原始网页:https://www.cockroachlabs.com/docs/stable/admin-ui-custom-chart-debug-page.html


在Admin界面中的 Custom Chart debug页能够创建一个或多个基于下述超过200个可用指标的组合图表。定制化的仪表盘也可以通过特定的编码的URL来定义,能够通过分享该编码好的URL分享定制化的内容,或是通过保存书签来随时查看仪表盘内容。

访问Custom Chart页

用户可以登陆Admin界面,查看Custom Chart debug页,或者:

使用Custom Chart页

图片

Custom Chart页中,用户可以调整图表的时间跨度,也可以添加更多的定制化图表并自定义图表内容:

属性列 简介
Metric Name 通过该名称引用指标,例如sql.bytesin
Downsampler “Downsampler”操作用于合并一段时间内多个独立的数据点为一个聚合的数据点。在存储上CockroachDB每10秒钟内保存一个数据点,而用户在查询一个大跨度的时间区间时,查询返回结果可能是每分钟、每5分钟甚至每小时的聚合结果,减少查询请求返回的数据量。

配置值:
AVG: 返回一段时间内的平均值。
MIN: 返回一段时间内的最小值。
MAX: 返回一段时间内的最高值。
SUM: 返回一段时间内的累加值。
Aggregator 用于合并不同节点的数据点,类似于Downsampler。

配置值:
AVG: 返回一段时间内的平均值。
MIN: 返回一段时间内的最小值。
MAX: 返回一段时间内的最高值。
SUM: 返回一段时间内的累加值。
Rate 决定反映指定时间区间内的指标数值变动速率的方式。

配置值:
Normal: 返回记录的真实值。
Rate: 返回每秒钟指标数值变动的速率。
Non-negative Rate: 反映指标数值变动的速率,将过滤掉负值并替换为0。实际记录的指标数值取至单调递增的计数器的累计值,可能是一个很大的数。该数值的变化率代表计数事件累计的变化率。“Non-negative Rate”类型是很有必要的,可以观测保存在节点内存中的计数值。在必要的时候节点会将该指标的计数值重置为0。
Source 查询节点的集合,可能取值:
整个集群
单个命名节点
Per Node 如果设置为ture,则在图表上会为每个节点绘制一条线显示该节点的指标值。

示例

Query user and system CPU usage

图片

如果需要对比系统和用户空间CPU使用率,可以选择以下指标:

Y轴刻度标志为Count,数值1代表100%的CPU使用率。聚合后的指标值可能会超过数值1,代表CPU使用率超过100%。

选中Per Node查看每个节点的情况,用户可以很容易地发现哪些节点的CPU使用率高于或低于平均值。

可用指标

NOTE: 以下列表直接提取自源码文件,可能会随着代码更新而有所变动。此外当中的一些指标可能已经使用在Admin界面的其他页面的图表中。

指标名 简介
addsstable.applications SSTable文件成功Ingest的次数(例如副本复制的时候)。
Number of SSTable ingestions applied (i.e. applied by Replicas)
addsstable.copies 所有SSTable文件Ingest作业中,需要copy操作的Ingest的次数。
Number of SSTable ingestions that required copying files during application
addsstable.proposals 倡议的SSTable文件Ingest的次数(例如租约持有者发送Raft日志)。
Number of SSTable ingestions proposed (i.e. sent to Raft by lease holders)
build.timestamp 编译时间
Build information
capacity.available 可用的存储容量
Available storage capacity
capacity.reserved 为快照预留的存储容量
Capacity reserved for snapshots
capacity.used 已使用的存储容量
Used storage capacity
capacity 总存储容量
Total storage capacity
clock-offset.meannanos 指定节点相比于其他节点的时间偏差值的平均值,单位为纳秒。
Mean clock offset with other nodes in nanoseconds
clock-offset.stddevnanos 指定节点相比于其他节点的时间偏差值的标准差,单位为纳秒。
Std dev clock offset with other nodes in nanoseconds
compactor.compactingnanos 范围Compaction执行的总时间,单位为纳秒。
Number of nanoseconds spent compacting ranges
compactor.compactions.failure 提交给存储引擎,执行失败的范围Compaction的数量。
Number of failed compaction requests sent to the storage engine
compactor.compactions.success 提交给存储引擎,执行成功的范围Compaction的数量。
Number of successful compaction requests sent to the storage engine
compactor.suggestionbytes.compacted 合并了的SuggestedCompactions的逻辑字节数。
Number of logical bytes compacted from suggested compactions
compactor.suggestionbytes.queued 队列中SuggestedCompactions的逻辑字节数。
Number of logical bytes in suggested compactions in the queue
compactor.suggestionbytes.skipped 未合并的SuggestedCompactions的逻辑字节数。
Number of logical bytes in suggested compactions which were not compacted
distsender.batches.partial 已处理的Partial Batch的数量。
Number of partial batches processed
distsender.batches 已处理的Batch的数量。
Number of batches processed
distsender.errors.notleaseholder NotLeaseHolderErrors错误出现的次数。
Number of NotLeaseHolderErrors encountered
distsender.rpc.sent.local 发送本地的RPCs数量。
Number of local RPCs sent
distsender.rpc.sent.nextreplicaerror 因副本错误导致的RPCs发送的次数。
Number of RPCs sent due to per-replica errors
distsender.rpc.sent 发送RPCs的次数。
Number of RPCs sent
exec.error 指定节点上执行失败的KV批量请求的数量。
Number of batch KV requests that failed to execute on this node
exec.latency 指定节点上KV批量请求的执行时间,单位为纳秒。
Latency in nanoseconds of batch KV requests executed on this node
exec.success 指定节点上执行成功的KV批量请求的数量。
Number of batch KV requests executed successfully on this node
gcbytesage non-live数据的累计年龄值,单位为秒。
Cumulative age of non-live data in seconds
gossip.bytes.received 接收到的Gossip数据的字节数。
Number of received gossip bytes
gossip.bytes.sent 发送的Gossip数据的字节数。
Number of sent gossip bytes
gossip.connections.incoming 活跃的Gossip连接的入度。
Number of active incoming gossip connections
gossip.connections.outgoing 活跃的Gossip连接的出度。
Number of active outgoing gossip connections
gossip.connections.refused Gossip连接中被拒绝的入度。
Number of refused incoming gossip connections
gossip.infos.received 接收到的Gossip消息对象的数量。
Number of received gossip Info objects
gossip.infos.sent 发送的Gossip消息对象的数量。
Number of sent gossip Info objects
intentage intents的累计年龄值,单位为秒。
Cumulative age of intents in seconds
intentbytes 所有intent键值对的字节数。
Number of bytes in intent KV pairs
intentcount intent的数量。
Count of intent keys
keybytes 所有Key值的字节数。
Number of bytes taken up by keys
keycount 所有Key值的数量。
Count of all keys
lastupdatenanos bytes/keys/intents相关指标最近更新的Unix时间,单位为纳秒。
Time in nanoseconds since Unix epoch at which bytes/keys/intents metrics were last updated
leases.epoch 持有epoch租约的租约持有者(副本)的数量。
Number of replica leaseholders using epoch-based leases
leases.error 租约请求失败的次数。
Number of failed lease requests
leases.expiration 持有expiration租约的租约持有者(副本)的数量。
Number of replica leaseholders using expiration-based leases
leases.success 租约请求成功的次数。
Number of successful lease requests
leases.transfers.error 租约传递失败的次数。
Number of failed lease transfers
leases.transfers.success 租约传递成功的次数。
Number of successful lease transfers
livebytes 存活数据(完整键值对)的字节数。
Number of bytes of live data (keys plus values)
livecount 存活的键值对的数量
Count of live keys
liveness.epochincrements 指定的节点延续其存活epoch的次数。
Number of times this node has incremented its liveness epoch
liveness.heartbeatfailures 该节点验证节点存活的心跳检测的失败次数
Number of failed node liveness heartbeats from this node
liveness.heartbeatlatency 该节点验证节点存活的心跳的时延,单位为纳秒。
Node liveness heartbeat latency in nanoseconds
liveness.heartbeatsuccesses 该节点验证节点存活的心跳检测的成功次数。
Number of successful node liveness heartbeats from this node
liveness.livenodes 集群中存活节点的数量(若发起查询的节点不是存活节点,则计数为0)
Number of live nodes in the cluster (will be 0 if this node is not itself live)
node-id 标记了Advertised RPC或HTTP地址的节点ID
node ID with labels for advertised RPC and HTTP addresses
queue.consistency.pending 一致性检查队列中待检测的副本数量。
Number of pending replicas in the consistency checker queue
queue.consistency.process.failure 一致性检查队列中检测失败的副本数量。
Number of replicas which failed processing in the consistency checker queue
queue.consistency.process.success 一致性检查队列中检测成功的副本数量。
Number of replicas successfully processed by the consistency checker queue
queue.consistency.processingnanos 一致性检查队列中处理副本的时间,单位为纳秒。
Nanoseconds spent processing replicas in the consistency checker queue
queue.gc.info.abortspanconsidered 因超龄而触发删除操作的AbortSpan实体的数量。
Number of AbortSpan entries old enough to be considered for removal
queue.gc.info.abortspangcnum 触发删除操作的AbortSpan实体的数量。
Number of AbortSpan entries fit for removal
queue.gc.info.abortspanscanned 从存储引擎中扫描出来的AbortSpan中存在的事务数量。
Number of transactions present in the AbortSpan scanned from the engine
queue.gc.info.intentsconsidered “老龄”intents的数量。
Number of 'old' intents
queue.gc.info.intenttxns 相关联的事务的数量(去重)。Number of associated distinct transactions
queue.gc.info.numkeysaffected 能够进行GC操作的数据中键值对数量。
Number of keys with GC'able data
queue.gc.info.pushtxn GC队列push的次数。
Number of attempted pushes
queue.gc.info.resolvesuccess intent成功回收的次数。
Number of successful intent resolutions
queue.gc.info.resolvetotal 尝试回收intent的次数。
Number of attempted intent resolutions
queue.gc.info.transactionspangcaborted 中止事务中能够进行GC操作的实体数量。
Number of GC'able entries corresponding to aborted txns
queue.gc.info.transactionspangccommitted 已提交事务中能够进行GC操作的实体数量。
Number of GC'able entries corresponding to committed txns
queue.gc.info.transactionspangcpending 等待事务中能够进行GC操作的实体数量。
Number of GC'able entries corresponding to pending txns
queue.gc.info.transactionspanscanned 存储引擎扫描出来的事务span中实体的数量。
Number of entries in transaction spans scanned from the engine
queue.gc.pending GC队列中等待GC的副本的数量。
Number of pending replicas in the GC queue
queue.gc.process.failure GC队列中GC失败的副本的数量。
Number of replicas which failed processing in the GC queue
queue.gc.process.success GC队列中成功GC的副本的数量。
Number of replicas successfully processed by the GC queue
queue.gc.processingnanos 用于进行GC队列中副本的GC操作的时间,单位为纳秒。
Nanoseconds spent processing replicas in the GC queue
queue.raftlog.pending Raft日志队列中等待的副本数量。
Number of pending replicas in the Raft log queue
queue.raftlog.process.failure Raft日志队列中处理失败的副本数量。
Number of replicas which failed processing in the Raft log queue
queue.raftlog.process.success Raft日志队列中处理成功的副本数量。
Number of replicas successfully processed by the Raft log queue
queue.raftlog.processingnanos Raft日志队列处理副本的时间,单位为纳秒。
Nanoseconds spent processing replicas in the Raft log queue
queue.raftsnapshot.pending Raft修复队列中等待的副本数量。
Number of pending replicas in the Raft repair queue
queue.raftsnapshot.process.failure Raft修复队列中处理失败的副本数量。
Number of replicas which failed processing in the Raft repair queue
queue.raftsnapshot.process.success Raft修复队列中处理成功的副本数量。
Number of replicas successfully processed by the Raft repair queue
queue.raftsnapshot.processingnanos Raft修复队列处理副本的时间,单位为纳秒。
Nanoseconds spent processing replicas in the Raft repair queue
queue.replicagc.pending 副本GC队列中等待的副本数量。
Number of pending replicas in the replica GC queue
queue.replicagc.process.failure 副本GC队列中处理失败的副本数量。
Number of replicas which failed processing in the replica GC queue
queue.replicagc.process.success 副本GC队列中处理成功的副本数量。
Number of replicas successfully processed by the replica GC queue
queue.replicagc.processingnanos 副本GC队列处理副本的时间,单位为纳秒。
Nanoseconds spent processing replicas in the replica GC queue
queue.replicagc.removereplica 副本GC队列尝试删除的副本数量
Number of replica removals attempted by the replica gc queue
queue.replicate.addreplica 复制队列尝试增加的副本数量
Number of replica additions attempted by the replicate queue
queue.replicate.pending 复制队列中等待副本的数量
Number of pending replicas in the replicate queue
queue.replicate.process.failure 复制队列中处理失败的副本数量
Number of replicas which failed processing in the replicate queue
queue.replicate.process.success 复制队列中处理成功的副本数量
Number of replicas successfully processed by the replicate queue
queue.replicate.processingnanos 复制队列中处理副本的时间,单位为纳秒
Nanoseconds spent processing replicas in the replicate queue
queue.replicate.purgatory 复制队列中因处理失败而加入到purgatory中等待分配选项的副本数量
Number of replicas in the replicate queue's purgatory, awaiting allocation options
queue.replicate.rebalancereplica 复制队列中重平衡器尝试增加的副本数量
Number of replica rebalancer-initiated additions attempted by the replicate queue
queue.replicate.removedeadreplica 复制队列中尝试删除的死亡副本的数量(通常是节点丢失的时候)。
Number of dead replica removals attempted by the replicate queue (typically in response to a node outage)
queue.replicate.removereplica 复制队列中尝试删除的副本数量(通常是重平衡器增加副本的时候)。
Number of replica removals attempted by the replicate queue (typically in response to a rebalancer-initiated addition)
queue.replicate.transferlease 复制队列中Range租约尝试传递的次数。
Number of range lease transfers attempted by the replicate queue
queue.split.pending Range切分队列中等待的副本数量。
Number of pending replicas in the split queue
queue.split.process.failure Range切分队列中处理失败的副本数量。
Number of replicas which failed processing in the split queue
queue.split.process.success Range切分队列中处理成功的副本数量。
Number of replicas successfully processed by the split queue
queue.split.processingnanos Range切分队列中处理副本的时间,单位为纳秒。
Nanoseconds spent processing replicas in the split queue
queue.tsmaintenance.pending 时间戳维护队列中等待的副本数量。
Number of pending replicas in the time series maintenance queue
queue.tsmaintenance.process.failure 时间戳维护队列中处理失败的副本数量。
Number of replicas which failed processing in the time series maintenance queue
queue.tsmaintenance.process.success 时间戳维护队列中处理成功的副本数量。
Number of replicas successfully processed by the time series maintenance queue
queue.tsmaintenance.processingnanos 时间戳维护队列中处理副本的时间,单位为纳秒。
Nanoseconds spent processing replicas in the time series maintenance queue
raft.commandsapplied Raft命令应用的次数。
Count of Raft commands applied
raft.enqueued.pending Raft传输队列中等待发送的消息的数量
Number of pending outgoing messages in the Raft Transport queue
raft.heartbeats.pending 等待合并的心跳和responses的数量。
Number of pending heartbeats and responses waiting to be coalesced
raft.process.commandcommit.latency 提交Raft命令的时延的直方图,单位为纳秒。
Latency histogram in nanoseconds for committing Raft commands
raft.process.logcommit.latency 提交的Raft日志项的时延的直方图,单位为纳秒。
Latency histogram in nanoseconds for committing Raft log entries
raft.process.tickingnanos store.processRaft()中执行replica.Tick()部分使用的时间,单位为纳秒。
Nanoseconds spent in store.processRaft() processing replica.Tick()
raft.process.workingnanos store.processRaft()执行的时间,单位为纳秒。Nanoseconds spent in store.processRaft() working
raft.rcvd.app 指定store接收到的MsgApp消息的数量。
Number of MsgApp messages received by this store
raft.rcvd.appresp 指定store接收到的MsgAppResp消息的数量。
Number of MsgAppResp messages received by this store
raft.rcvd.dropped 接收不成功的Raft消息的数量。
Number of dropped incoming Raft messages
raft.rcvd.heartbeat 指定store接收到的MsgHeartbeat消息的数量(不考虑消息合并的影响)。
Number of (coalesced, if enabled) MsgHeartbeat messages received by this store
raft.rcvd.heartbeatresp 指定store接收到的MsgHeartbeatResp消息的数量(不考虑消息合并的影响)。
Number of (coalesced, if enabled) MsgHeartbeatResp messages received by this store
raft.rcvd.prevote 指定store接收到的MsgPreVote消息的数量。
Number of MsgPreVote messages received by this store
raft.rcvd.prevoteresp 指定store接收到的MsgPreVoteResp消息的数量。
Number of MsgPreVoteResp messages received by this store
raft.rcvd.prop 指定store接收到的MsgProp消息的数量。
Number of MsgProp messages received by this store
raft.rcvd.snap 指定store接收到的MsgSnap消息的数量。
Number of MsgSnap messages received by this store
raft.rcvd.timeoutnow 指定store接收到的MsgTimeoutNow消息的数量。
Number of MsgTimeoutNow messages received by this store
raft.rcvd.transferleader 指定store接收到的MsgTransferLeader消息的数量。
Number of MsgTransferLeader messages received by this store
raft.rcvd.vote 指定store接收到的MsgVote消息的数量。
Number of MsgVote messages received by this store
raft.rcvd.voteresp 指定store接收到的MsgVoteResp消息的数量。
Number of MsgVoteResp messages received by this store
raft.ticks 排队中的Raft ticks的数量。
Number of Raft ticks queued
raftlog.behind Raft日志项在其他store上的Follower的数量。
Number of Raft log entries followers on other stores are behind
raftlog.truncated 被截断的Raft日志项的数量。
Number of Raft log entries truncated
range.adds 新增的Range数量。
Number of range additions
range.raftleadertransfers Raft leader传递的次数。
Number of raft leader transfers
range.removes 删除的Range数量。
Number of range removals
range.snapshots.generated 生成的快照数量。
Number of generated snapshots
range.snapshots.normal-applied 应用的快照数量。
Number of applied snapshots
range.snapshots.preemptive-applied 事先应用的快照数量。Number of applied pre-emptive snapshots
range.splits Range切分的次数。
Number of range splits
ranges.unavailable 因存活副本数量达不到最低可用要求而不可用的Range的数量。
Number of ranges with fewer live replicas than needed for quorum
ranges.underreplicated 正在复制的Range数量。
Number of ranges with fewer live replicas than the replication target
ranges Range的数量
Number of ranges
rebalancing.writespersecond 指定store每秒钟写入的Key的数量(例如Raft日志),在重平衡决策的时候将取很长一段时间内的平均值。
Number of keys written (i.e. applied by raft) per second to the store, averaged over a large time period as used in rebalancing decisions
replicas.commandqueue.combinedqueuesize 合并后的CommandQueues中command的数量。
Number of commands in all CommandQueues combined
replicas.commandqueue.combinedreadcount 合并后的CommandQueues中只读command的数量。
Number of read-only commands in all CommandQueues combined
replicas.commandqueue.combinedwritecount 合并后的CommandQueues中读写command的数量。
Number of read-write commands in all CommandQueues combined
replicas.commandqueue.maxoverlaps 一个command添加到任意CommandQueue的时候,其观测到的重叠的command的最大数量。
Largest number of overlapping commands seen when adding to any CommandQueue
replicas.commandqueue.maxreadcount 所有CommandQueue中只读command的最大数量。
Largest number of read-only commands in any CommandQueue
replicas.commandqueue.maxsize 所有CommandQueue中最大command数量。
Largest number of commands in any CommandQueue
replicas.commandqueue.maxtreesize 所有CommandQueue区间树中最大区间数量。
Largest number of intervals in any CommandQueue's interval tree
replicas.commandqueue.maxwritecount 所有CommandQueue中读写command的最大数量。
Largest number of read-write commands in any CommandQueue
replicas.leaders_not_leaseholders Raft领袖和Range租约不在同一个store的副本数量。
Number of replicas that are Raft leaders whose range lease is held by another store
replicas.leaders Raft领袖的数量。
Number of raft leaders
replicas.leaseholders 租约持有者的数量。
Number of lease holders
replicas.quiescent 静止副本的数量。
Number of quiesced replicas
replicas.reserved 用于快照的副本的数量。
Number of replicas reserved for snapshots
replicas 副本的数量。
Number of replicas
requests.backpressure.split 等待Range切分的Backpressure模式的写的数量。
Number of backpressured writes waiting on a Range split
requests.slow.commandqueue command队列中滞留许久的请求的数量。
Number of requests that have been stuck for a long time in the command queue
requests.slow.distsender distsender中滞留许久的请求的数量。
Number of requests that have been stuck for a long time in the dist sender
requests.slow.lease 长时间等待的、用于获取租约的请求的数量
Number of requests that have been stuck for a long time acquiring a lease
requests.slow.raft 长时间等待的Raft请求的数量
Number of requests that have been stuck for a long time in raft
rocksdb.block.cache.hits RocksDB中的BlockCache命中的计数。
Count of block cache hits
rocksdb.block.cache.misses RocksDB中的BlockCache未命中的计数。
Count of block cache misses
rocksdb.block.cache.pinned-usage BlockCache被固定的缓存的字节数。
Bytes pinned by the block cache
rocksdb.block.cache.usage BlockCache已缓存的字节数。Bytes used by the block cache
rocksdb.bloom.filter.prefix.checked BloomFilter使用的次数。
Number of times the bloom filter was checked
rocksdb.bloom.filter.prefix.useful 通过BloomFilter过滤,避免了迭代器创建的次数
Number of times the bloom filter helped avoid iterator creation
rocksdb.compactions RocksDB Compaction的次数。
Number of table compactions
rocksdb.flushes RocksDB Memtable Flush的次数。
Number of table flushes
rocksdb.memtable.total-size RocksDB当前Memtable的大小。
Current size of memtable in bytes
rocksdb.num-sstables RocksDB SSTable文件的数量。
Number of rocksdb SSTables
rocksdb.read-amplification RocksDB读放大的数值
Number of disk reads per query
rocksdb.table-readers-mem-estimate 索引块和BloomFilter块使用的内存大小。
Memory used by index and filter blocks
round-trip-latency 指定节点与其他节点的往返时延,单位为的往返时延,单位为纳秒。纳秒。
Distribution of round-trip latencies with other nodes in nanoseconds
security.certificate.expiration.ca CA证书到期时间(Unix时间,单位为秒),0代表无证书验证。(Unix时间,单位为秒),0代表无证书验证或存在异常或存在异常。
Expiration timestamp in seconds since Unix epoch for the CA certificate. 0 means no certificate or error.
security.certificate.expiration.node 节点证书到期时间(Unix时间,单位为秒),0代表无证书验证或存在异常。
Expiration timestamp in seconds since Unix epoch for the node certificate. 0 means no certificate or error.
sql.bytesin 接收到的SQL字节数。SQL字节数发送的SQL字节数
Number of sql bytes received
sql.bytesout 发送的SQL字节数。活动的SQL链接数量。
Number of sql bytes sent
sql.conns 活跃的SQL连接数量。
Number of active sql connections
sql.ddl.count DDL语句的数量。SQL DDL语句的数量。
Number of SQL DDL statements
sql.delete.count DELETE语句的数量。DELETE语句的数量。
Number of SQL DELETE statements
sql.distsql.exec.latency 活动的分布式SQL语句的时延,单位为纳秒。
Latency in nanoseconds of DistSQL statement execution
sql.distsql.flows.active 当前当前活动的分布式SQL流的数量流的数量
Number of distributed SQL flows currently active
sql.distsql.flows.total 已执行的已执行的分布式SQL流的数量。数量。
Number of distributed SQL flows executed
sql.distsql.queries.active 当前活动的分布式SQL查询的数量。数量。
Number of distributed SQL queries currently active
sql.distsql.queries.total 已执行的已执行的分布式SQL查询的的数量。
Number of distributed SQL queries executed
sql.distsql.select.count 分布式SELECT语句的数量。
Number of DistSQL SELECT statements
sql.distsql.service.latency 分布式请求执行的时延,单位为纳秒。纳秒。
Latency in nanoseconds of DistSQL request execution
sql.exec.latency SQL语句执行的时延,单位为纳秒。
Latency in nanoseconds of SQL statement execution
sql.insert.count INSERT语句的数量。INSERT语句的数量。
Number of SQL INSERT statements
sql.mem.current 当前SQL语句内存的使用量。使用量。
Current sql statement memory usage
sql.mem.distsql.current 当前分布式分布式SQL语句内存的使用量。使用量。Current sql statement memory usage for distsql
sql.mem.distsql.max 平均每个平均每个分布式SQL语句的内存使用量。。
Memory usage per sql statement for distsql
sql.mem.max 平均每个SQL语句的内存使用量。
Memory usage per sql statement
sql.mem.session.current 当前所有所有SQL session的内存使用量。Current sql session memory usage
sql.mem.session.max 每个SQL session的内存的使用量。每个SQL事务的内存使用量。
Memory usage per sql session
sql.mem.txn.current 当前所有所有SQL事务的内存使用量。
Current sql transaction memory usage
sql.mem.txn.max 每个SQL事务的内存使用量。
Memory usage per sql transaction
sql.misc.count 其他类型的SQL语句的数量。
Number of other SQL statements
sql.query.count SQL查询的数量。SQL查询的数量。
Number of SQL queries
sql.select.count SELECT语句的数量。
Number of SQL SELECT statements
sql.service.latency SQL请求执行的时延,单位为纳秒。
Latency in nanoseconds of SQL request execution
sql.txn.abort.count SQL事务中中BEGIN语句的数量。的数量。
Number of SQL transaction ABORT statements
sql.txn.begin.count SQL事务中ROLLBACK语句的数量。
Number of SQL transaction BEGIN statements
sql.txn.commit.count SQL事务中COMMIT语句的数量。
Number of SQL transaction COMMIT statements
sql.txn.rollback.count UPDATE语句的数量。
Number of SQL transaction ROLLBACK statements
sql.update.count UPDATE语句的数量。
Number of SQL UPDATE statements
sys.cgo.allocbytes 当前当前cgo分配的内存大小。
Current bytes of memory allocated by cgo
sys.cgo.totalbytes cgo分配的内存总的累积值累积值。
Total bytes of memory allocated by cgo, but not released
sys.cgocalls cgo调用的总总次数。。
Total number of cgo call
sys.cpu.sys.ns 用户CPU时间总和,单位为纳秒。
Total system cpu time in nanoseconds
sys.cpu.sys.percent 当前系统CPU时间所占时间所占比例。
Current system cpu percentage
sys.cpu.user.ns 用户CPU时间总和,单位为纳秒。
Total user cpu time in nanoseconds
sys.cpu.user.percent 当前用户CPU时间所占比例。
Current user cpu percentage
sys.fd.open 进程进程打开的文件描述符的数量。数量。
Process open file descriptors
sys.fd.softlimit 进程打开的文件描述符的软限制。
Process open FD soft limit
sys.gc.count GC执行的次数。,单位为
Total number of GC runs
sys.gc.pause.ns GC暂停的时间,单位为纳秒。纳秒。内存总字节数。
Total GC pause in nanoseconds
sys.gc.pause.percent GC暂停时间的占比。占比。
Current GC pause percentage
sys.go.allocbytes 当前go分配的内存字节数。当前进程的RSS值。
Current bytes of memory allocated by go
sys.go.totalbytes go分配的内存的累积值。
Total bytes of memory allocated by go, but not released
sys.goroutines 当前当前goroutine的数量。goroutine的数量。
Current number of goroutines
sys.rss 当前进程的RSS值。
Current process RSS
sys.uptime 当前进程的运行时间。
Process uptime in seconds
sysbytes 系统内部键值对的字节数。
Number of bytes in system KV pairs
syscount 系统内部键值对的数量。
Count of system KV pairs
timeseries.write.bytes 持久化的指标数据的样本数。
Total size in bytes of metric samples written to disk
timeseries.write.errors 指标数据写硬盘遇到遇到的错误的数量。
Total errors encountered while attempting to write metrics to disk
timeseries.write.samples 持久化的持久化的指标数据的数量。
Total number of metric samples written to disk
totalbytes 所有键值对数据的字节数。。
Total number of bytes taken up by keys and values including non-live data
tscache.skl.read.pages 读时间戳缓存中页的的数量。旋转的次数。
Number of pages in the read timestamp cache
tscache.skl.read.rotations 写时间戳缓存中页旋转的次数。
Number of page rotations in the read timestamp cache
tscache.skl.write.pages 写时间戳缓存中页的数量。数量。
Number of pages in the write timestamp cache
tscache.skl.write.rotations 写时间戳缓存中页旋转的次数。数量。
Number of page rotations in the write timestamp cache
txn.abandons 废弃的事务数量事务数量
Number of abandoned KV transactions
txn.aborts 中止的KV事务数量事务数量
Number of aborted KV transactions
txn.autoretries K事务事务自动重试的次数。
Number of automatic retries to avoid serializable restarts
txn.commits1PC 已提交的1PC KV事务次数。
Number of committed one-phase KV transactions
txn.commits 已提交的KV事务数量。。
Number of committed KV transactions (including 1PC)
txn.durations KV事务的时间,单位为纳秒。。
KV transaction durations in nanoseconds
txn.restarts.deleterange 在可串行化的隔离级别下,事务因其他事务更早的提交时间戳导致重启的次数。
Number of restarts due to a forwarded commit timestamp and a DeleteRange command
txn.restarts.possiblereplay 事务事务在存储层因命令batch可能的重播导致重启的次数。

Number of restarts due to possible replays of command batches at the storage layer
txn.restarts.serializable 在可串行化的隔离级别下,事务因其他事务更早的提交时间戳导致重启的次数。使用的字节数。
Number of restarts due to a forwarded commit timestamp and isolation=SERIALIZABLE
txn.restarts.writetooold 事务因其他并行写先提交而导致重启的次数。
Number of restarts due to a concurrent writer committing first
txn.restarts KV事务重启的次数。
Number of restarted KV transactions
valbytes 存储存储的Value值的字节数。
Number of bytes taken up by values
valcount 存储的Value值的数量。
Count of all values