From 0c8c88e766fc225cf7dff3bf0288c4443e459f49 Mon Sep 17 00:00:00 2001
From: Orfeas Kourkakis <orfeas.kourkakis@canonical.com>
Date: Tue, 9 Apr 2024 12:33:31 +0300
Subject: [PATCH] backport(fix): Fix dashboard panels not working from #157
 (#160)

* Add `ckf` tag to argo-controller's grafana dashboard.
* Fix dashboard panels not working by:
  * Replacing unavailable metrics with available ones
  * Change 2 minutes instead of 1 in places where rate() is used since
    this requires more than one scrape data points.
  * Remove rate() from panel that shows total number of log messages.

Ref canonical/bundle-kubeflow#856
Ref canonical/bundle-kubeflow#834
---
 .../src/grafana_dashboards/basic.json.tmpl            | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/charms/argo-controller/src/grafana_dashboards/basic.json.tmpl b/charms/argo-controller/src/grafana_dashboards/basic.json.tmpl
index bcf5671..9341693 100644
--- a/charms/argo-controller/src/grafana_dashboards/basic.json.tmpl
+++ b/charms/argo-controller/src/grafana_dashboards/basic.json.tmpl
@@ -980,7 +980,7 @@
         {
           "datasource": "${prometheusds}",
           "exemplar": true,
-          "expr": "  rate(argo_workflows_queue_latency_sum{queue_name=\"cron_wf_queue\",kubernetes_namespace=~\"^$ns$\"}[1m])\n/\n  rate(argo_workflows_queue_latency_count{queue_name=\"cron_wf_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[1m])",
+          "expr": "  rate(argo_workflows_queue_latency_sum{queue_name=\"cron_wf_queue\",kubernetes_namespace=~\"^$ns$\"}[2m])\n/\n  rate(argo_workflows_queue_latency_count{queue_name=\"cron_wf_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[2m])",
           "interval": "1m",
           "legendFormat": "{{origin_prometheus}} : {{app}} : {{kubernetes_namespace}} : {{queue_name}}",
           "queryType": "randomWalk",
@@ -989,7 +989,7 @@
         {
           "datasource": "${prometheusds}",
           "exemplar": true,
-          "expr": "  rate(argo_workflows_queue_latency_sum{queue_name=\"pod_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[1m])\n/\n  rate(argo_workflows_queue_latency_count{queue_name=\"pod_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[1m])",
+          "expr": "  rate(argo_workflows_queue_latency_sum{queue_name=\"pod_cleanup_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[2m])\n/\n  rate(argo_workflows_queue_latency_count{queue_name=\"pod_cleanup_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[2m])",
           "interval": "1m",
           "legendFormat": "{{origin_prometheus}} : {{app}} : {{kubernetes_namespace}} : {{queue_name}}",
           "refId": "B"
@@ -997,7 +997,7 @@
         {
           "datasource": "${prometheusds}",
           "exemplar": true,
-          "expr": "  rate(argo_workflows_queue_latency_sum{queue_name=\"wf_cron_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[1m])\n/\n  rate(argo_workflows_queue_latency_count{queue_name=\"wf_cron_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[1m])",
+          "expr": "  rate(argo_workflows_queue_latency_sum{queue_name=\"cron_wf_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[2m])\n/\n  rate(argo_workflows_queue_latency_count{queue_name=\"cron_wf_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[2m])",
           "hide": false,
           "interval": "1m",
           "legendFormat": "{{origin_prometheus}} : {{app}} : {{kubernetes_namespace}} : {{queue_name}}",
@@ -1006,7 +1006,7 @@
         {
           "datasource": "${prometheusds}",
           "exemplar": true,
-          "expr": "  rate(argo_workflows_queue_latency_sum{queue_name=\"workflow_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[1m])\n/\n  rate(argo_workflows_queue_latency_count{queue_name=\"workflow_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[1m])",
+          "expr": "  rate(argo_workflows_queue_latency_sum{queue_name=\"workflow_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[2m])\n/\n  rate(argo_workflows_queue_latency_count{queue_name=\"workflow_queue\",kubernetes_namespace=~\"^$ns$\",origin_prometheus=~\"^$dc$\"}[2m])",
           "hide": false,
           "interval": "1m",
           "legendFormat": "{{origin_prometheus}} : {{app}} : {{kubernetes_namespace}} : {{queue_name}}",
@@ -1112,7 +1112,7 @@
         {
           "datasource": "${prometheusds}",
           "exemplar": true,
-          "expr": "rate(log_messages{kubernetes_namespace=~\"$ns\",origin_prometheus=~\"^$dc$\"}[1m])",
+          "expr": "log_messages{kubernetes_namespace=~\"$ns\",origin_prometheus=~\"^$dc$\"}",
           "interval": "1m",
           "legendFormat": "{{origin_prometheus}} : {{app}} : {{kubernetes_namespace}} : {{level}}",
           "queryType": "randomWalk",
@@ -1159,6 +1159,7 @@
   "schemaVersion": 33,
   "style": "dark",
   "tags": [
+    "ckf",
     "argo"
   ],
   "templating": {