Clarify help text on scaling dashboard (#1666)

* Clarify help text on scaling dashboard * Remove unnecessary line * Fix linter * Apply suggestions from code review Co-authored-by: Ursula Kallio <[email protected]> * Rebuilt compiled mixin Signed-off-by: Marco Pracucci <[email protected]> Co-authored-by: Ursula Kallio <[email protected]> Co-authored-by: Marco Pracucci <[email protected]>
grafana · Apr 14, 2022 · 562ae26 · 562ae26
1 parent a3c1060
commit 562ae26
Show file tree

Hide file tree

Showing 3 changed files with 7 additions and 6 deletions.
diff --git a/docs/sources/operators-guide/visualizing-metrics/dashboards/scaling/index.md b/docs/sources/operators-guide/visualizing-metrics/dashboards/scaling/index.md
@@ -5,9 +5,9 @@ description: "View an example Scaling dashboard."
 weight: 140
 ---
 
-# Grafana Mimir Scaling dashboard
+# Grafana Mimir scaling dashboard
 
-The Scaling dashboard shows services that are not scaled correctly.
+The scaling dashboard displays services that you can optionally scale up, in the event of a failure that is caused by one or more specific reasons.
 
 ## Example
 

diff --git a/operations/mimir-mixin-compiled/dashboards/mimir-scaling.json b/operations/mimir-mixin-compiled/dashboards/mimir-scaling.json
@@ -29,7 +29,7 @@
                {
                   "id": 1,
                   "options": {
-                     "content": "This dashboards shows any services which are not scaled correctly.\nThe table below gives the required number of replicas and the reason why.\nWe only show services without enough replicas.\n\nReasons:\n- **sample_rate**: There are not enough replicas to handle the\n  sample rate.  Applies to distributor and ingesters.\n- **active_series**: There are not enough replicas\n  to handle the number of active series.  Applies to ingesters.\n- **cpu_usage**: There are not enough replicas\n  based on the CPU usage of the jobs vs the resource requests.\n  Applies to all jobs.\n- **memory_usage**: There are not enough replicas based on the memory\n  usage vs the resource requests.  Applies to all jobs.\n- **active_series_limits**: There are not enough replicas to hold 60% of the\n  sum of all the per tenant series limits.\n- **sample_rate_limits**: There are not enough replicas to handle 60% of the\n  sum of all the per tenant rate limits.\n",
+                     "content": "This dashboard identifies scaling-related issues by suggesting services that you might want to scale up.\nThe table that follows contains a suggested number of replicas and the reason why.\nIf the system is failing and depending on the reason, try scaling up to the specified number.\nThe specified numbers are intended as helpful guidelines when things go wrong, rather than prescriptive guidelines.\n\nReasons:\n- **sample_rate**: There are not enough replicas to handle the\n  sample rate.  Applies to distributor and ingesters.\n- **active_series**: There are not enough replicas\n  to handle the number of active series.  Applies to ingesters.\n- **cpu_usage**: There are not enough replicas\n  based on the CPU usage of the jobs vs the resource requests.\n  Applies to all jobs.\n- **memory_usage**: There are not enough replicas based on the memory\n  usage vs the resource requests.  Applies to all jobs.\n- **active_series_limits**: There are not enough replicas to hold 60% of the\n  sum of all the per tenant series limits.\n- **sample_rate_limits**: There are not enough replicas to handle 60% of the\n  sum of all the per tenant rate limits.\n",
                      "mode": "markdown"
                   },
                   "span": 12,

diff --git a/operations/mimir-mixin/dashboards/scaling.libsonnet b/operations/mimir-mixin/dashboards/scaling.libsonnet
@@ -11,9 +11,10 @@ local utils = import 'mixin-utils/utils.libsonnet';
         title: '',
         options: {
           content: |||
-            This dashboards shows any services which are not scaled correctly.
-            The table below gives the required number of replicas and the reason why.
-            We only show services without enough replicas.
+            This dashboard identifies scaling-related issues by suggesting services that you might want to scale up.
+            The table that follows contains a suggested number of replicas and the reason why.
+            If the system is failing and depending on the reason, try scaling up to the specified number.
+            The specified numbers are intended as helpful guidelines when things go wrong, rather than prescriptive guidelines.
 
             Reasons:
             - **sample_rate**: There are not enough replicas to handle the