Skip to content

Latest commit

 

History

History
238 lines (189 loc) · 9.46 KB

cloudwatch_adapter_metrics_events.md

File metadata and controls

238 lines (189 loc) · 9.46 KB
Raw

Cloudwatch Adapter Metrics and Events

This page describes the specific metrics and events supported by the AWS Adapter.

All Products

All events and metrics include a field product=<product>, which identifies the AWS Product related to the metric or event. <product> can be one of the following:

  • EC2
  • ELB
  • RDS
  • EBS
  • CloudFront
  • AutoScaling
  • ElastiCache
  • Lambda

All metrics include a field metric_type=<type> with the value AWS CloudWatch. The AWS Adapter defines additional metric_type fields.

All metrics also include a field namespace=<namespace> which has one of the following values:

  • AWS/EC2
  • AWS/EBS
  • AWS/ELB
  • AWS/RDS
  • AWS/CloudFront
  • AWS/AutoScaling
  • AWS/ElastiCache
  • AWS/Lambda

The namespace is generally the same as the product. The one exception is AutoScaling, which can return metrics in both the AWS/EC2 and AWS/AutoScaling namespaces.

All metrics also include a field units which names the units of the value field in the point.

All metrics include fields dimension=<dimension> and item=<item> which describe the item for which the CloudWatch metrics are calculated. This version of the adapter aggregates metrics by items for a given product (EC2 instance id, EBS volume, etc).

CloudWatch Metrics

The CloudWatch API documentation has a complete description of the metrics for each product.

###EC2

Also see the EC2 CloudWatch documentation.

metric_type="AWS CloudWatch", namespace="AWS/EC2", dimension="InstanceId", item=<instance-id>, plus the following:

  • name="CPUCreditUsage", value=<count>: The CPU Credit Usage over the last measurement period.
  • name="CPUCreditBalance", value=<count>: The CPU Credit Balance over the last measurement period.
  • name="CPUUtilization", value=<percent>
  • name="DiskReadOps", value=<count>
  • name="DiskWriteOps", value=<count>
  • name="DiskReadBytes", value=<bytes>
  • name="DiskWriteBytes", value=<bytes>
  • name="NetworkIn", value=<bytes>
  • name="NetworkOut", value=<bytes>
  • name="StatusCheckFailed_Instance", value=<count>: This metric will only be reported if the value is non-zero.
  • name="StatusCheckFailed_System", value=<count>: This metric will only be reported if the value is non-zero.

###EBS

Also see the EBS CloudWatch documentation.

metric_type="AWS CloudWatch", namespace="AWS/EBS", dimension="VolumeId", item=<volume-id>, plus the following:

  • name="VolumeReadBytes", value=<bytes>
  • name="VolumeWriteBytes", value=<bytes>
  • name="VolumeReadOps", value=<count>
  • name="VolumeWriteOps", value=<count>
  • name="VolumeTotalReadTime", value=<seconds>
  • name="VolumeTotalWriteTime", value=<seconds>
  • name="VolumeIdleTime", value=<seconds>
  • name="VolumeQueueLength", value=<count>
  • name="VolumeThroughputPercentage", value=<percent>
  • name="VolumeConsumedReadWriteOps", value=<count>

###ELB

Also see the ELB CloudWatch documentation.

metric_type="AWS CloudWatch", namespace="AWS/ELB", dimension="LoadBalancerName", item=<load-balancer-name>, plus the following:

  • name="HealthyHostCount", value=<count>
  • name="UnHealthyHostCount", value=<count>
  • name="RequestCount", value=<count>
  • name="Latency", value=<seconds>
  • name="HTTPCode_ELB_4XX", value=<count>
  • name="HTTPCode_ELB_5XX", value=<count>
  • name="HTTPCode_Backend_2XX", value=<count>
  • name="HTTPCode_Backend_3XX", value=<count>
  • name="HTTPCode_Backend_4XX", value=<count>
  • name="HTTPCode_Backend_5XX", value=<count>
  • name="BackendConnectionErrors", value=<count>
  • name="SurgeQueueLength", value=<count>
  • name="SpilloverCount", value=<count>

##RDS

Also see the RDS CloudWatch documentation.

metric_type="AWS CloudWatch", namespace="AWS/RDS", dimension="DBInstanceIdentifier", item=<db-instance-name>, plus the following:

  • name="BinLogDiskUsage", value=<bytes>
  • name="CPUUtilization", value=<percent>
  • name="CPUCreditUsage", value=<count>
  • name="CPUCreditBalance", value=<count>
  • name="DatabaseConnections", value=<count>
  • name="DiskQueueDepth", value=<count>
  • name="FreeableMemory", value=<bytes>
  • name="FreeStorageSpace", value=<bytes>
  • name="ReplicaLag", value=<seconds>
  • name="SwapUsage", value=<bytes>
  • name="ReadIOPS", value=<count/second>
  • name="WriteIOPS", value=<count/second>
  • name="ReadLatency", value=<seconds>
  • name="WriteLatency", value=<seconds>
  • name="ReadThroughput", value=<bytes/second>
  • name="WriteThroughput", value=<bytes/second>
  • name="NetworkReceiveThroughput", value=<bytes/second>
  • name="NetworkTransmitThroughput", value=<bytes/second>

###CloudFront

Also see the CloudFront CloudWatch documentation.

metric_type="AWS CloudWatch", namespace="AWS/CloudFront", dimension="DistributionId", item=<distribution-id>, plus the following:

  • name="Requests", value=<count>
  • name="BytesDownloaded", value=<bytes>
  • name="BytesUploaded", value=<bytes>
  • name="TotalErrorRate", value=<percent>
  • name="4xxErrorRate", value=<percent>
  • name="5xxErrorRate", value=<percent>

###AutoScaling

Also see the AutoScaling CloudWatch documentation.

metric_type="AWS CloudWatch", namespace="AWS/AutoScaling", dimension="AutoScalingGroupName", item=<group-name>, plus the following:

  • name="GroupMinSize", value=<count>
  • name="GroupMaxSize", value=<count>
  • name="GroupDesiredCapacity", value=<count>
  • name="GroupInServiceInstances", value=<count>
  • name="GroupPendingInstances", value=<count>
  • name="GroupStandbyInstances", value=<count>
  • name="GroupTerminatingInstances", value=<count>
  • name="GroupTotalInstances", value=<count>

###ElastiCache

Also see the ElastiCache CloudWatch documentation.

metric_type="AWS CloudWatch", namespace="AWS/ElastiCache", dimension="CacheClusterId", item=<cluster-id>, plus the following:

The following CloudWatch metrics are available for all engine types:

  • name="CPUUtilization", value=<percent>
  • name="FreeableMemory", value=<bytes>
  • name="NetworkBytesIn", value=<bytes>
  • name="NetworkBytesOut", value=<bytes>
  • name="SwapUsage", value=<bytes>

The following additional CloudWatch metrics are available for redis engines:

  • name="BytesUsedForCache", value=<bytes>
  • name="CacheHits", value=<count>
  • name="CacheMisses", value=<count>
  • name="CurrConnections", value=<count>
  • name="Evictions", value=<count>
  • name="HyperLogLogBasedCmds", value=<count>
  • name="NewConnections", value=<count>
  • name="Reclaimed", value=<count>
  • name="ReplicationBytes", value=<bytes>
  • name="ReplicationLag", value=<seconds>
  • name="SaveInProgress", value=<count>
  • name="CurrItems", value=<count>
  • name="GetTypeCmds", value=<count>
  • name="HashBasedCmds", value=<count>
  • name="KeyBasedCmds", value=<count>
  • name="ListBasedCmds", value=<count>
  • name="SetBasedCmds", value=<count>
  • name="SetTypeCmds", value=<count>
  • name="SortedSetBasedCmds", value=<count>
  • name="StringBasedCmds", value=<count>
  • name="BytesUsedForCache", value=<bytes>

The following additional CloudWatch metrics are available for memcached engines:

  • name="BytesUsedForCacheItems", value=<bytes>
  • name="BytesReadIntoMemcached", value=<bytes>
  • name="BytesWrittenOutFromMemcached", value=<bytes>
  • name="CasBadval", value=<count>
  • name="CasHits", value=<count>
  • name="CasMisses", value=<count>
  • name="CmdFlush", value=<count>
  • name="CmdGet", value=<count>
  • name="CmdSet", value=<count>
  • name="CurrConnections", value=<count>
  • name="CurrItems", value=<count>
  • name="DecrHits", value=<count>
  • name="DecrMisses", value=<count>
  • name="DeleteHits", value=<count>
  • name="DeleteMisses", value=<count>
  • name="Evictions", value=<count>
  • name="GetHits", value=<count>
  • name="GetMisses", value=<count>
  • name="IncrHits", value=<count>
  • name="IncrMisses", value=<count>
  • name="Reclaimed", value=<count>
  • name="BytesUsedForHash", value=<bytes>
  • name="CmdConfigGet", value=<count>
  • name="CmdConfigSet", value=<count>
  • name="CmdTouch", value=<count>
  • name="CurrConfig", value=<count>
  • name="EvictedUnfetched", value=<count>
  • name="ExpiredUnfetched", value=<count>
  • name="SlabsMoved", value=<count>
  • name="TouchHits", value=<count>
  • name="TouchMisses", value=<count>
  • name="NewConnections", value=<count>
  • name="NewItems", value=<count>
  • name="UnusedMemory", value=<bytes>

###Lambda

Also see the Lambda CloudWatch documentation.

metric_type="AWS CloudWatch", namespace="AWS/Lambda", dimension="FunctionName", item=<function-name>, plus the following:

  • name="Invocations", value=<count>
  • name="Errors", value=<count>
  • name="Duration", value=<milliseconds>
  • name="Throttles", value=<count>