ildkule: Move metrics->monitoring. Add uptime-kuma. Adjust prometheus exporters.

hosts/ildkule/services/monitoring/default.nix

@@ -0,0 +1,10 @@
+{ config, pkgs, ... }:
+
+{
+  imports = [
+    ./grafana.nix
+    ./loki.nix
+    ./prometheus
+    ./uptime-kuma.nix
+  ];
+}

hosts/ildkule/services/monitoring/grafana.nix

@@ -0,0 +1,98 @@
+{ config, pkgs, values, ... }: let
+  cfg = config.services.grafana;
+in {
+  sops.secrets = let
+    owner = "grafana";
+    group = "grafana";
+  in {
+    "keys/grafana/secret_key" = { inherit owner group; };
+    "keys/grafana/admin_password" = { inherit owner group; };
+  };
+
+  services.grafana = {
+    enable = true;
+
+    settings = let
+      # See https://grafana.com/docs/grafana/latest/setup-grafana/configure-grafana/#file-provider
+      secretFile = path: "$__file{${path}}";
+    in {
+      server = {
+        domain = "grafana.pvv.ntnu.no";
+        http_port = 2342;
+        http_addr = "127.0.0.1";
+      };
+
+      security = {
+        secret_key = secretFile config.sops.secrets."keys/grafana/secret_key".path;
+        admin_password = secretFile config.sops.secrets."keys/grafana/admin_password".path;
+      };
+    };
+
+    provision = {
+      enable = true;
+      datasources.settings.datasources = [
+        {
+          name = "Ildkule Prometheus";
+          type = "prometheus";
+          url = ("http://${config.services.prometheus.listenAddress}:${toString config.services.prometheus.port}");
+         isDefault = true;
+        }
+        {
+          name = "Ildkule loki";
+          type = "loki";
+          url = ("http://${config.services.loki.configuration.server.http_listen_address}:${toString config.services.loki.configuration.server.http_listen_port}");
+        }
+      ];
+      dashboards.settings.providers = [
+        {
+          name = "Node Exporter Full";
+          type = "file";
+          url = "https://grafana.com/api/dashboards/1860/revisions/29/download";
+          options.path = dashboards/node-exporter-full.json;
+        }
+        {
+          name = "Matrix Synapse";
+          type = "file";
+          url = "https://raw.githubusercontent.com/matrix-org/synapse/develop/contrib/grafana/synapse.json";
+          options.path = dashboards/synapse.json;
+        }
+	# TODO: enable once https://github.com/NixOS/nixpkgs/pull/242365 gets merged
+	# {
+	#   name = "MySQL";
+	#   type = "file";
+	#   url = "https://raw.githubusercontent.com/prometheus/mysqld_exporter/main/mysqld-mixin/dashboards/mysql-overview.json";
+	#   options.path = dashboards/mysql.json;
+	# }
+        {
+          name = "Postgresql";
+          type = "file";
+          url = "https://grafana.com/api/dashboards/9628/revisions/7/download";
+          options.path = dashboards/postgres.json;
+        }
+        {
+          name = "Go Processes (gogs)";
+          type = "file";
+          url = "https://grafana.com/api/dashboards/240/revisions/3/download";
+          options.path = dashboards/go-processes.json;
+        }
+      ];
+
+    };
+  };
+
+  services.nginx.virtualHosts.${cfg.settings.server.domain} = {
+    enableACME = true;
+    forceSSL = true;
+    kTLS = true;
+    locations = {
+      "/" = {
+        proxyPass = "http://127.0.0.1:${toString cfg.settings.server.http_port}";
+        proxyWebsockets = true;
+        extraConfig = ''
+          proxy_buffers 8 1024k;
+          proxy_buffer_size 1024k;
+        '';
+      };
+    };
+  };
+}

hosts/ildkule/services/monitoring/loki.nix

@@ -0,0 +1,86 @@
+{ config, pkgs, ... }:
+
+let
+  cfg = config.services.loki;
+in {
+  services.loki = {
+    enable = true;
+    configuration = {
+      auth_enabled = false;
+      server = {
+        http_listen_port = 3100;
+        http_listen_address = "0.0.0.0";
+        grpc_listen_port = 9096;
+      };
+
+      ingester = {
+        wal = {
+          enabled = true;
+          dir = "/var/lib/loki/wal";
+        };
+        lifecycler = {
+          address = "127.0.0.1";
+          ring = {
+            kvstore = {
+              store = "inmemory";
+            };
+            replication_factor = 1;
+          };
+          final_sleep = "0s";
+        };
+        chunk_idle_period = "1h";
+      };
+
+      schema_config = {
+        configs = [
+          {
+            from = "2022-12-01";
+            store = "boltdb-shipper";
+            object_store = "filesystem";
+            schema = "v11";
+            index = {
+              prefix = "index_";
+              period = "24h";
+            };
+          }
+        ];
+      };
+
+      storage_config = {
+        boltdb_shipper = {
+          active_index_directory = "/var/lib/loki/boltdb-shipper-index";
+          cache_location = "/var/lib/loki/boltdb-shipper-cache";
+          shared_store = "filesystem";
+          cache_ttl = "24h";
+        };
+        filesystem = {
+          directory = "/var/lib/loki/chunks";
+        };
+      };
+
+      limits_config = {
+        enforce_metric_name = false;
+        reject_old_samples = true;
+        reject_old_samples_max_age = "72h";
+      };
+
+      compactor = {
+        working_directory = "/var/lib/loki/compactor";
+        shared_store = "filesystem";
+      };
+
+      # ruler = {
+      #   storage = {
+      #     type = "local";
+      #     local = {
+      #       directory = "/var/lib/loki/rules";
+      #     };
+      #   };
+      #   rule_path = "/etc/loki/rules";
+      #   alertmanager_url = "http://localhost:9093";
+      # };
+    };
+  };
+
+  networking.firewall.allowedTCPPorts = [ cfg.configuration.server.http_listen_port ];
+}

hosts/ildkule/services/monitoring/prometheus/default.nix

@@ -0,0 +1,18 @@
+{ config, ... }: {
+  imports = [
+    ./gogs.nix
+    ./matrix-synapse.nix
+    # TODO: enable once https://github.com/NixOS/nixpkgs/pull/242365 gets merged
+    # ./mysqld.nix
+    ./node.nix
+    ./postgres.nix
+  ];
+
+  services.prometheus = {
+    enable = true;
+    listenAddress = "127.0.0.1";
+    port = 9001;
+
+    ruleFiles = [ rules/synapse-v2.rules ];
+  };
+}

hosts/ildkule/services/monitoring/prometheus/gogs.nix

@@ -0,0 +1,16 @@
+{ config, ... }: let
+  cfg = config.services.prometheus;
+in {
+  services.prometheus.scrapeConfigs = [{
+    job_name = "git-gogs";
+    scheme = "https";
+    metrics_path = "/-/metrics";
+    static_configs = [
+      {
+        targets = [
+          "essendrop.pvv.ntnu.no:443"
+        ];
+      }
+    ];
+  }];
+}

hosts/ildkule/services/monitoring/prometheus/matrix-synapse.nix

@@ -0,0 +1,40 @@
+{ ... }:
+{
+  services.prometheus.scrapeConfigs = [{
+    job_name = "synapse";
+    scrape_interval = "15s";
+    scheme = "https";
+
+    http_sd_configs = [{
+      url = "https://matrix.pvv.ntnu.no/metrics/config.json";
+    }];
+
+    relabel_configs = [
+      {
+        source_labels = [ "__address__" ];
+        regex = "[^/]+(/.*)";
+        target_label = "__metrics_path__";
+      }
+      {
+        source_labels = [ "__address__" ];
+        regex = "([^/]+)/.*";
+        target_label = "instance";
+      }
+      {
+        source_labels = [ "__address__" ];
+        regex = "[^/]+\\/+[^/]+/(.*)/\\d+$";
+        target_label = "job";
+      }
+      {
+        source_labels = [ "__address__" ];
+        regex = "[^/]+\\/+[^/]+/.*/(\\d+)$";
+        target_label = "index";
+      }
+      {
+        source_labels = [ "__address__" ];
+        regex = "([^/]+)/.*";
+        target_label = "__address__";
+      }
+    ];
+  }];
+}

hosts/ildkule/services/monitoring/prometheus/mysqld.nix

@@ -0,0 +1,25 @@
+{ config, ... }: let
+  cfg = config.services.prometheus;
+in {
+  sops.secrets."config/mysqld_exporter" = { };
+
+  services.prometheus = {
+    scrapeConfigs = [{
+      job_name = "mysql";
+      scheme = "http";
+      metrics_path = cfg.exporters.mysqld.telemetryPath;
+      static_configs = [
+        {
+          targets = [
+            "localhost:${toString cfg.exporters.mysqld.port}"
+          ];
+        }
+      ];
+    }];
+
+    exporters.mysqld = {
+      enable = true;
+      configFilePath = config.sops.secrets."config/mysqld_exporter".path;
+    };
+  };
+}

hosts/ildkule/services/monitoring/prometheus/node.nix

@@ -0,0 +1,22 @@
+{ config, ... }: let
+  cfg = config.services.prometheus;
+in {
+  services.prometheus.scrapeConfigs = [{
+    job_name = "node";
+    static_configs = [
+      {
+        targets = [
+          "ildkule.pvv.ntnu.no:${toString cfg.exporters.node.port}"
+          "microbel.pvv.ntnu.no:9100"
+          "isvegg.pvv.ntnu.no:9100"
+          "knakelibrak.pvv.ntnu.no:9100"
+          "hildring.pvv.ntnu.no:9100"
+          "bicep.pvv.ntnu.no:9100"
+          "essendrop.pvv.ntnu.no:9100"
+          "andresbu.pvv.ntnu.no:9100"
+          "bekkalokk.pvv.ntnu.no:9100"
+        ];
+      }
+    ];
+  }];
+}

hosts/ildkule/services/monitoring/prometheus/postgres.nix

@@ -0,0 +1,51 @@
+{ pkgs, lib, config, values, ... }: let
+  cfg = config.services.prometheus;
+in {
+  sops.secrets = {
+    "keys/postgres/postgres_exporter_env" = {};
+    "keys/postgres/postgres_exporter_knakelibrak_env" = {};
+  };
+
+  services.prometheus = {
+    scrapeConfigs = [
+      {
+        job_name = "postgres";
+        scrape_interval = "15s";
+        static_configs = [{
+          targets = [ "localhost:${toString cfg.exporters.postgres.port}" ];
+          labels = {
+            server = "bicep";
+          };
+        }];
+      }
+      {
+        job_name = "postgres-knakelibrak";
+        scrape_interval = "15s";
+        static_configs = [{
+          targets = [ "localhost:${toString (cfg.exporters.postgres.port + 1)}" ];
+          labels = {
+            server = "knakelibrak";
+          };
+        }];
+      }
+    ];
+
+    exporters.postgres = {
+      enable = true;
+      extraFlags = [ "--auto-discover-databases" ];
+      environmentFile = config.sops.secrets."keys/postgres/postgres_exporter_env".path;
+    };
+  };
+
+  systemd.services.prometheus-postgres-exporter-knakelibrak.serviceConfig = let
+    localCfg = config.services.prometheus.exporters.postgres; 
+  in lib.recursiveUpdate config.systemd.services.prometheus-postgres-exporter.serviceConfig {
+      EnvironmentFile = config.sops.secrets."keys/postgres/postgres_exporter_knakelibrak_env".path;
+      ExecStart = ''
+        ${pkgs.prometheus-postgres-exporter}/bin/postgres_exporter \
+          --web.listen-address ${localCfg.listenAddress}:${toString (localCfg.port + 1)} \
+          --web.telemetry-path ${localCfg.telemetryPath} \
+          ${lib.concatStringsSep " \\\n  " localCfg.extraFlags}
+      '';
+    };
+}

hosts/ildkule/services/monitoring/prometheus/rules/synapse-v2.rules

@@ -0,0 +1,74 @@
+groups:
+- name: synapse
+  rules:
+
+  ###
+  ### Prometheus Console Only
+  ### The following rules are only needed if you use the Prometheus Console
+  ### in contrib/prometheus/consoles/synapse.html
+  ###
+  - record: 'synapse_federation_client_sent'
+    labels:
+      type: "EDU"
+    expr: 'synapse_federation_client_sent_edus_total + 0'
+  - record: 'synapse_federation_client_sent'
+    labels:
+      type: "PDU"
+    expr: 'synapse_federation_client_sent_pdu_destinations_count_total + 0'
+  - record: 'synapse_federation_client_sent'
+    labels:
+      type: "Query"
+    expr: 'sum(synapse_federation_client_sent_queries) by (job)'
+
+  - record: 'synapse_federation_server_received'
+    labels:
+      type: "EDU"
+    expr: 'synapse_federation_server_received_edus_total + 0'
+  - record: 'synapse_federation_server_received'
+    labels:
+      type: "PDU"
+    expr: 'synapse_federation_server_received_pdus_total + 0'
+  - record: 'synapse_federation_server_received'
+    labels:
+      type: "Query"
+    expr: 'sum(synapse_federation_server_received_queries) by (job)'
+
+  - record: 'synapse_federation_transaction_queue_pending'
+    labels:
+      type: "EDU"
+    expr: 'synapse_federation_transaction_queue_pending_edus + 0'
+  - record: 'synapse_federation_transaction_queue_pending'
+    labels:
+      type: "PDU"
+    expr: 'synapse_federation_transaction_queue_pending_pdus + 0'
+  ###
+  ### End of 'Prometheus Console Only' rules block
+  ###
+
+
+  ###
+  ### Grafana Only
+  ### The following rules are only needed if you use the Grafana dashboard
+  ### in contrib/grafana/synapse.json
+  ###
+  - record: synapse_storage_events_persisted_by_source_type
+    expr: sum without(type, origin_type, origin_entity) (synapse_storage_events_persisted_events_sep_total{origin_type="remote"})
+    labels:
+      type: remote
+  - record: synapse_storage_events_persisted_by_source_type
+    expr: sum without(type, origin_type, origin_entity) (synapse_storage_events_persisted_events_sep_total{origin_entity="*client*",origin_type="local"})
+    labels:
+      type: local
+  - record: synapse_storage_events_persisted_by_source_type
+    expr: sum without(type, origin_type, origin_entity) (synapse_storage_events_persisted_events_sep_total{origin_entity!="*client*",origin_type="local"})
+    labels:
+      type: bridges
+
+  - record: synapse_storage_events_persisted_by_event_type
+    expr: sum without(origin_entity, origin_type) (synapse_storage_events_persisted_events_sep_total)
+
+  - record: synapse_storage_events_persisted_by_origin
+    expr: sum without(type) (synapse_storage_events_persisted_events_sep_total)
+  ###
+  ### End of 'Grafana Only' rules block
+  ###

hosts/ildkule/services/monitoring/uptime-kuma.nix

@@ -0,0 +1,20 @@
+{ config, pkgs, lib, ... }:
+let
+  cfg = config.services.uptime-kuma;
+  domain = "uptime.pvv.ntnu.no";
+in {
+  services.uptime-kuma = {
+    enable = true;
+    settings = {
+      PORT = "5059";
+      HOST = "127.0.1.2";
+    };
+  };
+
+  services.nginx.virtualHosts.${domain} = {
+    enableACME = true;
+    forceSSL = true;
+    kTLS = true;
+    locations."/".proxyPass = "http://${cfg.settings.HOST}:${cfg.settings.PORT}";
+  };
+}