feat: initial Prometheus analyzers (#855)

* feat: initial Prometheus analyzers Added a prometheus integration with two analyzers: 1. PrometheusConfigValidate 2. PrometheusConfigRelabelReport The integration does not deploy any Prometheus stack in the cluster. Instead, it searches the provided --namespace for a Prometheus configuration, stored in a ConfigMap or Secret. If it finds one, it unmarshals it into memory and runs the analyzers on it. PrometheusConfigValidate checks if the actual Prometheus configuration is valid or has any errors. PrometheusConfigRelabelReport tries to distill the scrape config relabeling rules to give a concise label set per job that targets need to have to be scraped. This analyzer is unconventional, in that it does not necessarily mean there are issues with the config. It merely tries to give a human-readable explanation of the relabel rules it discovers, leaning on the LLM and prompt. Tested on both kube-prometheus and Google Managed Prometheus stacks. Signed-off-by: Daniel Clark <danielclark@google.com> * review: feedback cycle 1 Simplify ConfigValidate prompt and add comments. Signed-off-by: Daniel Clark <danielclark@google.com> * review: feedback cycle 2 Add Prometheus configuration discovery to integration activate command. Also improve logging to make this more clear to users. Signed-off-by: Daniel Clark <danielclark@google.com> --------- Signed-off-by: Daniel Clark <danielclark@google.com>
k8sgpt-ai · Jan 12, 2024 · 45fa827 · 45fa827
1 parent 4106d39
commit 45fa827
Show file tree

Hide file tree

Showing 7 changed files with 1,810 additions and 8 deletions.
diff --git a/go.mod b/go.mod
@@ -7,7 +7,8 @@ require (
 	github.com/fatih/color v1.16.0
 	github.com/magiconair/properties v1.8.7
 	github.com/mittwald/go-helm-client v0.12.5
-	github.com/sashabaranov/go-openai v1.17.11
+	github.com/prometheus/prometheus v1.8.2-0.20211119115433-692a54649ed7
+	github.com/sashabaranov/go-openai v1.17.10
 	github.com/schollz/progressbar/v3 v3.14.1
 	github.com/spf13/cobra v1.8.0
 	github.com/spf13/viper v1.18.2
@@ -49,23 +50,29 @@ require (
 	github.com/Azure/azure-sdk-for-go/sdk/internal v1.5.1 // indirect
 	github.com/AzureAD/microsoft-authentication-library-for-go v1.2.0 // indirect
 	github.com/Microsoft/hcsshim v0.11.4 // indirect
+	github.com/alecthomas/units v0.0.0-20211218093645-b94a6e3cc137 // indirect
 	github.com/anchore/go-struct-converter v0.0.0-20230627203149-c72ef8859ca9 // indirect
 	github.com/cohere-ai/tokenizer v1.1.1 // indirect
 	github.com/containerd/log v0.1.0 // indirect
 	github.com/distribution/reference v0.5.0 // indirect
 	github.com/dlclark/regexp2 v1.10.0 // indirect
 	github.com/evanphx/json-patch/v5 v5.7.0 // indirect
 	github.com/felixge/httpsnoop v1.0.4 // indirect
+	github.com/go-kit/log v0.2.1 // indirect
+	github.com/go-logfmt/logfmt v0.5.1 // indirect
 	github.com/golang-jwt/jwt/v5 v5.0.0 // indirect
 	github.com/golang/groupcache v0.0.0-20210331224755-41bb18bfe9da // indirect
 	github.com/google/gnostic-models v0.6.9-0.20230804172637-c7be7c783f49 // indirect
 	github.com/google/s2a-go v0.1.7 // indirect
 	github.com/googleapis/enterprise-certificate-proxy v0.3.2 // indirect
 	github.com/googleapis/gax-go/v2 v2.12.0 // indirect
 	github.com/jmespath/go-jmespath v0.4.0 // indirect
+	github.com/jpillora/backoff v1.0.0 // indirect
 	github.com/kylelemons/godebug v1.1.0 // indirect
 	github.com/matttproud/golang_protobuf_extensions/v2 v2.0.0 // indirect
+	github.com/mwitkow/go-conntrack v0.0.0-20190716064945-2f068394615f // indirect
 	github.com/pkg/browser v0.0.0-20210911075715-681adbf594b8 // indirect
+	github.com/prometheus/common/sigv4 v0.1.0 // indirect
 	github.com/sagikazarmark/locafero v0.4.0 // indirect
 	github.com/sagikazarmark/slog-shim v0.1.0 // indirect
 	github.com/sourcegraph/conc v0.3.0 // indirect
@@ -209,7 +216,7 @@ require (
 	google.golang.org/protobuf v1.31.0 // indirect
 	gopkg.in/inf.v0 v0.9.1 // indirect
 	gopkg.in/ini.v1 v1.67.0 // indirect
-	gopkg.in/yaml.v2 v2.4.0 // indirect
+	gopkg.in/yaml.v2 v2.4.0
 	gopkg.in/yaml.v3 v3.0.1 // indirect
 	k8s.io/apiextensions-apiserver v0.28.4
 	k8s.io/apiserver v0.28.4 // indirect

diff --git a/go.sum b/go.sum
diff --git a/pkg/ai/prompts.go b/pkg/ai/prompts.go
@@ -8,10 +8,52 @@ const (
 	`
 	trivy_vuln_prompt = "Explain the following trivy scan result and the detail risk or root cause of the CVE ID, then provide a solution. Response in %s: %s"
 	trivy_conf_prompt = "Explain the following trivy scan result and the detail risk or root cause of the security check, then provide a solution."
+
+	prom_conf_prompt = `Simplify the following Prometheus error message delimited by triple dashes written in --- %s --- language; --- %s ---.
+	This error came when validating the Prometheus configuration file.
+	Provide step by step instructions to fix, with suggestions, referencing Prometheus documentation if relevant.
+	Write the output in the following format in no more than 300 characters:
+	Error: {Explain error here}
+	Solution: {Step by step solution here}
+	`
+
+	prom_relabel_prompt = `
+	Return your prompt in this language: %s, beginning with
+	The following is a list of the form:
+	job_name:
+	{Prometheus job_name}
+	relabel_configs:
+	{Prometheus relabel_configs}
+	kubernetes_sd_configs:
+	{Prometheus service discovery config}
+	---
+	%s
+	---
+	For each job_name, describe the Kubernetes service and pod labels,
+	namespaces, ports, and containers they match.
+	Return the message:
+	Discovered and parsed Prometheus scrape configurations.
+	For targets to be scraped by Prometheus, ensure they are running with
+	at least one of the following label sets:
+	Then for each job, write this format:
+	- Job: {job_name}
+	  - Service Labels:
+	    - {list of service labels}
+	  - Pod Labels:
+	    - {list of pod labels}
+	  - Namespaces:
+	    - {list of namespaces}
+	  - Ports:
+	    - {list of ports}
+	  - Containers:
+	    - {list of container names}
+	`
 )
 
 var PromptMap = map[string]string{
-	"default":             default_prompt,
-	"VulnerabilityReport": trivy_vuln_prompt, // for Trivy integration, the key should match `Result.Kind` in pkg/common/types.go
-	"ConfigAuditReport":   trivy_conf_prompt,
+	"default":                       default_prompt,
+	"VulnerabilityReport":           trivy_vuln_prompt, // for Trivy integration, the key should match `Result.Kind` in pkg/common/types.go
+	"ConfigAuditReport":             trivy_conf_prompt,
+	"PrometheusConfigValidate":      prom_conf_prompt,
+	"PrometheusConfigRelabelReport": prom_relabel_prompt,
 }
diff --git a/pkg/integration/integration.go b/pkg/integration/integration.go
@@ -18,6 +18,7 @@ import (
 	"fmt"
 
 	"github.com/k8sgpt-ai/k8sgpt/pkg/common"
+	"github.com/k8sgpt-ai/k8sgpt/pkg/integration/prometheus"
 	"github.com/k8sgpt-ai/k8sgpt/pkg/integration/trivy"
 	"github.com/k8sgpt-ai/k8sgpt/pkg/util"
 	"github.com/spf13/viper"
@@ -44,7 +45,8 @@ type Integration struct {
 }
 
 var integrations = map[string]IIntegration{
-	"trivy": trivy.NewTrivy(),
+	"trivy":      trivy.NewTrivy(),
+	"prometheus": prometheus.NewPrometheus(),
 }
 
 func NewIntegration() *Integration {