Red Hat values our customer experience and privacy. It is important to us that our customers understand exactly what we are sending back to engineering and why. During the developer preview or beta release of our software, we want to be able to make changes to our designs and coding practices in real-time based on customer environments. The faster the feedback loop during these development stages the better.
For the OpenShift 4 Developer Preview we will be sending back these exact attributes based on your cluster ID and pull secret from Red Hat:
apiVersion: v1
data:
metrics.yaml: |-
matches:
# cluster:usage recording rules summarize important usage information
# about the cluster that points to specific features or component usage
# that may help identify problems or specific workloads. For example,
# cluster:usage:openshift:build:rate24h would show the number of builds
# executed within a 24h period so as to determine whether the current
# cluster is using builds and may be susceptible to eviction due to high
# disk usage from build temporary directories.
# All metrics under this prefix must have low (1-5) cardinality and must
# be well-scoped and follow proper naming and scoping conventions.
- '{__name__=~"cluster:usage:.*"}'
# count:up0 contains the count of cluster monitoring sources being marked as down.
# This information is relevant to the health of the registered
# cluster monitoring sources on a cluster. This metric allows telemetry
# to identify when an update causes a service to begin to crash-loop or
# flake.
- '{__name__="count:up0"}'
# count:up1 contains the count of cluster monitoring sources being marked as up.
# This information is relevant to the health of the registered
# cluster monitoring sources on a cluster. This metric allows telemetry
# to identify when an update causes a service to begin to crash-loop or
# flake.
- '{__name__="count:up1"}'
# cluster_version reports what payload and version the cluster is being
# configured to and is used to identify what versions are on a cluster
# that is experiencing problems.
- '{__name__="cluster_version"}'
# cluster_version_available_updates reports the channel and version
# server the cluster is configured to use and how many updates are
# available. This is used to ensure that updates are being properly
# served to clusters.
- '{__name__="cluster_version_available_updates"}'
# cluster_operator_up reports the health status of the core cluster
# operators - like up, an upgrade that fails due to a configuration value
# on the cluster will help narrow down which component is affected.
- '{__name__="cluster_operator_up"}'
# cluster_operator_conditions exposes the status conditions cluster
# operators report for debugging. The condition and status are reported.
- '{__name__="cluster_operator_conditions"}'
# cluster_version_payload captures how far through a payload the cluster
# version operator has progressed and can be used to identify whether
# a particular payload entry is causing failures during upgrade.
- '{__name__="cluster_version_payload"}'
# cluster_installer reports what installed the cluster, along with its
# version number and invoker.
- '{__name__="cluster_installer"}'
# cluster_infrastructure_provider reports the configured cloud provider
# if any, along with the infrastructure region when running in the public
# cloud.
- '{__name__="cluster_infrastructure_provider"}'
# cluster_feature_set reports the configured cluster feature set and
# whether the feature set is considered supported or unsupported.
- '{__name__="cluster_feature_set"}'
# instance:etcd_object_counts:sum identifies two key metrics:
# - the rough size of the data stored in etcd and
# - the consistency between the etcd instances.
- '{__name__="instance:etcd_object_counts:sum"}'
# alerts are the key summarization of the system state. They are
# reported via telemetry to assess their value in detecting
# upgrade failure causes and also to prevent the need to gather
# large sets of metrics that are already summarized on the cluster.
# Reporting alerts also creates an incentive to improve per
# cluster alerting for the purposes of preventing upgrades from
# failing for end users.
- '{__name__="ALERTS",alertstate="firing"}'
# the following three metrics will be used for SLA analysis reports.
# code:apiserver_request_count:rate:sum identifies average of occurances
# of each http status code over 10 minutes
- '{__name__="code:apiserver_request_count:rate:sum"}'
# cluster:capacity_cpu_cores:sum is the total number of CPU cores
# in the cluster labeled by node role and type.
- '{__name__="cluster:capacity_cpu_cores:sum"}'
# cluster:capacity_memory_bytes:sum is the total bytes of memory
# in the cluster labeled by node role and type.
- '{__name__="cluster:capacity_memory_bytes:sum"}'
# cluster:cpu_usage_cores:sum is the current amount of CPU in
# use across the whole cluster.
- '{__name__="cluster:cpu_usage_cores:sum"}'
# cluster:memory_usage_bytes:sum is the current amount of memory in
# use across the whole cluster.
- '{__name__="cluster:memory_usage_bytes:sum"}'
# openshift:cpu_usage_cores:sum is the current amount of CPU
# used by OpenShift components, including the control plane and
# host services (including the kernel).
- '{__name__="openshift:cpu_usage_cores:sum"}'
# openshift:memory_usage_bytes:sum is the current amount of memory
# used by OpenShift components, including the control plane and
# host services (including the kernel).
- '{__name__="openshift:memory_usage_bytes:sum"}'
# workload:cpu_usage_cores:sum is the current amount of CPU
# used by cluster workloads, excluding infrastructure.
- '{__name__="workload:cpu_usage_cores:sum"}'
# workload:memory_usage_bytes:sum is the current amount of memory
# used by cluster workloads, excluding infrastructure.
- '{__name__="workload:memory_usage_bytes:sum"}'
# cluster:virt_platform_nodes:sum is the number of nodes reporting
# a particular virt_platform type (nodes may report multiple types).
# This metric helps identify issues specific to a virtualization
# type or bare metal.
- '{__name__="cluster:virt_platform_nodes:sum"}'
# cluster:node_instance_type_count:sum is the number of nodes
# of each instance type and role.
- '{__name__="cluster:node_instance_type_count:sum"}'
# cnv:vmi_status_running:count is the total number of VM instances running in the cluster.
- '{__name__="cnv:vmi_status_running:count"}'
# node_role_os_version_machine:cpu_capacity_cores:sum is the total number of CPU cores
# in the cluster labeled by master and/or infra node role, os, architecture, and hyperthreading state.
- '{__name__="node_role_os_version_machine:cpu_capacity_cores:sum"}'
# node_role_os_version_machine:cpu_capacity_sockets:sum is the total number of CPU sockets
# in the cluster labeled by master and/or infra node role, os, architecture, and hyperthreading state.
- '{__name__="node_role_os_version_machine:cpu_capacity_sockets:sum"}'
# subscription_sync_total is the number of times an OLM operator
# Subscription has been synced, labelled by name and installed csv
- '{__name__="subscription_sync_total"}'
# csv_succeeded is unique to the namespace, name, version, and phase labels.
# The metrics is always present and can be equal to 0 or 1, where 0 represents that the
# csv is not in the succeeded state while 1 represents that the csv is in the succeeded state.
- '{__name__="csv_succeeded"}'
# csv_abnormal represents the reason why a csv is not in the succeeded state and includes the
# namespace, name, version, phase, reason labels. When a csv is updated, the previous time series associated with the csv will be deleted.
- '{__name__="csv_abnormal"}'
# OCS metrics to be collected:
# ceph_cluster_total_bytes gives the size of ceph cluster in bytes.
- '{__name__="ceph_cluster_total_bytes"}'
# ceph_cluster_total_used_raw_bytes is the amount of ceph cluster storage used in bytes.
- '{__name__="ceph_cluster_total_used_raw_bytes"}'
# ceph_health_status gives the ceph cluster health status
- '{__name__="ceph_health_status"}'
# job:ceph_osd_metadata:count is the total count of osds.
- '{__name__="job:ceph_osd_metadata:count"}'
# job:kube_pv:count is the total number of Persistent Volumes present in OCP cluster.
- '{__name__="job:kube_pv:count"}'
# job:ceph_pools_iops:total is the total iops (reads+writes) value for all the pools in ceph cluster
- '{__name__="job:ceph_pools_iops:total"}'
# job:ceph_pools_iops:total is the total iops (reads+writes) value in bytes for all the pools in ceph cluster
- '{__name__="job:ceph_pools_iops_bytes:total"}'
# job:ceph_versions_running:count is the total count of ceph cluster versions running.
- '{__name__="job:ceph_versions_running:count"}'
# job:noobaa_total_unhealthy_buckets:sum is the total number of unhealthy noobaa buckets
- '{__name__="job:noobaa_total_unhealthy_buckets:sum"}'
# job:noobaa_bucket_count:sum is the total number of noobaa buckets
- '{__name__="job:noobaa_bucket_count:sum"}'
# job:noobaa_total_object_count:sum is the total number of noobaa objects
- '{__name__="job:noobaa_total_object_count:sum"}'
# noobaa_accounts_num gives the count of noobaa's accounts.
- '{__name__="noobaa_accounts_num"}'
# noobaa_total_usage gives the total usage of noobaa's storage in bytes.
- '{__name__="noobaa_total_usage"}'
# console_url is the url of the console running on the cluster.
- '{__name__="console_url"}'
# cluster:network_attachment_definition_instances:max" gives max no of instance
# in the cluster that are annotated with k8s.v1.cni.cncf.io/networks, labelled by networks.
- '{__name__="cluster:network_attachment_definition_instances:max"}'
# cluster:network_attachment_definition_enabled_instance_up informs (1 or 0) if the cluster has
# at least max of one instance with k8s.v1.cni.cncf.io/networks annotation, labelled by networks (any or sriov).
- '{__name__="cluster:network_attachment_definition_enabled_instance_up:max"}'
# insightsclient_request_send tracks the number of metrics sends.
- '{__name__="insightsclient_request_send_total"}'
# cam_app_workload_migrations tracks number of app workload migrations
# by current state. Tracked migration states are idle, running, completed, and failed.
- '{__name__="cam_app_workload_migrations"}'
# cluster:apiserver_current_inflight_requests:sum:max_over_time:2m gives maximum number of requests in flight
# over a 2-minute window. This metric is a constant 4 time series that monitors concurrency of kube-apiserver and
# openshift-apiserver with request type which can be either 'mutating' or 'readonly'.
# We want to have an idea of how loaded our api server(s) are globally.
- '{__name__="cluster:apiserver_current_inflight_requests:sum:max_over_time:2m"}'
# (monitoring-team) cluster:telemetry_selected_series:count tracks the total number of series
# sent via telemetry from each cluster.
- '{__name__="cluster:telemetry_selected_series:count"}'
kind: ConfigMap
metadata:
name: telemetry-config
namespace: openshift-monitoringThese attributes are focused on the health of the cluster based on the CPU/MEM environmental attributes. From this telemetry we hope to be able to determine the immediate functionality of the framework components and whether or not we have a correlation of issues across similar developer preview environmental characteristics. This information will allow us to immediately make changes to the OpenShift solution to improve our customer's experience and software resiliency.
We are extremely excited about showing you where the product is headed during this developer preview and we hope you will allow us this information to enhance the solution for all those involved.