Monitoring AWS Resources

Introduced在GitLab 9.4

GitLab has support for automatically detecting and monitoring AWS resources, starting with theElastic Load Balancer. This is provided by leveraging the officialCloudwatch exporter, which translatesCloudwatch metricsinto a Prometheus readable form.

Requirements

ThePrometheus servicemust be enabled.

Metrics supported

Name	Query
Throughput (req/sec)	sum(aws_elb_request_count_sum{%{environment_filter}}) / 60
Latency (ms)	avg(aws_elb_latency_average{%{environment_filter}}) * 1000
HTTP Error Rate (%)	sum(aws_elb_httpcode_backend_5_xx_sum{%{environment_filter}}) / sum(aws_elb_request_count_sum{%{environment_filter}})

Configuring Prometheus to monitor for Cloudwatch metrics

To get started with Cloudwatch monitoring, you should install and configure theCloudwatch exporterwhich retrieves and parses the specified Cloudwatch metrics and translates them into a Prometheus monitoring endpoint.

Right now, the only AWS resource supported is the Elastic Load Balancer, whose Cloudwatch metrics can be found在这里.

一个示例Cloudwatch出口er configuration file, configured for basic AWS ELB monitoring, isavailable for download.

Specifying the Environment label

In order to isolate and only display relevant metrics for a given environment however, GitLab needs a method to detect which labels are associated. To do this, GitLab willlook for anenvironmentlabel.