feature/namespace #2

Merged
jem merged 15 commits from feature/namespace into master 2024-02-20 17:24:37 +00:00
21 changed files with 157 additions and 106 deletions
Showing only changes of commit ab2473bdf4 - Show all commits

View file

@ -1,20 +1,18 @@
(ns dda.c4k-common.monitoring (ns dda.c4k-common.monitoring
(:require (:require
[clojure.spec.alpha :as s] [clojure.spec.alpha :as s]
#?(:cljs [shadow.resource :as rc])
#?(:clj [orchestra.core :refer [defn-spec]] #?(:clj [orchestra.core :refer [defn-spec]]
:cljs [orchestra.core :refer-macros [defn-spec]]) :cljs [orchestra.core :refer-macros [defn-spec]])
[clojure.string :as str]
[dda.c4k-common.yaml :as yaml] [dda.c4k-common.yaml :as yaml]
[dda.c4k-common.predicate :as cp] [dda.c4k-common.predicate :as cp]
[dda.c4k-common.common :as cm] [dda.c4k-common.monitoring.monitoring-internal :as int]))
[clojure.string :as str]))
(s/def ::grafana-cloud-user cp/bash-env-string?) (s/def ::grafana-cloud-user cp/bash-env-string?)
(s/def ::grafana-cloud-password cp/bash-env-string?) (s/def ::grafana-cloud-password cp/bash-env-string?)
(s/def ::grafana-cloud-url string?) (s/def ::grafana-cloud-url string?)
(s/def ::cluster-name string?) (s/def ::cluster-name string?)
(s/def ::cluster-stage cp/stage?) (s/def ::cluster-stage cp/stage?)
(s/def ::pvc-storage-class-name cp/pvc-storage-class-name?)
(s/def ::node-regex string?) (s/def ::node-regex string?)
(s/def ::traefik-regex string?) (s/def ::traefik-regex string?)
(s/def ::kube-state-regex string?) (s/def ::kube-state-regex string?)
@ -23,11 +21,11 @@
::cluster-stage])) ::cluster-stage]))
(s/def ::mon-auth (s/keys :req-un [::grafana-cloud-user (s/def ::mon-auth (s/keys :req-un [::grafana-cloud-user
::grafana-cloud-password])) ::grafana-cloud-password]))
(s/def ::storage (s/keys :opt-un [::pvc-storage-class-name]))
(s/def ::filter-regex (s/keys :req-un [::node-regex (s/def ::filter-regex (s/keys :req-un [::node-regex
::traefik-regex ::traefik-regex
::kube-state-regex])) ::kube-state-regex]))
(def metric-regex {:node-regex (def metric-regex {:node-regex
(str "node_cpu_sec.+|node_load[0-9]+|node_memory_Buf.*|node_memory_Mem.*|" (str "node_cpu_sec.+|node_load[0-9]+|node_memory_Buf.*|node_memory_Mem.*|"
"node_memory_Cached.*|node_disk_[r,w,i].*|node_filesystem_[s,a].*|" "node_memory_Cached.*|node_disk_[r,w,i].*|node_filesystem_[s,a].*|"
@ -46,63 +44,24 @@
(def filter-regex-string (def filter-regex-string
(str/join "|" (vals metric-regex))) (str/join "|" (vals metric-regex)))
#?(:cljs
(defmethod yaml/load-resource :monitoring [resource-name]
(get (inline-resources "monitoring") resource-name)))
(defn-spec generate-stateful-set cp/map-or-seq? (defn-spec generate seq?
[config ::storage]
(let [{:keys [pvc-storage-class-name]
:or {pvc-storage-class-name :manual}} config]
(->
(yaml/load-as-edn "monitoring/stateful-set.yaml")
(assoc-in [:spec :volumeClaimTemplates 0 :spec :storageClassName] (name pvc-storage-class-name)))))
(defn-spec generate-prometheus-config cp/map-or-seq?
[config ::mon-cfg
auth ::mon-auth]
(let [{:keys [grafana-cloud-url cluster-name cluster-stage]} config
{:keys [grafana-cloud-user grafana-cloud-password]} auth]
(->
(yaml/load-as-edn "monitoring/prometheus/prometheus.yaml")
(assoc-in [:global :external_labels :cluster]
cluster-name)
(assoc-in [:global :external_labels :stage]
cluster-stage)
(assoc-in [:remote_write 0 :url]
grafana-cloud-url)
(assoc-in [:remote_write 0 :basic_auth :username]
grafana-cloud-user)
(assoc-in [:remote_write 0 :basic_auth :password]
grafana-cloud-password)
(cm/replace-all-matching-values-by-new-value "FILTER_REGEX" filter-regex-string))))
(defn-spec generate-config cp/map-or-seq?
[config ::mon-cfg
auth ::mon-auth]
(->
(yaml/load-as-edn "monitoring/prometheus/config.yaml")
(assoc-in [:stringData :prometheus.yaml]
(yaml/to-string
(generate-prometheus-config config auth)))))
(defn-spec generate cp/map-or-seq?
[config ::mon-cfg [config ::mon-cfg
auth ::mon-auth] auth ::mon-auth]
[(yaml/load-as-edn "monitoring/namespace.yaml") [(yaml/load-as-edn "monitoring/namespace.yaml")
(yaml/load-as-edn "monitoring/prometheus/cluster-role.yaml") (yaml/load-as-edn "monitoring/prometheus-cluster-role.yaml")
(yaml/load-as-edn "monitoring/prometheus/cluster-role-binding.yaml") (yaml/load-as-edn "monitoring/prometheus-cluster-role-binding.yaml")
(yaml/load-as-edn "monitoring/prometheus/service.yaml") (yaml/load-as-edn "monitoring/prometheus-service.yaml")
(yaml/load-as-edn "monitoring/prometheus/service-account.yaml") (yaml/load-as-edn "monitoring/prometheus-service-account.yaml")
(generate-config config auth) (int/generate-config config auth)
(yaml/load-as-edn "monitoring/prometheus/deployment.yaml") (yaml/load-as-edn "monitoring/prometheus-deployment.yaml")
(yaml/load-as-edn "monitoring/node-exporter/service-account.yaml") (yaml/load-as-edn "monitoring/node-exporter-service-account.yaml")
(yaml/load-as-edn "monitoring/node-exporter/cluster-role.yaml") (yaml/load-as-edn "monitoring/node-exporter-cluster-role.yaml")
(yaml/load-as-edn "monitoring/node-exporter/cluster-role-binding.yaml") (yaml/load-as-edn "monitoring/node-exporter-cluster-role-binding.yaml")
(yaml/load-as-edn "monitoring/node-exporter/daemon-set.yaml") (yaml/load-as-edn "monitoring/node-exporter-daemon-set.yaml")
(yaml/load-as-edn "monitoring/node-exporter/service.yaml") (yaml/load-as-edn "monitoring/node-exporter-service.yaml")
(yaml/load-as-edn "monitoring/kube-state-metrics/cluster-role-binding.yaml") (yaml/load-as-edn "monitoring/kube-state-metrics-cluster-role-binding.yaml")
(yaml/load-as-edn "monitoring/kube-state-metrics/cluster-role.yaml") (yaml/load-as-edn "monitoring/kube-state-metrics-cluster-role.yaml")
(yaml/load-as-edn "monitoring/kube-state-metrics/deployment.yaml") (yaml/load-as-edn "monitoring/kube-state-metrics-deployment.yaml")
(yaml/load-as-edn "monitoring/kube-state-metrics/service-account.yaml") (yaml/load-as-edn "monitoring/kube-state-metrics-service-account.yaml")
(yaml/load-as-edn "monitoring/kube-state-metrics/service.yaml")]) (yaml/load-as-edn "monitoring/kube-state-metrics-service.yaml")])

View file

@ -0,0 +1,77 @@
(ns dda.c4k-common.monitoring.monitoring-internal
(:require
[clojure.spec.alpha :as s]
#?(:clj [orchestra.core :refer [defn-spec]]
:cljs [orchestra.core :refer-macros [defn-spec]])
#?(:cljs [dda.c4k-common.macros :refer-macros [inline-resources]])
[dda.c4k-common.yaml :as yaml]
[dda.c4k-common.predicate :as cp]
[dda.c4k-common.common :as cm]
[clojure.string :as str]))
(s/def ::grafana-cloud-user cp/bash-env-string?)
(s/def ::grafana-cloud-password cp/bash-env-string?)
(s/def ::grafana-cloud-url string?)
(s/def ::cluster-name string?)
(s/def ::cluster-stage cp/stage?)
(s/def ::node-regex string?)
(s/def ::traefik-regex string?)
(s/def ::kube-state-regex string?)
(s/def ::mon-cfg (s/keys :req-un [::grafana-cloud-url
::cluster-name
::cluster-stage]))
(s/def ::mon-auth (s/keys :req-un [::grafana-cloud-user
::grafana-cloud-password]))
(s/def ::filter-regex (s/keys :req-un [::node-regex
::traefik-regex
::kube-state-regex]))
(def metric-regex {:node-regex
(str "node_cpu_sec.+|node_load[0-9]+|node_memory_Buf.*|node_memory_Mem.*|"
"node_memory_Cached.*|node_disk_[r,w,i].*|node_filesystem_[s,a].*|"
"node_network_receive_bytes_total|node_network_transmit_bytes_total")
:traefik-regex (str "traefik_entrypoint_.*_total|"
"traefik_entrypoint_.*_seconds_count|"
"traefik_router_.*_total|"
"traefik_router_.*_seconds_count|"
"traefik_service_.*_total|"
"traefik_service_.*_seconds_count|"
"traefik_tls_certs_not_after")
:kube-state-regex (str "kube_pod_container_status_restarts_total|"
"kube_pod_status_reason|kube_node_status_capacity|kube_node_status_allocatable|"
"kube_cronjob_status_active|kube_job_status_failed")})
(def filter-regex-string
(str/join "|" (vals metric-regex)))
(defn-spec generate-prometheus-config cp/map-or-seq?
[config ::mon-cfg
auth ::mon-auth]
(let [{:keys [grafana-cloud-url cluster-name cluster-stage]} config
{:keys [grafana-cloud-user grafana-cloud-password]} auth]
(->
(yaml/load-as-edn "monitoring/prometheus-prometheus.yaml")
(assoc-in [:global :external_labels :cluster]
cluster-name)
(assoc-in [:global :external_labels :stage]
cluster-stage)
(assoc-in [:remote_write 0 :url]
grafana-cloud-url)
(assoc-in [:remote_write 0 :basic_auth :username]
grafana-cloud-user)
(assoc-in [:remote_write 0 :basic_auth :password]
grafana-cloud-password)
(cm/replace-all-matching-values-by-new-value "FILTER_REGEX" filter-regex-string))))
(defn-spec generate-config cp/map-or-seq?
[config ::mon-cfg
auth ::mon-auth]
(->
(yaml/load-as-edn "monitoring/prometheus-config.yaml")
(assoc-in [:stringData :prometheus.yaml]
(yaml/to-string
(generate-prometheus-config config auth)))))
#?(:cljs
(defmethod yaml/load-resource :monitoring [resource-name]
(get (inline-resources "monitoring") resource-name)))

View file

@ -0,0 +1,59 @@
(ns dda.c4k-common.monitoring.monitoring-internal-test
(:require
#?(:clj [clojure.test :refer [deftest is are testing run-tests]]
:cljs [cljs.test :refer-macros [deftest is are testing run-tests]])
[clojure.string :as str]
[clojure.spec.test.alpha :as st]
[dda.c4k-common.monitoring.monitoring-internal :as cut]))
(st/instrument `cut/generate-stateful-set)
(st/instrument `cut/generate-agent-config)
(st/instrument `cut/generate-config)
(def conf {:cluster-name "clustername"
:cluster-stage "test"
:grafana-cloud-url "https://some.url/with/path"})
(def auth {:grafana-cloud-user "user"
:grafana-cloud-password "password"
:hetzner-cloud-ro-token "ro-token"})
(def invalid-conf {:cluster-name "clustername"
:cluster-stage "test"
:grafana-clud-url "https://some.url/with/path"})
(def invalid-auth {:grafana-cloud-user "user"
:grafana-clod-password "password"
:hetzner-cloud-ro-token "ro-token"})
(deftest should-not-generate-config
(is (thrown?
#?(:clj Exception :cljs js/Error)
(cut/generate-config invalid-conf auth))))
(deftest should-not-generate-auth
(is (thrown?
#?(:clj Exception :cljs js/Error)
(cut/generate-config conf invalid-auth))))
(deftest should-generate-prometheus-remote-write-auth
(is (= {:username "user",
:password "password"}
(get-in
(cut/generate-prometheus-config conf auth)
[:remote_write 0 :basic_auth]))))
(deftest should-generate-prometheus-external-labels
(is (= {:cluster "clustername",
:stage "test"}
(get-in
(cut/generate-prometheus-config conf auth)
[:global :external_labels]))))
(deftest should-generate-config
(is (str/starts-with?
(get-in
(cut/generate-config conf auth)
[:stringData :prometheus.yaml])
"global:\n scrape_interval:")))

View file

@ -2,16 +2,10 @@
(:require (:require
#?(:clj [clojure.test :refer [deftest is are testing run-tests]] #?(:clj [clojure.test :refer [deftest is are testing run-tests]]
:cljs [cljs.test :refer-macros [deftest is are testing run-tests]]) :cljs [cljs.test :refer-macros [deftest is are testing run-tests]])
[clojure.string :as s]
[clojure.spec.test.alpha :as st] [clojure.spec.test.alpha :as st]
[dda.c4k-common.monitoring :as cut] [dda.c4k-common.monitoring :as cut]))
[dda.c4k-common.yaml :as yaml]
[clojure.string :as str]))
(st/instrument `cut/generate) (st/instrument `cut/generate)
(st/instrument `cut/generate-stateful-set)
(st/instrument `cut/generate-agent-config)
(st/instrument `cut/generate-config)
(def conf {:cluster-name "clustername" (def conf {:cluster-name "clustername"
:cluster-stage "test" :cluster-stage "test"
@ -21,45 +15,7 @@
:grafana-cloud-password "password" :grafana-cloud-password "password"
:hetzner-cloud-ro-token "ro-token"}) :hetzner-cloud-ro-token "ro-token"})
(def invalid-conf {:cluster-name "clustername"
:cluster-stage "test"
:grafana-clud-url "https://some.url/with/path"})
(def invalid-auth {:grafana-cloud-user "user"
:grafana-clod-password "password"
:hetzner-cloud-ro-token "ro-token"})
(deftest should-not-generate-config
(is (thrown?
#?(:clj Exception :cljs js/Error)
(cut/generate-config invalid-conf auth))))
(deftest should-not-generate-auth
(is (thrown?
#?(:clj Exception :cljs js/Error)
(cut/generate-config conf invalid-auth))))
(deftest should-generate (deftest should-generate
(is (= 17 (is (= 17
(count (cut/generate conf auth))))) (count (cut/generate conf auth)))))
(deftest should-generate-prometheus-remote-write-auth
(is (= {:username "user",
:password "password"}
(get-in
(cut/generate-prometheus-config conf auth)
[:remote_write 0 :basic_auth]))))
(deftest should-generate-prometheus-external-labels
(is (= {:cluster "clustername",
:stage "test"}
(get-in
(cut/generate-prometheus-config conf auth)
[:global :external_labels]))))
(deftest should-generate-config
(is (s/starts-with?
(get-in
(cut/generate-config conf auth)
[:stringData :prometheus.yaml])
"global:\n scrape_interval:")))