><td>从GC之前到GC之后老年代内存池大小正增长的累计</td><td>jvm_gc_memory_promoted_bytes_total 8425512.0</td></tr><tr><td>jvm_gc_max_data_size_bytes</td><td>无</td><td>important</td><td>老年代内存的历史最大值</td><td>jvm_gc_max_data_size_bytes 2.863661056E9</td></tr><tr><td>jvm_gc_live_data_size_bytes</td><td>无</td><td>important</td><td>GC后老年代内存的大小</td><td>jvm_gc_live_data_size_bytes 8450088.0</td></tr><tr><td>jvm_gc_memory_allocated_bytes_total</td><td>无</td><td>important</td><td>在一个GC之后到下一个GC之前年轻代增加的内存</td><td>jvm_gc_memory_allocated_bytes_total 4.2979144E7</td></tr></tbody></table><h5 id="_1-3-4-1-3-内存" tabindex="-1"><a class="header-anchor" href="#_1-3-4-1-3-内存"><span>1.3.4.1.3. 内存</span></a></h5>',8)),e[45]||(e[45]=t("table",null,[t("thead",null,[t("tr",null,[t("th",null,"Metric"),t("th",null,"Tag"),t("th",null,"level"),t("th",null,"说明"),t("th",null,"示例")])]),t("tbody",null,[t("tr",null,[t("td",null,"jvm_buffer_memory_used_bytes"),t("td",null,'id="direct/mapped"'),t("td",null,"important"),t("td",null,"已经使用的缓冲区大小"),t("td",null,'jvm_buffer_memory_used_bytes{id="direct",} 3.46728099E8')]),t("tr",null,[t("td",null,"jvm_buffer_total_capacity_bytes"),t("td",null,'id="direct/mapped"'),t("td",null,"important"),t("td",null,"最大缓冲区大小"),t("td",null,'jvm_buffer_total_capacity_bytes{id="mapped",} 0.0')]),t("tr",null,[t("td",null,"jvm_buffer_count_buffers"),t("td",null,'id="direct/mapped"'),t("td",null,"important"),t("td",null,"当前缓冲区数量"),t("td",null,'jvm_buffer_count_buffers{id="direct",} 183.0')]),t("tr",null,[t("td",null,"jvm_memory_committed_bytes"),t("td",{area:'heap/nonheap,id="xxx",'}),t("td",null,"important"),t("td",null,"当前向JVM申请的内存大小"),t("td",null,[a('jvm_memory_committed_bytes{area="heap",id="Par Survivor Space",} 2.44252672E8'),t("br"),a('jvm_memory_committed_bytes{area="nonheap",id="Metaspace",} 3.9051264E7'),t("br")])]),t("tr",null,[t("td",null,"jvm_memory_max_bytes"),t("td",{area:'heap/nonheap,id="xxx",'}),t("td",null,"important"),t("td",null,"JVM最大内存"),t("td",null,[a('jvm_memory_max_bytes{area="heap",id="Par Survivor Space",} 2.44252672E8'),t("br"),a('jvm_memory_max_bytes{area="nonheap",id="Compressed Class Space",} 1.073741824E9')])]),t("tr",null,[t("td",null,"jvm_memory_used_bytes"),t("td",{area:'heap/nonheap,id="xxx",'}),t("td",null,"important"),t("td",null,"JVM已使用内存大小"),t("td",null,[a('jvm_memory_used_bytes{area="heap",id="Par Eden Space",} 1.000128376E9'),t("br"),a('jvm_memory_used_bytes{area="nonheap",id="Code Cache",} 2.9783808E7'),t("br")])])])],-1)),e[46]||(e[46]=t("h5",{id:"_1-3-4-1-4-classes",tabindex:"-1"},[t("a",{class:"header-anchor",href:"#_1-3-4-1-4-classes"},[t("span",null,"1.3.4.1.4. Classes")])],-1)),e[47]||(e[47]=t("table",null,[t("thead",null,[t("tr",null,[t("th",null,"Metric"),t("th",null,"Tag"),t("th",null,"level"),t("th",null,"说明"),t("th",null,"示例")])]),t("tbody",null,[t("tr",null,[t("td",null,"jvm_classes_unloaded_classes_total"),t("td",null,"无"),t("td",null,"important"),t("td",null,"jvm累计卸载的class数量"),t("td",null,"jvm_classes_unloaded_classes_total 680.0")]),t("tr",null,[t("td",null,"jvm_classes_loaded_classes"),t("td",null,"无"),t("td",null,"important"),t("td",null,"jvm累计加载的class数量"),t("td",null,"jvm_classes_loaded_classes 5975.0")]),t("tr",null,[t("td",null,"jvm_compilation_time_ms_total"),t("td",{compiler:"HotSpot 64-Bit Tiered Compilers,"}),t("td",null,"important"),t("td",null,"jvm耗费在编译上的时间"),t("td",null,'jvm_compilation_time_ms_total{compiler="HotSpot 64-Bit Tiered Compilers",} 107092.0')])])],-1)),e[48]||(e[48]=l('<h4 id="_1-3-4-2-文件-file" tabindex="-1"><a class="header-anchor" href="#_1-3-4-2-文件-file"><span>1.3.4.2. 文件（File）</span></a></h4><table><thead><tr><th>Metric</th><th>Tag</th><th>level</th><th>说明</th><th>示例</th></tr></thead><tbody><tr><td>file_size</td><td>name="wal/seq/unseq"</td><td>important</td><td>当前时间wal/seq/unseq文件大小(byte)</td><td>file_size{name="wal",} 67.0</td></tr><tr><td>file_count</td><td>name="wal/seq/unseq"</td><td>important</td><td>当前时间wal/seq/unseq文件个数</td><td>file_count{name="seq",} 1.0</td></tr></tbody></table><h4 id="_1-3-4-3-日志-logback" tabindex="-1"><a class="header-anchor" href="#_1-3-4-3-日志-logback"><span>1.3.4.3. 日志(logback)</span></a></h4>',3)),e[49]||(e[49]=t("table",null,[t("thead",null,[t("tr",null,[t("th",null,"Metric"),t("th",null,"Tag"),t("th",null,"level"),t("th",null,"说明"),t("th",null,"示例")])]),t("tbody",null,[t("tr",null,[t("td",null,"logback_events_total"),t("td",{level:"trace/debug/info/warn/error,"}),t("td",null,"important"),t("td",null,"trace/debug/info/warn/error日志累计数量"),t("td",null,'logback_events_total{level="warn",} 0.0')])])],-1)),e[50]||(e[50]=l(`<h5 id="_1-3-4-3-1-进程-process" tabindex="-1"><a class="header-anchor" href="#_1-3-4-3-1-进程-process"><span>1.3.4.3.1. 进程（Process）</span></a></h5><table><thead><tr><th>Metric</th><th>Tag</th><th>level</th><th>说明</th><th>示例</th></tr></thead><tbody><tr><td>process_cpu_load</td><td>name="cpu"</td><td>core</td><td>process当前CPU占用率（%）</td><td>process_cpu_load{name="process",} 5.0</td></tr><tr><td>process_cpu_time</td><td>name="cpu"</td><td>core</td><td>process累计占用CPU时间（ns)</td><td>process_cpu_time{name="process",} 3.265625E9</td></tr><tr><td>process_max_mem</td><td>name="memory"</td><td>core</td><td>JVM最大可用内存</td><td>process_max_mem{name="process",} 3.545759744E9</td></tr><tr><td>process_used_mem</td><td>name="memory"</td><td>core</td><td>JVM当前使用内存</td><td>process_used_mem{name="process",} 4.6065456E7</td></tr><tr><td>process_total_mem</td><td>name="memory"</td><td>core</td><td>JVM当前已申请内存</td><td>process_total_mem{name="process",} 2.39599616E8</td></tr><tr><td>process_free_mem</td><td>name="memory"</td><td>core</td><td>JVM当前剩余可用内存</td><td>process_free_mem{name="process",} 1.94035584E8</td></tr><tr><td>process_mem_ratio</td><td>name="memory"</td><td>core</td><td>进程的内存占用比例</td><td>process_mem_ratio{name="process",} 0.0</td></tr><tr><td>process_threads_count</td><td>name="process"</td><td>core</td><td>当前线程数</td><td>process_threads_count{name="process",} 11.0</td></tr><tr><td>process_status</td><td>name="process"</td><td>core</td><td>进程存活状态，1.0为存活，0.0为终止</td><td>process_status{name="process",} 1.0</td></tr></tbody></table><h4 id="_1-3-4-5-系统-system" tabindex="-1"><a class="header-anchor" href="#_1-3-4-5-系统-system"><span>1.3.4.5. 系统（System）</span></a></h4><table><thead><tr><th>Metric</th><th>Tag</th><th>level</th><th>说明</th><th>示例</th></tr></thead><tbody><tr><td>sys_cpu_load</td><td>name="cpu"</td><td>core</td><td>system当前CPU占用率（%）</td><td>sys_cpu_load{name="system",} 15.0</td></tr><tr><td>sys_cpu_cores</td><td>name="cpu"</td><td>core</td><td>jvm可用处理器数</td><td>sys_cpu_cores{name="system",} 16.0</td></tr><tr><td>sys_total_physical_memory_size</td><td>name="memory"</td><td>core</td><td>system最大物理内存</td><td>sys_total_physical_memory_size{name="system",} 1.5950999552E10</td></tr><tr><td>sys_free_physical_memory_size</td><td>name="memory"</td><td>core</td><td>system当前剩余可用内存</td><td>sys_free_physical_memory_size{name="system",} 4.532396032E9</td></tr><tr><td>sys_total_swap_space_size</td><td>name="memory"</td><td>core</td><td>system交换区最大空间</td><td>sys_total_swap_space_size{name="system",} 2.1051273216E10</td></tr><tr><td>sys_free_swap_space_size</td><td>name="memory"</td><td>core</td><td>system交换区剩余可用空间</td><td>sys_free_swap_space_size{name="system",} 2.931576832E9</td></tr><tr><td>sys_committed_vm_size</td><td>name="memory"</td><td>important</td><td>system保证可用于正在运行的进程的虚拟内存量</td><td>sys_committed_vm_size{name="system",} 5.04344576E8</td></tr><tr><td>sys_disk_total_space</td><td>name="disk"</td><td>core</td><td>磁盘总大小</td><td>sys_disk_total_space{name="system",} 5.10770798592E11</td></tr><tr><td>sys_disk_free_space</td><td>name="disk"</td><td>core</td><td>磁盘可用大小</td><td>sys_disk_free_space{name="system",} 3.63467845632E11</td></tr></tbody></table><h3 id="_1-3-5-自定义添加埋点" tabindex="-1"><a class="header-anchor" href="#_1-3-5-自定义添加埋点"><span>1.3.5. 自定义添加埋点</span></a></h3><ul><li>如果想自己在IoTDB中添加更多系统监控指标埋点，可以参考<a href="https://github.com/apache/iotdb/tree/master/metrics" target="_blank" rel="noopener noreferrer">IoTDB Metrics Framework</a>使用说明</li><li>Metric 埋点定义规则 <ul><li><code>Metric</code>：监控项的名称，比如<code>entry_seconds_count</code>为接口累计访问次数，file_size 为文件总数。</li><li><code>Tags</code>：Key-Value对，用来明确被监控项，可选项 <ul><li><code>name = xxx</code>：被监控项的名称，比如对<code>entry_seconds_count</code>这个监控项，name 的含义是被监控的接口名称。</li><li><code>status = xxx</code>：被监控项的状态细分，比如监控 Task 的监控项可以通过该参数，将运行的 Task 和停止的 Task 分开。</li><li><code>user = xxx</code>：被监控项和某个特定用户相关，比如统计root用户的写入总次数。</li><li>根据具体情况自定义......</li></ul></li></ul></li><li>监控指标级别含义： <ul><li>线上运行默认启动级别为<code>Important</code>级，线下调试默认启动级别为<code>Normal</code>级，审核严格程度<code>Core > Important > Normal > All</code></li><li><code>Core</code>：系统的核心指标，供<strong>运维人员</strong>使用，关乎系统的<strong>性能、稳定性、安全性</strong>，比如实例的状况，系统的负载等。</li><li><code>Important</code>：模块的重要指标，供<strong>运维和测试人员</strong>使用，直接关乎<strong>每个模块的运行状态</strong>，比如合并文件个数、执行情况等。</li><li><code>Normal</code>：模块的一般指标，供<strong>开发人员</strong>使用，方便在出现问题时<strong>定位模块</strong>，比如合并中的特定关键操作情况。</li><li><code>All</code>：模块的全部指标，供<strong>模块开发人员</strong>使用，往往在复现问题的时候使用，从而快速解决问题。</li></ul></li></ul><h2 id="_1-4-怎样获取这些系统监控指标" tabindex="-1"><a class="header-anchor" href="#_1-4-怎样获取这些系统监控指标"><span>1.4. 怎样获取这些系统监控指标？</span></a></h2><p>metric采集默认是关闭的，需要先到conf/iotdb-metric.yml中打开后启动server，目前也支持启动后，通过<code>load configuration</code>热加载。</p><h3 id="_1-4-1-配置文件" tabindex="-1"><a class="header-anchor" href="#_1-4-1-配置文件"><span>1.4.1. 配置文件</span></a></h3><div class="language-yaml line-numbers-mode" data-highlighter="prismjs" data-ext="yml" data-title="yml"><pre><code><span class="line"><span class="token comment"># 是否启动监控模块，默认为false</span></span> |
| <span class="line"></span></code></pre><div class="line-numbers" aria-hidden="true" style="counter-reset:line-number 0;"><div class="line-number"></div><div class="line-number"></div><div class="line-number"></div><div class="line-number"></div><div class="line-number"></div><div class="line-number"></div><div class="line-number"></div><div class="line-number"></div><div class="line-number"></div><div class="line-number"></div><div class="line-number"></div></div></div><p>更多细节可以参考下面的文档：</p><p><a href="https://prometheus.io/docs/prometheus/latest/getting_started/" target="_blank" rel="noopener noreferrer">Prometheus安装使用文档</a></p><p><a href="https://prometheus.io/docs/prometheus/latest/configuration/configuration/#scrape_config" target="_blank" rel="noopener noreferrer">Prometheus从HTTP接口拉取metrics数据的配置说明</a></p><p><a href="https://grafana.com/docs/grafana/latest/getting-started/getting-started/" target="_blank" rel="noopener noreferrer">Grafana安装使用文档</a></p><p><a href="https://prometheus.io/docs/visualization/grafana/#grafana-support-for-prometheus" target="_blank" rel="noopener noreferrer">Grafana从Prometheus查询数据并绘图的文档</a></p><h3 id="_1-4-3-apache-iotdb-dashboard" tabindex="-1"><a class="header-anchor" href="#_1-4-3-apache-iotdb-dashboard"><span>1.4.3. Apache IoTDB Dashboard</span></a></h3><p>我们提供了Apache IoTDB Dashboard，在Grafana中显示的效果图如下所示：</p><figure><img src="https://alioss.timecho.com/docs/img/UserGuide/System-Tools/Metrics/dashboard.png" alt="Apache IoTDB Dashboard" tabindex="0" loading="lazy"><figcaption>Apache IoTDB Dashboard</figcaption></figure><p>Apache IoTDB Dashboard的获取方式：</p><p>方式一： 您可以在grafana-metrics-example文件夹下获取到对应不同iotdb版本的Dashboard的json文件。<br> 方式二： 您可以访问Grafana Dashboard官网，点击下载 <a href="https://grafana.com/grafana/dashboards/16132-apache-iotdb-dashboard/?tab=revisions" target="_blank" rel="noopener noreferrer">Apache IoTDB Dashboard</a>。</p><p>在创建Grafana时，您可以选择Import刚刚下载的json文件，并为Apache IoTDB Dashboard选择对应目标数据源。</p><h1 id="_2-系统状态监控" tabindex="-1"><a class="header-anchor" href="#_2-系统状态监控"><span>2. 系统状态监控</span></a></h1><p>进入 Jconsole 监控页面后，首先看到的是 IoTDB 各类运行情况的概览。在这里，您可以看到堆内存信息、线程信息、类信息以及服务器的 CPU 使用情况。</p><h1 id="_3-jmx-mbean-监控" tabindex="-1"><a class="header-anchor" href="#_3-jmx-mbean-监控"><span>3. JMX MBean 监控</span></a></h1><p>通过使用 JConsole 工具并与 JMX 连接，您可以查看一些系统统计信息和参数。<br> 本节描述如何使用 JConsole 的 "Mbean" 选项卡来监视 IoTDB 的一些系统配置、写入数据统计等等。 连接到 JMX 后，您可以通过 "MBeans" 标签找到名为 "org.apache.iotdb.service" 的 "MBean"，如下图所示。</p><p><img style="width:100%;max-width:800px;max-height:600px;margin-left:auto;margin-right:auto;display:block;" src="https://alioss.timecho.com/docs/img/github/149951720-707f1ee8-32ee-4fde-9252-048caebd232e.png"> <br></p><h1 id="_4-性能监控" tabindex="-1"><a class="header-anchor" href="#_4-性能监控"><span>4. 性能监控</span></a></h1><h2 id="_4-1-介绍" tabindex="-1"><a class="header-anchor" href="#_4-1-介绍"><span>4.1. 介绍</span></a></h2><p>性能监控模块用来监控 IOTDB 每一个操作的耗时，以便用户更好的了解数据库的整体性能。此模块会统计每一种操作的平均耗时，以及耗时在一定时间区间内（1ms，4ms，16ms，64ms，256ms，1024ms，以上）的操作的比例。输出文件在 log_measure.log 中。输出样例如下：</p><img style="width:100%;max-width:800px;max-height:600px;margin-left:auto;margin-right:auto;display:block;" src="https://alioss.timecho.com/docs/img/github/60937461-14296f80-a303-11e9-9602-a7bed624bfb3.png"><h2 id="_4-2-配置参数" tabindex="-1"><a class="header-anchor" href="#_4-2-配置参数"><span>4.2. 配置参数</span></a></h2><p>配置文件位置：conf/iotdb-datanode.properties</p><center><p><strong>表 -配置参数以及描述项</strong></p><table><thead><tr><th style="text-align:left;">参数</th><th style="text-align:left;">默认值</th><th style="text-align:left;">描述</th></tr></thead><tbody><tr><td style="text-align:left;">enable_performance_stat</td><td style="text-align:left;">false</td><td style="text-align:left;">是否开启性能监控模块</td></tr></tbody></table></center><h1 id="_5-cache-命中率统计" tabindex="-1"><a class="header-anchor" href="#_5-cache-命中率统计"><span>5. Cache 命中率统计</span></a></h1><p>为了提高查询性能，IOTDB 对 ChunkMetaData 和 TsFileMetaData 进行了缓存。用户可以通过 debug 级别的日志以及 MXBean 两种方式来查看缓存的命中率，并根据缓存命中率以及系统内存来调节缓存所使用的内存大小。使用 MXBean 查看缓存命中率的方法为：</p><ol><li>通过端口 31999 连接 jconsole，并在上方菜单项中选择‘MBean’.</li><li>展开侧边框并选择 'org.apache.iotdb.db.service'. 将会得到如下图所示结果：</li></ol><img style="width:100%;max-width:800px;max-height:600px;margin-left:auto;margin-right:auto;display:block;" src="https://alioss.timecho.com/docs/img/github/112426760-73e3da80-8d73-11eb-9a8f-9232d1f2033b.png">`,48))])}const m=o(r,[["render",u],["__file","Metric-Tool.html.vue"]]),c=JSON.parse('{"path":"/zh/UserGuide/V0.13.x/Maintenance-Tools/Metric-Tool.html","title":"1. 监控框架","lang":"zh-CN","frontmatter":{"description":"当前用户可以使用多种手段对正在运行的IoTDB进程进行系统监控，包括使用 Java 的 Jconsole 工具对正在运行的 IoTDB 进程进行系统状态监控，使用 IoTDB 为用户开发的接口查看数据统计量，使用监控框架进行 IoTDB 的运行状态监控 1. 监控框架 在IoTDB运行过程中，我们希望对IoTDB的状态进行观测，以便于排查系统问题或者及...","head":[["link",{"rel":"alternate","hreflang":"en-us","href":"https://iotdb.apache.org/UserGuide/V0.13.x/Maintenance-Tools/Metric-Tool.html"}],["meta",{"property":"og:url","content":"https://iotdb.apache.org/zh/UserGuide/V0.13.x/Maintenance-Tools/Metric-Tool.html"}],["meta",{"property":"og:site_name","content":"IoTDB Website"}],["meta",{"property":"og:title","content":"1. 监控框架"}],["meta",{"property":"og:description","content":"当前用户可以使用多种手段对正在运行的IoTDB进程进行系统监控，包括使用 Java 的 Jconsole 工具对正在运行的 IoTDB 进程进行系统状态监控，使用 IoTDB 为用户开发的接口查看数据统计量，使用监控框架进行 IoTDB 的运行状态监控 1. 监控框架 在IoTDB运行过程中，我们希望对IoTDB的状态进行观测，以便于排查系统问题或者及..."}],["meta",{"property":"og:type","content":"article"}],["meta",{"property":"og:image","content":"https://alioss.timecho.com/docs/img/UserGuide/System-Tools/Metrics/iotdb_prometheus_grafana.png"}],["meta",{"property":"og:locale","content":"zh-CN"}],["meta",{"property":"og:locale:alternate","content":"en-US"}],["meta",{"property":"og:updated_time","content":"2024-05-09T10:21:49.000Z"}],["meta",{"property":"article:modified_time","content":"2024-05-09T10:21:49.000Z"}],["script",{"type":"application/ld+json"},"{\\"@context\\":\\"https://schema.org\\",\\"@type\\":\\"Article\\",\\"headline\\":\\"1. 监控框架\\",\\"image\\":[\\"https://alioss.timecho.com/docs/img/UserGuide/System-Tools/Metrics/iotdb_prometheus_grafana.png\\",\\"https://alioss.timecho.com/docs/img/UserGuide/System-Tools/Metrics/dashboard.png\\"],\\"dateModified\\":\\"2024-05-09T10:21:49.000Z\\",\\"author\\":[]}"]]},"headers":[{"level":2,"title":"1.1. 什么场景下会使用到监控框架?","slug":"_1-1-什么场景下会使用到监控框架","link":"#_1-1-什么场景下会使用到监控框架","children":[]},{"level":2,"title":"1.2. 什么人需要使用监控框架?","slug":"_1-2-什么人需要使用监控框架","link":"#_1-2-什么人需要使用监控框架","children":[]},{"level":2,"title":"1.3. IoTDB都有哪些监控指标?","slug":"_1-3-iotdb都有哪些监控指标","link":"#_1-3-iotdb都有哪些监控指标","children":[{"level":3,"title":"1.3.1. 名词解释","slug":"_1-3-1-名词解释","link":"#_1-3-1-名词解释","children":[]},{"level":3,"title":"1.3.2. 数据格式","slug":"_1-3-2-数据格式","link":"#_1-3-2-数据格式","children":[]},{"level":3,"title":"1.3.3. IoTDB 默认指标","slug":"_1-3-3-iotdb-默认指标","link":"#_1-3-3-iotdb-默认指标","children":[]},{"level":3,"title":"1.3.4. IoTDB 预定义指标集","slug":"_1-3-4-iotdb-预定义指标集","link":"#_1-3-4-iotdb-预定义指标集","children":[]},{"level":3,"title":"1.3.5. 自定义添加埋点","slug":"_1-3-5-自定义添加埋点","link":"#_1-3-5-自定义添加埋点","children":[]}]},{"level":2,"title":"1.4. 怎样获取这些系统监控指标？","slug":"_1-4-怎样获取这些系统监控指标","link":"#_1-4-怎样获取这些系统监控指标","children":[{"level":3,"title":"1.4.1. 配置文件","slug":"_1-4-1-配置文件","link":"#_1-4-1-配置文件","children":[]},{"level":3,"title":"1.4.2. 对接Prometheus和Grafana","slug":"_1-4-2-对接prometheus和grafana","link":"#_1-4-2-对接prometheus和grafana","children":[]},{"level":3,"title":"1.4.3. Apache IoTDB Dashboard","slug":"_1-4-3-apache-iotdb-dashboard","link":"#_1-4-3-apache-iotdb-dashboard","children":[]}]},{"level":2,"title":"4.1. 介绍","slug":"_4-1-介绍","link":"#_4-1-介绍","children":[]},{"level":2,"title":"4.2. 配置参数","slug":"_4-2-配置参数","link":"#_4-2-配置参数","children":[]}],"git":{"createdTime":1688958677000,"updatedTime":1715250109000,"contributors":[{"name":"CritasWang","email":"critas@outlook.com","commits":1},{"name":"wanghui42","email":"105700158+wanghui42@users.noreply.github.com","commits":1}]},"readingTime":{"minutes":13.02,"words":3907},"filePathRelative":"zh/UserGuide/V0.13.x/Maintenance-Tools/Metric-Tool.md","localizedDate":"2023年7月10日","autoDesc":true}');export{m as comp,c as data}; |
