feat(observability): Add record attribute to alertgroups

docs/data-sources/observability_alertgroup.md

@@ -45,3 +45,4 @@ Read-Only:
 - `expression` (String) The PromQL expression to evaluate. Every evaluation cycle this is evaluated at the current time, and all resultant time series become pending/firing alerts.
 - `for` (String) Alerts are considered firing once they have been returned for this long. Alerts which have not yet fired for long enough are considered pending. Default is 0s
 - `labels` (Map of String) A map of key:value. Labels to add or overwrite for each alert
+- `record` (String) The name of the metric. It's the identifier and must be unique in the group.

docs/resources/observability_alertgroup.md

@@ -32,16 +32,11 @@ resource "stackit_observability_alertgroup" "example" {
       }
     },
     {
-      alert      = "example-alert-name-2"
       expression = "kube_node_status_condition{condition=\"Ready\", status=\"false\"} > 0"
-      for        = "1m"
       labels = {
         severity = "critical"
       },
-      annotations = {
-        summary : "example summary"
-        description : "example description"
-      }
+      record = "example_record_name"
     },
   ]
 }
@@ -76,11 +71,12 @@ import {
 
 Required:
 
-- `alert` (String) The name of the alert rule. Is the identifier and must be unique in the group.
 - `expression` (String) The PromQL expression to evaluate. Every evaluation cycle this is evaluated at the current time, and all resultant time series become pending/firing alerts.
 
 Optional:
 
+- `alert` (String) The name of the alert rule. Is the identifier and must be unique in the group.
 - `annotations` (Map of String) A map of key:value. Annotations to add or overwrite for each alert
 - `for` (String) Alerts are considered firing once they have been returned for this long. Alerts which have not yet fired for long enough are considered pending. Default is 0s
 - `labels` (Map of String) A map of key:value. Labels to add or overwrite for each alert
+- `record` (String) The name of the metric. It's the identifier and must be unique in the group.

examples/resources/stackit_observability_alertgroup/resource.tf

@@ -17,16 +17,11 @@ resource "stackit_observability_alertgroup" "example" {
       }
     },
     {
-      alert      = "example-alert-name-2"
       expression = "kube_node_status_condition{condition=\"Ready\", status=\"false\"} > 0"
-      for        = "1m"
       labels = {
         severity = "critical"
       },
-      annotations = {
-        summary : "example summary"
-        description : "example description"
-      }
+      record = "example_record_name"
     },
   ]
 }

stackit/internal/services/observability/alertgroup/datasource.go

@@ -123,6 +123,10 @@ func (a *alertGroupDataSource) Schema(_ context.Context, _ datasource.SchemaRequ
 							ElementType: types.StringType,
 							Computed:    true,
 						},
+						"record": schema.StringAttribute{
+							Description: descriptions["record"],
+							Computed:    true,
+						},
 					},
 				},
 			},

stackit/internal/services/observability/alertgroup/resource.go

@@ -53,6 +53,7 @@ type rule struct {
 	Labels      types.Map    `tfsdk:"labels"`
 	Expression  types.String `tfsdk:"expression"`
 	For         types.String `tfsdk:"for"`
+	Record      types.String `tfsdk:"record"`
 }
 
 var ruleTypes = map[string]attr.Type{
@@ -61,6 +62,7 @@ var ruleTypes = map[string]attr.Type{
 	"labels":      basetypes.MapType{ElemType: types.StringType},
 	"expression":  basetypes.StringType{},
 	"for":         basetypes.StringType{},
+	"record":      basetypes.StringType{},
 }
 
 // Descriptions for the resource and data source schemas are centralized here.
@@ -75,6 +77,7 @@ var descriptions = map[string]string{
 	"for":         "Alerts are considered firing once they have been returned for this long. Alerts which have not yet fired for long enough are considered pending. Default is 0s",
 	"labels":      "A map of key:value. Labels to add or overwrite for each alert",
 	"annotations": "A map of key:value. Annotations to add or overwrite for each alert",
+	"record":      "The name of the metric. It's the identifier and must be unique in the group.",
 }
 
 // NewAlertGroupResource is a helper function to simplify the provider implementation.
@@ -107,6 +110,40 @@ func (a *alertGroupResource) Configure(ctx context.Context, req resource.Configu
 	tflog.Info(ctx, "Observability alert group client configured")
 }
 
+func (a *alertGroupResource) ValidateConfig(ctx context.Context, req resource.ValidateConfigRequest, resp *resource.ValidateConfigResponse) {
+	var resourceModel Model
+	resp.Diagnostics.Append(req.Config.Get(ctx, &resourceModel)...)
+	if resp.Diagnostics.HasError() {
+		return
+	}
+
+	rules := &[]rule{}
+	diags := resourceModel.Rules.ElementsAs(ctx, rules, false)
+	resp.Diagnostics.Append(diags...)
+	if resp.Diagnostics.HasError() {
+		return
+	}
+
+	// check every rule if the requirements are met, if one fails the whole config fails
+	rs := *rules
+	for i := range rs {
+		// either `alert` or `record` is needed
+		if rs[i].Alert.IsNull() && rs[i].Record.IsNull() {
+			core.LogAndAddError(ctx, &resp.Diagnostics, "Error configuring alertgroup", "You need to provide either `alert` or `record` for a `rule`.")
+		}
+
+		// both are set, only one is allowed
+		if !rs[i].Alert.IsNull() && !rs[i].Record.IsNull() {
+			core.LogAndAddError(ctx, &resp.Diagnostics, "Error configuring alertgroup", "Both `alert` and `record` were set for a`rule`. Only one is allowed.")
+		}
+
+		// if record is set, `annotations` and `for` are not allowed
+		if (!rs[i].Record.IsNull() && !rs[i].Record.IsUnknown()) && ((!rs[i].Annotations.IsNull() && !rs[i].Annotations.IsUnknown()) || (!rs[i].For.IsNull() && !rs[i].For.IsUnknown())) {
+			core.LogAndAddError(ctx, &resp.Diagnostics, "Error configuring alertgroup", "Setting either `annotations` or `for` when using `record` is not allowed.")
+		}
+	}
+}
+
 // Schema defines the schema for the resource.
 func (a *alertGroupResource) Schema(_ context.Context, _ resource.SchemaRequest, resp *resource.SchemaResponse) {
 	resp.Schema = schema.Schema{
@@ -174,7 +211,7 @@ func (a *alertGroupResource) Schema(_ context.Context, _ resource.SchemaRequest,
 					Attributes: map[string]schema.Attribute{
 						"alert": schema.StringAttribute{
 							Description: descriptions["alert"],
-							Required:    true,
+							Optional:    true,
 							Validators: []validator.String{
 								stringvalidator.RegexMatches(
 									regexp.MustCompile(`^[a-zA-Z0-9-]+$`),
@@ -222,6 +259,17 @@ func (a *alertGroupResource) Schema(_ context.Context, _ resource.SchemaRequest,
 								mapvalidator.SizeAtMost(5),
 							},
 						},
+						"record": schema.StringAttribute{
+							Description: descriptions["record"],
+							Optional:    true,
+							Validators: []validator.String{
+								stringvalidator.RegexMatches(
+									regexp.MustCompile(`^[a-zA-Z0-9:_]+$`),
+									"must match expression",
+								),
+								stringvalidator.LengthBetween(1, 300),
+							},
+						},
 					},
 				},
 			},
@@ -468,6 +516,14 @@ func toRulesPayload(ctx context.Context, model *Model) ([]observability.UpdateAl
 			oarr.Annotations = &annotations
 		}
 
+		if !utils.IsUndefined(rule.Record) {
+			record := conversion.StringValueToPointer(rule.Record)
+			if record == nil {
+				return nil, fmt.Errorf("found nil record for rule[%d]", i)
+			}
+			oarr.Record = record
+		}
+
 		oarrs = append(oarrs, oarr)
 	}
 
@@ -539,6 +595,7 @@ func mapRules(_ context.Context, alertGroup *observability.AlertGroup, model *Mo
 			"for":         types.StringPointerValue(r.For),
 			"labels":      types.MapNull(types.StringType),
 			"annotations": types.MapNull(types.StringType),
+			"record":      types.StringPointerValue(r.Record),
 		}
 
 		if r.Labels != nil {

stackit/internal/services/observability/alertgroup/resource_test.go

@@ -72,6 +72,7 @@ func TestToCreatePayload(t *testing.T) {
 										"k": types.StringValue("v"),
 									},
 								),
+								"record": types.StringValue("record"),
 							},
 						),
 					},
@@ -91,6 +92,7 @@ func TestToCreatePayload(t *testing.T) {
 						Labels: &map[string]interface{}{
 							"k": "v",
 						},
+						Record: utils.Ptr("record"),
 					},
 				},
 			},
@@ -153,6 +155,7 @@ func TestToRulesPayload(t *testing.T) {
 						"annotations": types.MapValueMust(types.StringType, map[string]attr.Value{
 							"note": types.StringValue("important"),
 						}),
+						"record": types.StringValue("record"),
 					}),
 				}),
 			},
@@ -167,6 +170,7 @@ func TestToRulesPayload(t *testing.T) {
 					Annotations: &map[string]interface{}{
 						"note": "important",
 					},
+					Record: utils.Ptr("record"),
 				},
 			},
 			expectErr: false,
@@ -181,6 +185,7 @@ func TestToRulesPayload(t *testing.T) {
 						"for":         types.StringValue("5s"),
 						"labels":      types.MapNull(types.StringType),
 						"annotations": types.MapNull(types.StringType),
+						"record":      types.StringValue("record1"),
 					}),
 					types.ObjectValueMust(ruleTypes, map[string]attr.Value{
 						"alert":      types.StringValue("alert2"),
@@ -192,14 +197,16 @@ func TestToRulesPayload(t *testing.T) {
 						"annotations": types.MapValueMust(types.StringType, map[string]attr.Value{
 							"note": types.StringValue("important"),
 						}),
+						"record": types.StringValue("record2"),
 					}),
 				}),
 			},
 			expect: []observability.UpdateAlertgroupsRequestInnerRulesInner{
 				{
-					Alert: utils.Ptr("alert1"),
-					Expr:  utils.Ptr("expr1"),
-					For:   utils.Ptr("5s"),
+					Alert:  utils.Ptr("alert1"),
+					Expr:   utils.Ptr("expr1"),
+					For:    utils.Ptr("5s"),
+					Record: utils.Ptr("record1"),
 				},
 				{
 					Alert: utils.Ptr("alert2"),
@@ -211,6 +218,7 @@ func TestToRulesPayload(t *testing.T) {
 					Annotations: &map[string]interface{}{
 						"note": "important",
 					},
+					Record: utils.Ptr("record2"),
 				},
 			},
 			expectErr: false,

stackit/internal/services/observability/observability_acc_test.go

@@ -39,6 +39,7 @@ var testConfigVarsMin = config.Variables{
 	"alertgroup_name":           config.StringVariable(fmt.Sprintf("tf-acc-ag%s", acctest.RandStringFromCharSet(5, acctest.CharSetAlphaNum))),
 	"alert_rule_name":           config.StringVariable("alert1"),
 	"alert_rule_expression":     config.StringVariable(alert_rule_expression),
+	"record_rule_name":          config.StringVariable("record1"),
 	"instance_name":             config.StringVariable(fmt.Sprintf("tf-acc-i%s", acctest.RandStringFromCharSet(5, acctest.CharSetAlphaNum))),
 	"plan_name":                 config.StringVariable("Observability-Medium-EU01"),
 	"logalertgroup_name":        config.StringVariable(fmt.Sprintf("tf-acc-lag%s", acctest.RandStringFromCharSet(5, acctest.CharSetAlphaNum))),
@@ -55,6 +56,7 @@ var testConfigVarsMax = config.Variables{
 	"alertgroup_name":            config.StringVariable(fmt.Sprintf("tf-acc-ag%s", acctest.RandStringFromCharSet(5, acctest.CharSetAlphaNum))),
 	"alert_rule_name":            config.StringVariable("alert1"),
 	"alert_rule_expression":      config.StringVariable(alert_rule_expression),
+	"record_rule_name":           config.StringVariable("record1"),
 	"instance_name":              config.StringVariable(fmt.Sprintf("tf-acc-i%s", acctest.RandStringFromCharSet(5, acctest.CharSetAlphaNum))),
 	"plan_name":                  config.StringVariable("Observability-Medium-EU01"),
 	"logalertgroup_name":         config.StringVariable(fmt.Sprintf("tf-acc-lag%s", acctest.RandStringFromCharSet(5, acctest.CharSetAlphaNum))),
@@ -215,6 +217,8 @@ func TestAccResourceMin(t *testing.T) {
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "name", testutil.ConvertConfigVariable(testConfigVarsMin["alertgroup_name"])),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.alert", testutil.ConvertConfigVariable(testConfigVarsMin["alert_rule_name"])),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.expression", alert_rule_expression),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.record", testutil.ConvertConfigVariable(testConfigVarsMin["record_rule_name"])),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.expression", alert_rule_expression),
 
 					// logalertgroup
 					resource.TestCheckResourceAttr("stackit_observability_logalertgroup.logalertgroup", "project_id", testutil.ConvertConfigVariable(testConfigVarsMin["project_id"])),
@@ -302,6 +306,8 @@ func TestAccResourceMin(t *testing.T) {
 					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "name", testutil.ConvertConfigVariable(testConfigVarsMin["alertgroup_name"])),
 					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "rules.0.alert", testutil.ConvertConfigVariable(testConfigVarsMin["alert_rule_name"])),
 					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "rules.0.expression", alert_rule_expression),
+					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "rules.1.record", testutil.ConvertConfigVariable(testConfigVarsMin["record_rule_name"])),
+					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "rules.1.expression", alert_rule_expression),
 
 					// logalertgroup
 					resource.TestCheckResourceAttr("data.stackit_observability_logalertgroup.logalertgroup", "project_id", testutil.ConvertConfigVariable(testConfigVarsMin["project_id"])),
@@ -460,6 +466,8 @@ func TestAccResourceMin(t *testing.T) {
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "name", testutil.ConvertConfigVariable(testConfigVarsMin["alertgroup_name"])),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.alert", testutil.ConvertConfigVariable(configVarsMinUpdated()["alert_rule_name"])),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.expression", alert_rule_expression),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.record", testutil.ConvertConfigVariable(configVarsMinUpdated()["record_rule_name"])),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.expression", alert_rule_expression),
 
 					// logalertgroup
 					resource.TestCheckResourceAttr("stackit_observability_logalertgroup.logalertgroup", "project_id", testutil.ConvertConfigVariable(testConfigVarsMin["project_id"])),
@@ -606,6 +614,9 @@ func TestAccResourceMax(t *testing.T) {
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.expression", alert_rule_expression),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.for", testutil.ConvertConfigVariable(testConfigVarsMax["alert_for_time"])),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.labels.label1", testutil.ConvertConfigVariable(testConfigVarsMax["alert_label"])),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.record", testutil.ConvertConfigVariable(testConfigVarsMax["record_rule_name"])),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.expression", alert_rule_expression),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.labels.label1", testutil.ConvertConfigVariable(testConfigVarsMax["alert_label"])),
 
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.annotations.annotation1", testutil.ConvertConfigVariable(testConfigVarsMax["alert_annotation"])),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "interval", testutil.ConvertConfigVariable(testConfigVarsMax["alert_interval"])),
@@ -776,6 +787,9 @@ func TestAccResourceMax(t *testing.T) {
 					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "rules.0.labels.label1", testutil.ConvertConfigVariable(testConfigVarsMax["alert_label"])),
 					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "rules.0.annotations.annotation1", testutil.ConvertConfigVariable(testConfigVarsMax["alert_annotation"])),
 					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "interval", testutil.ConvertConfigVariable(testConfigVarsMax["alert_interval"])),
+					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "rules.1.record", testutil.ConvertConfigVariable(testConfigVarsMax["record_rule_name"])),
+					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "rules.1.expression", alert_rule_expression),
+					resource.TestCheckResourceAttr("data.stackit_observability_alertgroup.alertgroup", "rules.1.labels.label1", testutil.ConvertConfigVariable(testConfigVarsMax["alert_label"])),
 
 					// logalertgroup
 					resource.TestCheckResourceAttr("stackit_observability_logalertgroup.logalertgroup", "project_id", testutil.ConvertConfigVariable(testConfigVarsMax["project_id"])),
@@ -1006,6 +1020,9 @@ func TestAccResourceMax(t *testing.T) {
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.expression", alert_rule_expression_updated),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.for", testutil.ConvertConfigVariable(testConfigVarsMax["alert_for_time"])),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.labels.label1", testutil.ConvertConfigVariable(testConfigVarsMax["alert_label"])),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.record", testutil.ConvertConfigVariable(testConfigVarsMax["record_rule_name"])),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.expression", alert_rule_expression_updated),
+					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.1.labels.label1", testutil.ConvertConfigVariable(testConfigVarsMax["alert_label"])),
 
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "rules.0.annotations.annotation1", testutil.ConvertConfigVariable(testConfigVarsMax["alert_annotation"])),
 					resource.TestCheckResourceAttr("stackit_observability_alertgroup.alertgroup", "interval", testutil.ConvertConfigVariable(configVarsMaxUpdated()["alert_interval"])),

stackit/internal/services/observability/testdata/resource-max.tf

@@ -8,6 +8,7 @@ variable "alert_for_time" {}
 variable "alert_label" {}
 variable "alert_annotation" {}
 variable "alert_interval" {}
+variable "record_rule_name" {}
 
 variable "instance_name" {}
 variable "plan_name" {}
@@ -89,6 +90,13 @@ resource "stackit_observability_alertgroup" "alertgroup" {
       annotations = {
         annotation1 = var.alert_annotation
       }
+    },
+    {
+      record     = var.record_rule_name
+      expression = var.alert_rule_expression
+      labels = {
+        label1 = var.alert_label
+      }
     }
   ]
   interval = var.alert_interval

stackit/internal/services/observability/testdata/resource-min.tf

@@ -4,6 +4,7 @@ variable "project_id" {}
 variable "alertgroup_name" {}
 variable "alert_rule_name" {}
 variable "alert_rule_expression" {}
+variable "record_rule_name" {}
 
 variable "instance_name" {}
 variable "plan_name" {}
@@ -27,6 +28,10 @@ resource "stackit_observability_alertgroup" "alertgroup" {
     {
       alert      = var.alert_rule_name
       expression = var.alert_rule_expression
+    },
+    {
+      record     = var.record_rule_name
+      expression = var.alert_rule_expression
     }
   ]
 }