changeset 19:10017def57ce

update scrapes config
author drewp@bigasterisk.com
date Sat, 24 Jun 2023 01:43:55 -0700
parents 92dce2fa8954
children f5777b65f035
files config/scrape_main.yaml
diffstat 1 files changed, 24 insertions(+), 42 deletions(-) [+]
line wrap: on
line diff
--- a/config/scrape_main.yaml	Sat Jun 24 01:43:21 2023 -0700
+++ b/config/scrape_main.yaml	Sat Jun 24 01:43:55 2023 -0700
@@ -11,12 +11,6 @@
 scrape_configs:
   # some based on https://github.com/prometheus/prometheus/blob/main/documentation/examples/prometheus-kubernetes.yml
 
-  # - job_name: "vmalert"
-  #   metrics_path: /vmalert/metrics
-  #   static_configs:
-  #     - targets:
-  #         - vmalert.default.svc.cluster.local
-
   - job_name: "kubernetes-apiservers"
     scheme: https
     tls_config: { ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt }
@@ -63,9 +57,9 @@
         regex: "false"
         action: drop
 
-      - source_labels: [__meta_kubernetes_namespace]
-        regex: default
-        action: keep
+      # - source_labels: [__meta_kubernetes_namespace]
+      #   regex: default
+      #   action: keep
 
       # promote these to display
       - source_labels: [__meta_kubernetes_service_name]
@@ -92,32 +86,31 @@
         replacement: "/m/metrics"
 
       # discovery is matching extra ports that don't serve metrics- remove these targets
+      - {if: '{job="cert-manager-webhook"}',                    action: drop}
+      - {if: '{job="cert-manager",      __port_number="9403"}', action: drop}
+      - {if: '{job="filesync-syncthing",__port_number="8384"}', action: drop}
+      - {if: '{job="jsregistry",        __port_number="4873"}', action: drop}
+      - {if: '{job="kube-dns",          __port_number="53"}',   action: drop}
       - {if: '{job="kubernetes"}', action: drop}
       - {if: '{job="mongodb",           __port_number="27017"}', action: drop}
       - {if: '{job="mosquitto-ext",     __port_number="1883"}', action: drop}
-      - {if: '{job="filesync-syncthing",__port_number="8384"}', action: drop}
-      - {if: '{job="jsregistry",        __port_number="4873"}', action: drop}
+      - {if: '{job="net-route-input",   __port_number="80"}',   action: drop}
       - {if: '{job="photoprism",        __port_number="2342"}', action: drop}
-      - {if: '{job="net-route-input",   __port_number="80"}', action: drop}
+      - {if: '{job="pomerium-metrics",  __port_number="8080"}', action: drop}
+      - {if: '{job="pomerium-metrics",  __port_number="8443"}', action: drop}
+      - {if: '{job="pomerium-proxy",    __port_number="8080"}', action: drop}
+      - {if: '{job="pomerium-proxy",    __port_number="8443"}', action: drop}
+      
+      # Needs https. Used by `kubectl top`
+      - {if: '{job="metrics-server",    __port_number="4443"}', action: drop}
 
       # discovery is also matching redundant container ports that it also catches with the service
-      - {if: '{job="lanscape",        __port_number="8001"}', action: drop}
-      - {if: '{job="lanscape",        __port_number="8002"}', action: drop}
-      - {if: '{job="collector",        __port_number="8001"}', action: drop}
-      - {if: '{job="collector",        __port_number="8002"}', action: drop}
-      - {if: '{job="racc-console",        __port_number="8002"}', action: drop}
-      - {if: '{job="antigen-web",        __port_number="8001"}', action: drop}
-
-  
-  # # seems like this would match more stuff, but all I get is coredns
-  # - job_name: 'old_coredns'
-  #   kubernetes_sd_configs: [{role: pod}]
-  #   relabel_configs:
-  #     - source_labels: [__meta_kubernetes_pod_container_port_name]
-  #       regex: metrics
-  #       action: keep
-  #     - source_labels: [__meta_kubernetes_pod_container_name]
-  #       target_label: job
+      - {if: '{job="antigen-web",       __port_number="8001"}', action: drop}
+      - {if: '{job="collector",         __port_number="8001"}', action: drop}
+      - {if: '{job="collector",         __port_number="8002"}', action: drop}
+      - {if: '{job="lanscape",          __port_number="8001"}', action: drop}
+      - {if: '{job="lanscape",          __port_number="8002"}', action: drop}
+      - {if: '{job="racc-console",      __port_number="8002"}', action: drop}
 
   - job_name: "telegraf"
     scheme: http
@@ -155,18 +148,11 @@
       - targets:
           - pipe:9991
 
-  - job_name: "pomerium"
-    static_configs:
-      - targets:
-          - pomerium-metrics.pomerium.svc:9090
-          - cert-manager.cert-manager.svc:9402
-
-
   - job_name: "zfs"
     scrape_interval: 1h
     static_configs:
       - targets:
-          # running in in k8s, but not in SD
+          # running in in k8s, but as daemonset so it's not in SD above
           - ditto:9634
           - ditto:9986
 
@@ -179,16 +165,12 @@
       - targets:
           # printer, since it falls out of ntop with no traffic at all. Or, we could poll ink status at http://10.2.0.37/general/status.html?pageid=1
           - printer014032ED
-          # frontbed, for monitoring
-#          - 10.5.0.17 # too flaky
-          # asher bulb, not sure why it sleeps so long
-          - bulb1
-
     relabel_configs:
       - {source_labels: [__address__],    target_label: __param_target}
       - {source_labels: [__param_target], target_label: instance}
       - target_label: __address__
         replacement: prober
+
   - job_name: "racc"
     scrape_interval: 30s
     static_configs: