banzaicloud-stable/thanos

Chart version: 0.3.31
Api version: v1
App version: 0.17.1
Thanos is a set of components that can be composed into a highl...
application
Chart Type
Active
Status
Unknown
License
5908
Downloads
https://kubernetes-charts.banzaicloud.com
Set me up:
helm repo add center https://repo.chartcenter.io
Install Chart:
helm install thanos center/banzaicloud-stable/thanos
Versions (0)

Thanos Helm chart

This is a Helm Chart for Thanos. It does not include the required Prometheus and sidecar installation.

Thanos

Thanos is a set of components that can be composed into a highly available metric system with unlimited storage capacity, which can be added seamlessly on top of existing Prometheus deployments.

Thanos leverages the Prometheus 2.0 storage format to cost-efficiently store historical metric data in any object storage while retaining fast query latencies. Additionally, it provides a global query view across all Prometheus installations and can merge data from Prometheus HA pairs on the fly.

Concretely the aims of the project are:

  • Global query view of metrics.
  • Unlimited retention of metrics.
  • High availability of components, including Prometheus.

Helm Chart

This chart is in Beta state to provide easy installation via Helm chart. Things that we are improving in near future: - [ ] Automatic TLS generation for communicating between in-cluster components - [ ] Support for tracing configuration - [ ] Grafana dashboards - [ ] Informative NOTES.txt

Architecture

This Chart will install a complete Thanos solution. To understand how Thanos works please read it’s official Architecture design.

Installing the Chart

Add Banzai Cloud repository:

$ helm repo add banzaicloud-stable https://kubernetes-charts.banzaicloud.com

Storage examples

Example GCS configuration for object-store.yaml

type: GCS
config:
  bucket: "thanos"
  service_account: |-
    {
      "type": "service_account",
      "project_id": "project",
      "private_key_id": "abcdefghijklmnopqrstuvwxyz12345678906666",
      "private_key": "-----BEGIN PRIVATE KEY-----\...\n-----END PRIVATE KEY-----\n",
      "client_email": "project@thanos.iam.gserviceaccount.com",
      "client_id": "123456789012345678901",
      "auth_uri": "https://accounts.google.com/o/oauth2/auth",
      "token_uri": "https://oauth2.googleapis.com/token",
      "auth_provider_x509_cert_url": "https://www.googleapis.com/oauth2/v1/certs",
      "client_x509_cert_url": "https://www.googleapis.com/robot/v1/metadata/x509/thanos%40gitpods.iam.gserviceaccount.com"
    }

Example S3 configuration for object-store.yaml

This is an example configuration using thanos with S3. Check endpoints here: https://docs.aws.amazon.com/general/latest/gr/rande.html#s3_region

type: S3
config:
  bucket: ""
  endpoint: ""
  region: ""
  access_key: ""
  insecure: false
  signature_version2: false
  secret_key: ""
  put_user_metadata: {}
  http_config:
    idle_conn_timeout: 0s
    response_header_timeout: 0s
    insecure_skip_verify: false
  trace:
    enable: false
  part_size: 0

Example Azure configuration for object-store.yaml

type: AZURE
config:
  storage_account: ""
  storage_account_key: ""
  container: ""
  endpoint: ""
  max_retries: 0

Create the Service Account and Bucket at Google cloud.

Install the chart:

helm install banzaicloud-stable/thanos --name thanos -f my-values.yaml --set-file objstoreFile=object-store.yaml

Visit the Bucket browser

kubectl port-forward svc/thanos-bucket 8080 &
open http://localhost:8080

Install prometheus-operator

Extra configuration for prometheus operator.

Note: Prometheus-operator and Thanos MUST be in the same namespace.

prometheus:
  prometheusSpec:
    thanos:
      image: quay.io/thanos/thanos:v0.9.0
      version: v0.9.0
      objectStorageConfig:
        name: thanos
        key: object-store.yaml

Install prometheus-operator

helm install stable/prometheus-operator -f thanos-sidecar.yaml

Configuration

This section describes the values available

General

Name Description Default Value
image.repository Thanos image repository and name ‘quay.io/thanos/thanos’ For Thanos version 0.6.0 or older change this to ‘improbable/thanos’
image.tag Thanos image tag v0.9.0
image.pullPolicy Image Kubernetes pull policy IfNotPresent
objstore  Configuration for the backend object storage in yaml format. Mutually exclusive with other objstore options. {}
objstoreFile  Configuration for the backend object storage in string format. Mutually exclusive with other objstore options. ””
objstoreSecretOverride  Configuration for the backend object storage in an existing secret. Mutually exclusive with other objstore options. ””

Common settings for all components

These setting applicable to nearly all components.

Name Description Default Value
$component.labels Additional labels to the Pod {}
$component.annotations Additional annotations to the Pod {}
$component.deploymentLabels Additional labels to the deployment {}
$component.deploymentAnnotations Additional annotations to the deployment {}
$component.extraEnv Add extra environment variables []
$component.strategy Kubernetes deployment update strategy object {}
$component.updateStrategy Kubernetes statefulset update strategy object {}
$component.metrics.annotations.enabled Prometheus annotation for component false
$component.metrics.serviceMonitor.enabled Prometheus ServiceMonitor definition for component false
$component.securityContext SecurityContext for Pod {}
$component.resources Resource definition for container {}
$component.tolerations Node tolerations for server scheduling to nodes with taints {}
$component.nodeSelector Node labels for compact pod assignment {}
$component.affinity Pod affinity {}
$component.grpc.port grpc listen port number  10901
$component.grpc.service.annotations Service definition for grpc service {}
$component.grpc.service.matchLabels Pod label selector to match grpc service on. {}
$component.grpc.ingress.enabled Set up ingress for the grpc service false
$component.grpc.ingress.defaultBackend Set up default backend for ingress false
$component.grpc.ingress.annotations Add annotations to ingress {}
$component.grpc.ingress.labels Add labels to ingress {}
$component.grpc.ingress.path Ingress path ”/”
$component.grpc.ingress.hosts Ingress hosts []
$component.grpc.ingress.tls Ingress TLS configuration []
$component.http.port http listen port number  10902
$component.http.service.annotations Service definition for http service {}
$component.http.service.matchLabels Pod label selector to match http service on. {}
$component.http.ingress.enabled Set up ingress for the http service false
$component.http.ingress.apiVersion Set API version for ingress extensions/v1beta1
$component.http.ingress.defaultBackend Set up default backend for ingress false
$component.http.ingress.annotations Add annotations to ingress {}
$component.http.ingress.labels Add labels to ingress {}
$component.http.ingress.path Ingress path ”/”
$component.http.ingress.hosts Ingress hosts []
$component.http.ingress.tls Ingress TLS configuration []

Store

These values are just samples, for more fine-tuning please check the values.yaml.

Name Description Default Value
store.enabled Enable component true
store.replicaCount Pod replica count 1
store.logLevel Log level info
store.logFormat Log format to use. Possible options: logfmt or json. logfmt
store.indexCacheSize Maximum size of items held in the index cache. 250MB
store.chunkPoolSize Maximum size of concurrently allocatable bytes for chunks. 2GB
store.grpcSeriesSampleLimit Maximum amount of samples returned via a single series call. 0 means no limit. NOTE: for efficiency we take 120 as the number of samples in chunk (it cannot be bigger than that), so the actual number of samples might be lower, even though the maximum could be hit. 0
store.grpcSeriesMaxConcurrency Maximum number of concurrent Series calls. 20
store.syncBlockDuration Repeat interval for syncing the blocks between local and remote view. 3m
store.blockSyncConcurrency Number of goroutines to use when syncing blocks from object storage. 20
store.extraEnv Add extra environment variables []
store.extraArgs Add extra arguments []
store.serviceAccount Name of the Kubernetes service account to use ””
store.livenessProbe Set up liveness probe for store available for Thanos v0.8.0+) {}
store.readinessProbe Set up readinessProbe for store (available for Thanos v0.8.0+) {}
timePartioning list of min/max time for store partitions. See more details below. Setting this will create mutlipale thanos store deployments based on the number of items in the list [{min: “”, max: “”}]
hashPartioning.shards The number of shared used to partition the blocks based on the hashmod of the blocks. Can not be used with time partitioning ””
initContainers InitContainers allows injecting specialized containers that run before app containers. This is meant to pre-configure and tune mounted volume permissions. []

Store time partions

Thanos store supports partition based on time. Setting time partitions will create n number of store deployment based on the number of items in the list. Each item must contain min and max time for querying in the supported format (see details here See details at https://thanos.io/components/store.md/#time-based-partioning ). Leaving this empty list ([]) will create a single store for all data. Example - This will create 3 stores:

timePartioning:
  # One store for data older than 6 weeks
  - min: ""
    max: -6w
  # One store for data newer than 6 weeks and older than 2 weeks
  - min: -6w
    max: -2w
  # One store for data newer than 2 weeks
  - min: -2w
    max: ""

Query

Name Description Default Value
query.enabled Enable component true
query.replicaCount Pod replica count 1
query.logLevel Log level info
query.logFormat Log format to use. Possible options: logfmt or json. logfmt
query.replicaLabels Labels to treat as a replica indicator along which data is deduplicated. Still you will be able to query without deduplication using ‘dedup=false’ parameter. []
query.autoDownsampling Enable –query.auto-downsampling option for query. true
query.webRoutePrefix Prefix for API and UI endpoints. This allows thanos UI to be served on a sub-path. This option is analogous to –web.route-prefix of Promethus. ””
query.webExternalPrefix Static prefix for all HTML links and redirect URLs in the UI query web interface. Actual endpoints are still served on / or the web.route-prefix. This allows thanos UI to be served behind a reverse proxy that strips a URL sub-path ””
query.webPrefixHeader Name of HTTP request header used for dynamic prefixing of UI links and redirects. This option is ignored if web.external-prefix argument is set. Security risk: enable this option only if a reverse proxy in front of thanos is resetting the header. The –web.prefix-header=X-Forwarded-Prefix option can be useful, for example, if Thanos UI is served via Traefik reverse proxy with PathPrefixStrip option enabled, which sends the stripped prefix value in X-Forwarded-Prefix header. This allows thanos UI to be served on a sub-path ””
query.storeDNSResolver Custome DNS resolver because of issue miekgdns
query.storeDNSDiscovery Enable DNS discovery for stores true
query.sidecarDNSDiscovery Enable DNS discovery for sidecars (this is for the chart built-in sidecar service) true
query.stores Addresses of statically configured store API servers (repeatable). The scheme may be prefixed with ‘dns+’ or ‘dnssrv+’ to detect store API servers through respective DNS lookups. []
query.serviceDiscoveryFiles Path to files that contains addresses of store API servers. The path can be a glob pattern (repeatable). []
query.serviceDiscoveryFileConfigMaps Names of configmaps that contain addresses of store API servers, used for file service discovery. []
query.serviceDiscoveryInterval Refresh interval to re-read file SD files. It is used as a resync fallback. 5m
query.extraEnv Add extra environment variables []
query.extraArgs Add extra arguments []
query.podDisruptionBudget.enabled Enabled and config podDisruptionBudget resource for this component false
query.podDisruptionBudget.minAvailable Minimum number of available query pods for PodDisruptionBudget 1
query.podDisruptionBudget.maxUnavailable Maximum number of unavailable query pods for PodDisruptionBudget []
query.autoscaling.enabled Enabled and config horizontalPodAutoscaling resource for this component false
query.autoscaling.minReplicas If autoscaling enabled, this field sets minimum replica count 2
query.autoscaling.maxReplicas If autoscaling enabled, this field sets maximum replica count 3
query.autoscaling.targetCPUUtilizationPercentage Target CPU utilization percentage to scale 50
query.autoscaling.targetMemoryUtilizationPercentage Target memory utilization percentage to scale 50
query.serviceAccount Name of the Kubernetes service account to use ””
query.psp.enabled Enable pod security policy, it also requires the query.rbac.enabled to be set to true. false
query.rbac.enabled Enable RBAC to use the PSP false
query.livenessProbe Set up liveness probe for query {}
query.readinessProbe Set up readinessProbe for query {}

Rule

Name Description Default Value
rule.enabled Enable component false
rule.logLevel Log level info
rule.logFormat Log format to use. Possible options: logfmt or json. logfmt
rule.ruleLabels Labels to be applied to all generated metrics (repeated). Similar to external labels for Prometheus, used to identify ruler and its blocks as unique source. {}
rule.resendDelay Minimum amount of time to wait before resending an alert to Alertmanager. ””
rule.evalInterval The default evaluation interval to use. ””
rule.tsdbBlockDuration Block duration for TSDB block. ””
rule.tsdbRetention Block retention time on local disk. ””
rule.webRoutePrefix Prefix for API and UI endpoints. This allows thanos UI to be served on a sub-path. This option is analogous to –web.route-prefix of Promethus. ””
rule.webExternalPrefix Static prefix for all HTML links and redirect URLs in the UI query web interface. Actual endpoints are still served on / or the web.route-prefix. This allows thanos UI to be served behind a reverse proxy that strips a URL sub-path ””
rule.webPrefixHeader Name of HTTP request header used for dynamic prefixing of UI links and redirects. This option is ignored if web.external-prefix argument is set. Security risk: enable this option only if a reverse proxy in front of thanos is resetting the header. The –web.prefix-header=X-Forwarded-Prefix option can be useful, for example, if Thanos UI is served via Traefik reverse proxy with PathPrefixStrip option enabled, which sends the stripped prefix value in X-Forwarded-Prefix header. This allows thanos UI to be served on a sub-path ””
rule.queryDNSDiscovery Enable DNS discovery for query insances true
rule.alertmanagers # Alertmanager replica URLs to push firing alerts. Ruler claims success if push to at least one alertmanager from discovered succeeds. The scheme may be prefixed with ‘dns+’ or ‘dnssrv+’ to detect Alertmanager IPs through respective DNS lookups. The port defaults to 9093 or the SRV record’s value. The URL path is used as a prefix for the regular Alertmanager API path. []]
rule.alertmanagersSendTimeout Timeout for sending alerts to alertmanagert ””
rule.alertQueryUrl The external Thanos Query URL that would be set in all alerts ‘Source’ field ””
rule.alertLabelDrop Labels by name to drop before sending to alertmanager. This allows alert to be deduplicated on replica label (repeated). Similar Prometheus alert relabelling []
rule.ruleOverrideName Override rules file with custom configmap ””
rule.ruleFiles See example in values.yaml {}”
rule.persistentVolumeClaim Create the specified persistentVolumeClaim in case persistentVolumeClaim is used for the dataVolume.backend above and needs to be created. {}

Compact

Name Description Default Value
compact.enabled Enable component true
compact.replicaCount Pod replica count 1
compact.logLevel Log level info
compact.logFormat Log format to use. Possible options: logfmt or json. logfmt
compact.serviceAccount Name of the Kubernetes service account to use ””
compact.consistencyDelay Minimum age of fresh (non-compacted) blocks before they are being processed. Malformed blocks older than the maximum of consistency-delay and 30m0s will be removed. 30m
compact.retentionResolutionRaw How long to retain raw samples in bucket. 0d - disables this retention 30d
compact.retentionResolution5m How long to retain samples of resolution 1 (5 minutes) in bucket. 0d - disables this retention 120d
compact.retentionResolution1h How long to retain samples of resolution 2 (1 hour) in bucket. 0d - disables this retention 1y
compact.blockSyncConcurrency Number of goroutines to use when syncing block metadata from object storage. 20
compact.compactConcurrency Number of goroutines to use when compacting groups. 1
compact.dataVolume.backend Data volume for the compactor to store temporary data defaults to emptyDir. {}
compact.persistentVolumeClaim Create the specified persistentVolumeClaim in case persistentVolumeClaim is used for the dataVolume.backend above and needs to be created. {}

Bucket

Name Description Default Value
bucket.enabled Enable component true
bucket.replicaCount Pod replica count 1
bucket.logLevel Log level info
bucket.logFormat Log format to use. Possible options: logfmt or json. logfmt
bucket.refresh Refresh interval to download metadata from remote storage 30m
bucket.timeout Timeout to download metadata from remote storage 5m
bucket.label Prometheus label to use as timeline title ””
bucket.http.port Listening port for bucket web 8080
bucket.serviceAccount Name of the Kubernetes service account to use ””
bucket.podDisruptionBudget.enabled Enabled and config podDisruptionBudget resource for this component false
bucket.podDisruptionBudget.minAvailable Minimum number of available query pods for PodDisruptionBudget 1
bucket.podDisruptionBudget.maxUnavailable Maximum number of unavailable query pods for PodDisruptionBudget []

Sidecar

Name Description Default Value
sidecar.enabled NOTE: This is only the service references for the sidecar. true
sidecar.selector Pod label selector to match sidecar services on. {"app": "prometheus"}

Query Frontend

Name Description Default Value
queryFrontend.enabled Enable component false
queryFrontend.replicaCount Pod replica count 1
queryFrontend.logLevel Log level info
queryFrontend.logFormat Log format to use. Possible options: logfmt or json. logfmt
queryFrontend.downstreamUrl URL of downstream Prometheus Query compatible API.
queryFrontend.compressResponses Compress HTTP responses. true
queryFrontend.logQueriesLongerThan Log queries that are slower than the specified duration. 0 (disabled)
queryFrontend.cacheCompressionType Use compression in results cache. Supported values are: snappy and “ (disable compression).
queryFrontend.queryRange.alignRangeWithStep See https://thanos.io/tip/components/query-frontend.md/#flags false
queryFrontend.queryRange.splitInterval See https://thanos.io/tip/components/query-frontend.md/#flags 24h
queryFrontend.queryRange.maxRetriesPerRequest See https://thanos.io/tip/components/query-frontend.md/#flags 5
queryFrontend.queryRange.maxQueryLength See https://thanos.io/tip/components/query-frontend.md/#flags 0
queryFrontend.queryRange.maxQueryParallelism See https://thanos.io/tip/components/query-frontend.md/#flags 14
queryFrontend.queryRange.responseCacheMaxFreshness See https://thanos.io/tip/components/query-frontend.md/#flag 1m
queryFrontend.queryRange.noPartialResponse See https://thanos.io/tip/components/query-frontend.md/#flags false
queryFrontend.cache.inMemory Use inMemory cache? false
queryFrontend.cache.maxSize Maximum Size of the cache. Use either this or maxSizeItems.
queryFrontend.cache.maxSizeItems Maximum number of items in the cache. Use either this or maxSize.
queryFrontend.cache.validity
queryFrontend.log.request.decision Request Logging for logging the start and end of requests LogFinishCall

Contributing

Contributions are very welcome!