query, rule: make endpoint discovery dynamically reloadable (#7890)

* Removed previously deprecated and hidden flags to configure endpoints ( --rule, --target, ...)
* Added new flags --endpoint.sd-config, --endpoint-sd-config-reload-interval to configure a dynamic SD file
* Moved endpoint set construction into cmd/thanos/endpointset.go for a little cleanup
* Renamed "thanos_(querier/ruler)_duplicated_store_addresses_total" to
  "thanos_(querier/ruler)_duplicated_endpoint_addresses_total"

The new config makes it possible to also set "strict" and "group" flags on the endpoint instead
of only their addresses, making it possible to have file based service discovery for endpoint groups too.

Signed-off-by: Michael Hoffmann <[email protected]>
Signed-off-by: Michael Hoffmann <[email protected]>

CHANGELOG.md

@@ -26,6 +26,8 @@ We use *breaking :warning:* to mark changes that are not backward compatible (re
 
 ### Changed
 
+- [#7890](https://github.com/thanos-io/thanos/pull/7890) Query,Ruler: *breaking :warning:* deprecated `--store.sd-file` and `--store.sd-interval` to be replaced with `--endpoint.sd-config` and `--endpoint-sd-config-reload-interval`; removed legacy flags to pass endpoints `--store`, `--metadata`, `--rule`, `--exemplar`.
+
 ### Removed
 
 ## [v0.37.2](https://github.com/thanos-io/thanos/tree/release-0.37) - 11.12.2024

cmd/thanos/config.go

@@ -14,11 +14,17 @@ import (
 
 	"github.com/KimMachineGun/automemlimit/memlimit"
 	extflag "github.com/efficientgo/tools/extkingpin"
+	"github.com/go-kit/log"
+	"github.com/opentracing/opentracing-go"
 	"github.com/pkg/errors"
+	"google.golang.org/grpc"
 
+	"github.com/prometheus/client_golang/prometheus"
 	"github.com/prometheus/common/model"
 	"github.com/prometheus/prometheus/model/labels"
 
+	"github.com/thanos-io/thanos/pkg/extgrpc"
+	"github.com/thanos-io/thanos/pkg/extgrpc/snappy"
 	"github.com/thanos-io/thanos/pkg/extkingpin"
 	"github.com/thanos-io/thanos/pkg/shipper"
 )
@@ -58,6 +64,38 @@ func (gc *grpcConfig) registerFlag(cmd extkingpin.FlagClause) *grpcConfig {
 	return gc
 }
 
+type grpcClientConfig struct {
+	secure            bool
+	skipVerify        bool
+	cert, key, caCert string
+	serverName        string
+	compression       string
+}
+
+func (gc *grpcClientConfig) registerFlag(cmd extkingpin.FlagClause) *grpcClientConfig {
+	cmd.Flag("grpc-client-tls-secure", "Use TLS when talking to the gRPC server").Default("false").BoolVar(&gc.secure)
+	cmd.Flag("grpc-client-tls-skip-verify", "Disable TLS certificate verification i.e self signed, signed by fake CA").Default("false").BoolVar(&gc.skipVerify)
+	cmd.Flag("grpc-client-tls-cert", "TLS Certificates to use to identify this client to the server").Default("").StringVar(&gc.cert)
+	cmd.Flag("grpc-client-tls-key", "TLS Key for the client's certificate").Default("").StringVar(&gc.key)
+	cmd.Flag("grpc-client-tls-ca", "TLS CA Certificates to use to verify gRPC servers").Default("").StringVar(&gc.caCert)
+	cmd.Flag("grpc-client-server-name", "Server name to verify the hostname on the returned gRPC certificates. See https://tools.ietf.org/html/rfc4366#section-3.1").Default("").StringVar(&gc.serverName)
+	compressionOptions := strings.Join([]string{snappy.Name, compressionNone}, ", ")
+	cmd.Flag("grpc-compression", "Compression algorithm to use for gRPC requests to other clients. Must be one of: "+compressionOptions).Default(compressionNone).EnumVar(&gc.compression, snappy.Name, compressionNone)
+
+	return gc
+}
+
+func (gc *grpcClientConfig) dialOptions(logger log.Logger, reg prometheus.Registerer, tracer opentracing.Tracer) ([]grpc.DialOption, error) {
+	dialOpts, err := extgrpc.StoreClientGRPCOpts(logger, reg, tracer, gc.secure, gc.skipVerify, gc.cert, gc.key, gc.caCert, gc.serverName)
+	if err != nil {
+		return nil, errors.Wrapf(err, "building gRPC client")
+	}
+	if gc.compression != compressionNone {
+		dialOpts = append(dialOpts, grpc.WithDefaultCallOptions(grpc.UseCompressor(gc.compression)))
+	}
+	return dialOpts, nil
+}
+
 type httpConfig struct {
 	bindAddress string
 	tlsConfig   string