feat: add minio support (#1048)

* feat: add minio support

Signed-off-by: Aris Boutselis <arisboutselis08@gmail.com>

* feat: add TLS skip for custom https minio endpoint

Signed-off-by: Aris Boutselis <arisboutselis08@gmail.com>

* feat: update cache with the new proto schema

Signed-off-by: Aris Boutselis <arisboutselis08@gmail.com>

---------

Signed-off-by: Aris Boutselis <arisboutselis08@gmail.com>

README.md

@@ -431,6 +431,8 @@ _Adding a remote cache_
  * AWS S3
    *  _As a prerequisite `AWS_ACCESS_KEY_ID` and `AWS_SECRET_ACCESS_KEY` are required as environmental variables._
    * Configuration, ``` k8sgpt cache add s3 --region <aws region> --bucket <name> ```
+   * Minio Configuration with HTTP endpoint ``` k8sgpt cache add s3 --bucket <name> --endpoint <http://localhost:9000>```
+   * Minio Configuration with HTTPs endpoint, skipping TLS verification ``` k8sgpt cache add s3 --bucket <name> --endpoint <https://localhost:9000> --insecure```
      * K8sGPT will create the bucket if it does not exist
  * Azure Storage
    * We support a number of [techniques](https://learn.microsoft.com/en-us/azure/developer/go/azure-sdk-authentication?tabs=bash#2-authenticate-with-azure) to authenticate against Azure

cmd/cache/add.go

@@ -30,6 +30,8 @@ var (
 	storageAccount string
 	containerName  string
 	projectId      string
+	endpoint       string
+	insecure       bool
 )
 
 // addCmd represents the add command
@@ -48,7 +50,7 @@ var addCmd = &cobra.Command{
 		}
 		fmt.Println(color.YellowString("Adding remote based cache"))
 		cacheType := args[0]
-		remoteCache, err := cache.NewCacheProvider(cacheType, bucketname, region, storageAccount, containerName, projectId)
+		remoteCache, err := cache.NewCacheProvider(cacheType, bucketName, region, endpoint, storageAccount, containerName, projectId, insecure)
 		if err != nil {
 			color.Red("Error: %v", err)
 			os.Exit(1)
@@ -63,9 +65,10 @@ var addCmd = &cobra.Command{
 
 func init() {
 	CacheCmd.AddCommand(addCmd)
-	addCmd.Flags().StringVarP(&region, "region", "r", "", "The region to use for the AWS S3 or GCS cache")
+	addCmd.Flags().StringVarP(&region, "region", "r", "us-east-1", "The region to use for the AWS S3 or GCS cache")
-	addCmd.Flags().StringVarP(&bucketname, "bucket", "b", "", "The name of the AWS S3 bucket to use for the cache")
+	addCmd.Flags().StringVarP(&endpoint, "endpoint", "e", "", "The S3 or minio endpoint")
-	addCmd.MarkFlagsRequiredTogether("region", "bucket")
+	addCmd.Flags().BoolVarP(&insecure, "insecure", "i", false, "Skip TLS verification for S3/Minio custom endpoint")
+	addCmd.Flags().StringVarP(&bucketName, "bucket", "b", "", "The name of the AWS S3 bucket to use for the cache")
 	addCmd.Flags().StringVarP(&projectId, "projectid", "p", "", "The GCP project ID")
 	addCmd.Flags().StringVarP(&storageAccount, "storageacc", "s", "", "The Azure storage account name of the container")
 	addCmd.Flags().StringVarP(&containerName, "container", "c", "", "The Azure container name to use for the cache")

cmd/cache/cache.go

@@ -18,10 +18,6 @@ import (
 	"github.com/spf13/cobra"
 )
 
-var (
-	bucketname string
-)
-
 // cacheCmd represents the cache command
 var CacheCmd = &cobra.Command{
 	Use:   "cache",

go.mod

@@ -26,7 +26,7 @@ require github.com/adrg/xdg v0.4.0
 require (
 	buf.build/gen/go/k8sgpt-ai/k8sgpt/grpc-ecosystem/gateway/v2 v2.19.1-20240213144542-6e830f3fdf19.1
 	buf.build/gen/go/k8sgpt-ai/k8sgpt/grpc/go v1.3.0-20240213144542-6e830f3fdf19.2
-	buf.build/gen/go/k8sgpt-ai/k8sgpt/protocolbuffers/go v1.32.0-20240213144542-6e830f3fdf19.1
+	buf.build/gen/go/k8sgpt-ai/k8sgpt/protocolbuffers/go v1.33.0-20240406062209-1cc152efbf5c.1
 	cloud.google.com/go/storage v1.40.0
 	cloud.google.com/go/vertexai v0.7.1
 	github.com/Azure/azure-sdk-for-go/sdk/azidentity v1.5.1

go.sum

@@ -17,6 +17,8 @@ buf.build/gen/go/k8sgpt-ai/k8sgpt/grpc/go v1.3.0-20240213144542-6e830f3fdf19.2/g
 buf.build/gen/go/k8sgpt-ai/k8sgpt/protocolbuffers/go v1.28.1-20240213144542-6e830f3fdf19.4/go.mod h1:WyRj8OIsAABLNsAELw73BT16v7vvJdEVv771fxX9pJI=
 buf.build/gen/go/k8sgpt-ai/k8sgpt/protocolbuffers/go v1.32.0-20240213144542-6e830f3fdf19.1 h1:YJ13kOhQHoOe4eMd3CqFFeTyQvVBRmeBGLqH/QLuOjQ=
 buf.build/gen/go/k8sgpt-ai/k8sgpt/protocolbuffers/go v1.32.0-20240213144542-6e830f3fdf19.1/go.mod h1:4QGFkgjJ3Wm1EBhQ6tOkaKihV4bFF6DvhTk1r9ZhFOE=
+buf.build/gen/go/k8sgpt-ai/k8sgpt/protocolbuffers/go v1.33.0-20240406062209-1cc152efbf5c.1 h1:rx7Res/Ji345EbuTWps4sxH2JQHmvEyoe/5wLFZW8nA=
+buf.build/gen/go/k8sgpt-ai/k8sgpt/protocolbuffers/go v1.33.0-20240406062209-1cc152efbf5c.1/go.mod h1:BQLbAK4GBQ4xEyMX/G1KEt+4vsa6EiOPD/Rb3VswwI0=
 cloud.google.com/go v0.26.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
 cloud.google.com/go v0.34.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
 cloud.google.com/go v0.38.0/go.mod h1:990N+gfupTy94rShfmMCWGDn0LpTmnzTp2qbd1dvSRU=

pkg/cache/cache.go

@@ -47,7 +47,7 @@ func ParseCacheConfiguration() (CacheProvider, error) {
 	return cacheInfo, nil
 }
 
-func NewCacheProvider(cacheType, bucketname, region, storageAccount, containerName, projectId string) (CacheProvider, error) {
+func NewCacheProvider(cacheType, bucketname, region, endpoint, storageAccount, containerName, projectId string, insecure bool) (CacheProvider, error) {
 	cProvider := CacheProvider{}
 
 	switch {
@@ -61,6 +61,8 @@ func NewCacheProvider(cacheType, bucketname, region, storageAccount, containerNa
 	case cacheType == "s3":
 		cProvider.S3.BucketName = bucketname
 		cProvider.S3.Region = region
+		cProvider.S3.Endpoint = endpoint
+		cProvider.S3.InsecureSkipVerify = insecure
 	default:
 		return CacheProvider{}, status.Error(codes.Internal, fmt.Sprintf("%s is not a valid option", cacheType))
 	}

pkg/cache/s3_based.go

@@ -2,7 +2,9 @@ package cache
 
 import (
 	"bytes"
+	"crypto/tls"
 	"log"
+	"net/http"
 
 	"github.com/aws/aws-sdk-go/aws"
 	"github.com/aws/aws-sdk-go/aws/session"
@@ -19,15 +21,14 @@ type S3Cache struct {
 type S3CacheConfiguration struct {
 	Region             string `mapstructure:"region" yaml:"region,omitempty"`
 	BucketName         string `mapstructure:"bucketname" yaml:"bucketname,omitempty"`
+	Endpoint           string `mapstructure:"endpoint" yaml:"endpoint,omitempty"`
+	InsecureSkipVerify bool   `mapstructure:"insecure" yaml:"insecure,omitempty"`
 }
 
 func (s *S3Cache) Configure(cacheInfo CacheProvider) error {
 	if cacheInfo.S3.BucketName == "" {
 		log.Fatal("Bucket name not configured")
 	}
-	if cacheInfo.S3.Region == "" {
-		log.Fatal("Region not configured")
-	}
 	s.bucketName = cacheInfo.S3.BucketName
 
 	sess := session.Must(session.NewSessionWithOptions(session.Options{
@@ -36,6 +37,15 @@ func (s *S3Cache) Configure(cacheInfo CacheProvider) error {
 			Region: aws.String(cacheInfo.S3.Region),
 		},
 	}))
+	if cacheInfo.S3.Endpoint != "" {
+		sess.Config.Endpoint = &cacheInfo.S3.Endpoint
+		sess.Config.S3ForcePathStyle = aws.Bool(true)
+		transport := &http.Transport{
+			TLSClientConfig: &tls.Config{InsecureSkipVerify: cacheInfo.S3.InsecureSkipVerify},
+		}
+		customClient := &http.Client{Transport: transport}
+		sess.Config.HTTPClient = customClient
+	}
 
 	s3Client := s3.New(sess)
 

pkg/server/config.go

@@ -9,6 +9,16 @@ import (
 	"google.golang.org/grpc/status"
 )
 
+const (
+	notUsedBucket        = ""
+	notUsedRegion        = ""
+	notUsedEndpoint      = ""
+	notUsedStorageAcc    = ""
+	notUsedContainerName = ""
+	notUsedProjectId     = ""
+	notUsedInsecure      = false
+)
+
 func (h *handler) AddConfig(ctx context.Context, i *schemav1.AddConfigRequest) (*schemav1.AddConfigResponse, error,
 ) {
 
@@ -23,11 +33,11 @@ func (h *handler) AddConfig(ctx context.Context, i *schemav1.AddConfigRequest) (
 
 		switch i.Cache.GetCacheType().(type) {
 		case *schemav1.Cache_AzureCache:
-			remoteCache, err = cache.NewCacheProvider("azure", "", "", i.Cache.GetAzureCache().StorageAccount, i.Cache.GetAzureCache().ContainerName, "")
+			remoteCache, err = cache.NewCacheProvider("azure", notUsedBucket, notUsedRegion, notUsedEndpoint, i.Cache.GetAzureCache().StorageAccount, i.Cache.GetAzureCache().ContainerName, notUsedProjectId, notUsedInsecure)
 		case *schemav1.Cache_S3Cache:
-			remoteCache, err = cache.NewCacheProvider("s3", i.Cache.GetS3Cache().BucketName, i.Cache.GetS3Cache().Region, "", "", "")
+			remoteCache, err = cache.NewCacheProvider("s3", i.Cache.GetS3Cache().BucketName, i.Cache.GetS3Cache().Region, i.Cache.GetS3Cache().Endpoint, notUsedStorageAcc, notUsedContainerName, notUsedProjectId, i.Cache.GetS3Cache().Insecure)
 		case *schemav1.Cache_GcsCache:
-			remoteCache, err = cache.NewCacheProvider("gcs", i.Cache.GetGcsCache().BucketName, i.Cache.GetGcsCache().Region, "", "", i.Cache.GetGcsCache().GetProjectId())
+			remoteCache, err = cache.NewCacheProvider("gcs", i.Cache.GetGcsCache().BucketName, i.Cache.GetGcsCache().Region, notUsedEndpoint, notUsedStorageAcc, notUsedContainerName, i.Cache.GetGcsCache().GetProjectId(), notUsedInsecure)
 		default:
 			return resp, status.Error(codes.InvalidArgument, "Invalid cache configuration")
 		}