Support load-balanced runner groups for multitenant compute isolation (#814)

* Initial stab at the protocol

* initial protocol sketch for node pool manager

* Added http header frame as a message

* Force the use of WithAgent variants when creating a server

* adds grpc models for node pool manager plus go deps

* Naming things is really hard

* Merge (and optionally purge) details received by the NPM

* WIP: starting to add the runner-side functionality of the new data plane

* WIP: Basic startup of grpc server for pure runner. Needs proper certs.

* Go fmt

* Initial agent for LB nodes.

* Agent implementation for LB nodes.

* Pass keys and certs to LB node agent.

* Remove accidentally left reference to env var.

* Add env variables for certificate files

* stub out the capacity and group membership server channels

* implement server-side runner manager service

* removes unused variable

* fixes build error

* splits up GetCall and GetLBGroupId

* Change LB node agent to use TLS connection.

* Encode call model as JSON to send to runner node.

* Use hybrid client in LB node agent.

This should provide access to get app and route information for the call
from an API node.

* More error handling on the pure runner side

* Tentative fix for GetCall problem: set deadlines correctly when reserving slot

* Connect loop for LB agent to runner nodes.

* Extract runner connection function in LB agent.

* drops committed capacity counts

* Bugfix - end state tracker only in submit

* Do logs properly

* adds first pass of tracking capacity metrics in agent

* maked memory capacity metric uint64

* maked memory capacity metric uint64

* removes use of old capacity field

* adds remove capacity call

* merges overwritten reconnect logic

* First pass of a NPM

Provide a service that talks to a (simulated) CP.

- Receive incoming capacity assertions from LBs for LBGs
- expire LB requests after a short period
- ask the CP to add runners to a LBG
- note runner set changes and readvertise
- scale down by marking runners as "draining"
- shut off draining runners after some cool-down period

* add capacity update on schedule

* Send periodic capcacity metrics

Sending capcacity metrics to node pool manager

* splits grpc and api interfaces for capacity manager

* failure to advertise capacity shouldn't panic

* Add some instructions for starting DP/CP parts.

* Create the poolmanager server with TLS

* Use logrus

* Get npm compiling with cert fixups.

* Fix: pure runner should not start async processing

* brings runner, nulb and npm together

* Add field to acknowledgment to record slot allocation latency; fix a bug too

* iterating on pool manager locking issue

* raises timeout of placement retry loop

* Fix up NPM

Improve logging

Ensure that channels etc. are actually initialised in the structure
creation!

* Update the docs - runners GRPC port is 9120

* Bugfix: return runner pool accurately.

* Double locking

* Note purges as LBs stop talking to us

* Get the purging of old LBs working.

* Tweak: on restart, load runner set before making scaling decisions.

* more agent synchronization improvements

* Deal with teh CP pulling out active hosts from under us.

* lock at lbgroup level

* Send request and receive response from runner.

* Add capacity check right before slot reservation

* Pass the full Call into the receive loop.

* Wait for the data from the runner before finishing

* force runner list refresh every time

* Don't init db and mq for pure runners

* adds shutdown of npm

* fixes broken log line

* Extract an interface for the Predictor used by the NPM

* purge drained connections from npm

* Refactor of the LB agent into the agent package

* removes capacitytest wip

* Fix undefined err issue

* updating README for poolmanager set up

* ues retrying dial for lb to npm connections

* Rename lb_calls to lb_agent now that all functionality is there

* Use the right deadline and errors in LBAgent

* Make stream error flag per-call rather than global otherwise the whole runner is damaged by one call dropping

* abstracting gRPCNodePool

* Make stream error flag per-call rather than global otherwise the whole runner is damaged by one call dropping

* Add some init checks for LB and pure runner nodes

* adding some useful debug

* Fix default db and mq for lb node

* removes unreachable code, fixes typo

* Use datastore as logstore in API nodes.

This fixes a bug caused by trying to insert logs into a nil logstore. It
was nil because it wasn't being set for API nodes.

* creates placement abstraction and moves capacity APIs to NodePool

* removed TODO, added logging

* Dial reconnections for LB <-> runners

LB grpc connections to runners are established using a backoff stategy
in event of reconnections, this allows to let the LB up even in case one
of the runners go away and reconnect to it as soon as it is back.

* Add a status call to the Runner protocol

Stub at the moment. To be used for things like draindown, health checks.

* Remove comment.

* makes assign/release capacity lockless

* Fix hanging issue in lb agent when connections drop

* Add the CH hash from fnlb

Select this with FN_PLACER=ch when launching the LB.

* small improvement for locking on reloadLBGmembership

* Stabilise the list of Runenrs returned by NodePool

The NodePoolManager makes some attempt to keep the list of runner nodes advertised as
stable as possible. Let's preserve this effort in the client side. The main point of this
is to attempt to keep the same runner at the same inxed in the []Runner returned by
NodePool.Runners(lbgid); the ch algorithm likes it when this is the case.

* Factor out a generator function for the Runners so that mocks can be injected

* temporarily allow lbgroup to be specified in HTTP header, while we sort out changes to the model

* fixes bug with nil runners

* Initial work for mocking things in tests

* fix for anonymouse go routine error

* fixing lb_test to compile

* Refactor: internal objects for gRPCNodePool are now injectable, with defaults for the real world case

* Make GRPC port configurable, fix weird handling of web port too

* unit test reload Members

* check on runner creation failure

* adding nullRunner in case of failure during runner creation

* Refactored capacity advertisements/aggregations. Made grpc advertisement post asynchronous and non-blocking.

* make capacityEntry private

* Change the runner gRPC bind address.

This uses the existing `whoAmI` function, so that the gRPC server works
when the runner is running on a different host.

* Add support for multiple fixed runners to pool mgr

* Added harness for dataplane system tests, minor refactors

* Add Dockerfiles for components, along with docs.

* Doc fix: second runner needs a different name.

* Let us have three runners in system tests, why not

* The first system test running a function in API/LB/PureRunner mode

* Add unit test for Advertiser logic

* Fix issue with Pure Runner not sending the last data frame

* use config in models.Call as a temporary mechanism to override lb group ID

* make gofmt happy

* Updates documentation for how to configure lb groups for an app/route

* small refactor unit test

* Factor NodePool into its own package

* Lots of fixes to Pure Runner - concurrency woes with errors and cancellations

* New dataplane with static runnerpool (#813)

Added static node pool as default implementation

* moved nullRunner to grpc package

* remove duplication in README

* fix go vet issues

* Fix server initialisation in api tests

* Tiny logging changes in pool manager.

Using `WithError` instead of `Errorf` when appropriate.

* Change some log levels in the pure runner

* fixing readme

* moves multitenant compute documentation

* adds introduction to multitenant readme

* Proper triggering of system tests in makefile

* Fix insructions about starting up the components

* Change db file for system tests to avoid contention in parallel tests

* fixes revisions from merge

* Fix merge issue with handling of reserved slot

* renaming nulb to lb in the doc and images folder

* better TryExec sleep logic clean shutdown

In this change we implement a better way to deal with the sleep inside
the for loop during the attempt for placing a call.
Plus we added a clean way to shutdown the connections with external
component when we shut down the server.

* System_test mysql port

set mysql port for system test to a different value to the one set for
the api tests to avoid conflicts as they can run in parallel.

* change the container name for system-test

* removes flaky test TestRouteRunnerExecution pending resolution by issue #796

* amend remove_containers to remove new added containers

* Rework capacity reservation logic at a higher level for now

* LB agent implements Submit rather than delegating.

* Fix go vet linting errors

* Changed a couple of error levels

* Fix formatting

* removes commmented out test

* adds snappy to vendor directory

* updates Gopkg and vendor directories, removing snappy and addhing siphash

* wait for db containers to come up before starting the tests

* make system tests start API node on 8085 to avoid port conflict with api_tests

* avoid port conflicts with api_test.sh which are run in parallel

* fixes postgres port conflict and issue with removal of old containers

* Remove spurious println

api/server/server.go

@@ -4,17 +4,20 @@ import (
 	"bytes"
 	"context"
 	"encoding/base64"
+	"errors"
 	"fmt"
 	"net"
 	"net/http"
 	"os"
 	"path"
+	"path/filepath"
 	"strconv"
 	"strings"
 	"syscall"
 
 	"github.com/fnproject/fn/api/agent"
 	"github.com/fnproject/fn/api/agent/hybrid"
+	agent_grpc "github.com/fnproject/fn/api/agent/nodepool/grpc"
 	"github.com/fnproject/fn/api/common"
 	"github.com/fnproject/fn/api/datastore"
 	"github.com/fnproject/fn/api/id"
@@ -34,22 +37,31 @@ import (
 )
 
 const (
-	EnvLogLevel  = "FN_LOG_LEVEL"
-	EnvLogDest   = "FN_LOG_DEST"
-	EnvLogPrefix = "FN_LOG_PREFIX"
-	EnvMQURL     = "FN_MQ_URL"
-	EnvDBURL     = "FN_DB_URL"
-	EnvLOGDBURL  = "FN_LOGSTORE_URL"
-	EnvRunnerURL = "FN_RUNNER_API_URL"
-	EnvNodeType  = "FN_NODE_TYPE"
-	EnvPort      = "FN_PORT" // be careful, Gin expects this variable to be "port"
-	EnvAPICORS   = "FN_API_CORS"
-	EnvZipkinURL = "FN_ZIPKIN_URL"
+	EnvLogLevel        = "FN_LOG_LEVEL"
+	EnvLogDest         = "FN_LOG_DEST"
+	EnvLogPrefix       = "FN_LOG_PREFIX"
+	EnvMQURL           = "FN_MQ_URL"
+	EnvDBURL           = "FN_DB_URL"
+	EnvLOGDBURL        = "FN_LOGSTORE_URL"
+	EnvRunnerURL       = "FN_RUNNER_API_URL"
+	EnvNPMAddress      = "FN_NPM_ADDRESS"
+	EnvRunnerAddresses = "FN_RUNNER_ADDRESSES"
+	EnvLBPlacementAlg  = "FN_PLACER"
+	EnvNodeType        = "FN_NODE_TYPE"
+	EnvPort            = "FN_PORT" // be careful, Gin expects this variable to be "port"
+	EnvGRPCPort        = "FN_GRPC_PORT"
+	EnvAPICORS         = "FN_API_CORS"
+	EnvZipkinURL       = "FN_ZIPKIN_URL"
+	// Certificates to communicate with other FN nodes
+	EnvCert     = "FN_NODE_CERT"
+	EnvCertKey  = "FN_NODE_CERT_KEY"
+	EnvCertAuth = "FN_NODE_CERT_AUTHORITY"
 
 	// Defaults
 	DefaultLogLevel = "info"
 	DefaultLogDest  = "stderr"
 	DefaultPort     = 8080
+	DefaultGRPCPort = 9190
 )
 
 type ServerNodeType int32
@@ -57,7 +69,9 @@ type ServerNodeType int32
 const (
 	ServerTypeFull ServerNodeType = iota
 	ServerTypeAPI
+	ServerTypeLB
 	ServerTypeRunner
+	ServerTypePureRunner
 )
 
 func (s ServerNodeType) String() string {
@@ -66,8 +80,12 @@ func (s ServerNodeType) String() string {
 		return "full"
 	case ServerTypeAPI:
 		return "api"
+	case ServerTypeLB:
+		return "lb"
 	case ServerTypeRunner:
 		return "runner"
+	case ServerTypePureRunner:
+		return "pure-runner"
 	}
 }
 
@@ -75,11 +93,16 @@ type Server struct {
 	// TODO this one maybe we have `AddRoute` in extensions?
 	Router *gin.Engine
 
+	webListenPort   int
+	grpcListenPort  int
 	agent           agent.Agent
 	datastore       models.Datastore
 	mq              models.MessageQueue
 	logstore        models.LogStore
 	nodeType        ServerNodeType
+	cert            string
+	certKey         string
+	certAuthority   string
 	appListeners    *appListeners
 	rootMiddlewares []fnext.Middleware
 	apiMiddlewares  []fnext.Middleware
@@ -90,8 +113,12 @@ func nodeTypeFromString(value string) ServerNodeType {
 	switch value {
 	case "api":
 		return ServerTypeAPI
+	case "lb":
+		return ServerTypeLB
 	case "runner":
 		return ServerTypeRunner
+	case "pure-runner":
+		return ServerTypePureRunner
 	default:
 		return ServerTypeFull
 	}
@@ -102,11 +129,17 @@ func NewFromEnv(ctx context.Context, opts ...ServerOption) *Server {
 	curDir := pwd()
 	var defaultDB, defaultMQ string
 	nodeType := nodeTypeFromString(getEnv(EnvNodeType, "")) // default to full
-	if nodeType != ServerTypeRunner {
+	switch nodeType {
+	case ServerTypeLB: // nothing
+	case ServerTypeRunner: // nothing
+	case ServerTypePureRunner: // nothing
+	default:
 		// only want to activate these for full and api nodes
 		defaultDB = fmt.Sprintf("sqlite3://%s/data/fn.db", curDir)
 		defaultMQ = fmt.Sprintf("bolt://%s/data/fn.mq", curDir)
 	}
+	opts = append(opts, WithWebPort(getEnvInt(EnvPort, DefaultPort)))
+	opts = append(opts, WithGRPCPort(getEnvInt(EnvGRPCPort, DefaultGRPCPort)))
 	opts = append(opts, WithLogLevel(getEnv(EnvLogLevel, DefaultLogLevel)))
 	opts = append(opts, WithLogDest(getEnv(EnvLogDest, DefaultLogDest), getEnv(EnvLogPrefix, "")))
 	opts = append(opts, WithTracer(getEnv(EnvZipkinURL, ""))) // do this early on, so below can use these
@@ -115,6 +148,17 @@ func NewFromEnv(ctx context.Context, opts ...ServerOption) *Server {
 	opts = append(opts, WithLogURL(getEnv(EnvLOGDBURL, "")))
 	opts = append(opts, WithRunnerURL(getEnv(EnvRunnerURL, "")))
 	opts = append(opts, WithType(nodeType))
+	opts = append(opts, WithNodeCert(getEnv(EnvCert, "")))
+	opts = append(opts, WithNodeCertKey(getEnv(EnvCertKey, "")))
+	opts = append(opts, WithNodeCertAuthority(getEnv(EnvCertAuth, "")))
+
+	// Agent handling depends on node type and several other options so it must be the last processed option.
+	// Also we only need to create an agent if this is not an API node.
+	if nodeType != ServerTypeAPI {
+		opts = append(opts, WithAgentFromEnv())
+	} else {
+		opts = append(opts, WithLogstoreFromDatastore())
+	}
 	return New(ctx, opts...)
 }
 
@@ -127,6 +171,20 @@ func pwd() string {
 	return strings.Replace(cwd, "\\", "/", -1)
 }
 
+func WithWebPort(port int) ServerOption {
+	return func(ctx context.Context, s *Server) error {
+		s.webListenPort = port
+		return nil
+	}
+}
+
+func WithGRPCPort(port int) ServerOption {
+	return func(ctx context.Context, s *Server) error {
+		s.grpcListenPort = port
+		return nil
+	}
+}
+
 func WithLogLevel(ll string) ServerOption {
 	return func(ctx context.Context, s *Server) error {
 		common.SetLogLevel(ll)
@@ -200,6 +258,57 @@ func WithType(t ServerNodeType) ServerOption {
 	}
 }
 
+func WithNodeCert(cert string) ServerOption {
+	return func(ctx context.Context, s *Server) error {
+		if cert != "" {
+			abscert, err := filepath.Abs(cert)
+			if err != nil {
+				return fmt.Errorf("Unable to resolve %v: please specify a valid and readable cert file", cert)
+			}
+			_, err = os.Stat(abscert)
+			if err != nil {
+				return fmt.Errorf("Cannot stat %v: please specify a valid and readable cert file", abscert)
+			}
+			s.cert = abscert
+		}
+		return nil
+	}
+}
+
+func WithNodeCertKey(key string) ServerOption {
+	return func(ctx context.Context, s *Server) error {
+		if key != "" {
+			abskey, err := filepath.Abs(key)
+			if err != nil {
+				return fmt.Errorf("Unable to resolve %v: please specify a valid and readable cert key file", key)
+			}
+			_, err = os.Stat(abskey)
+			if err != nil {
+				return fmt.Errorf("Cannot stat %v: please specify a valid and readable cert key file", abskey)
+			}
+			s.certKey = abskey
+		}
+		return nil
+	}
+}
+
+func WithNodeCertAuthority(ca string) ServerOption {
+	return func(ctx context.Context, s *Server) error {
+		if ca != "" {
+			absca, err := filepath.Abs(ca)
+			if err != nil {
+				return fmt.Errorf("Unable to resolve %v: please specify a valid and readable cert authority file", ca)
+			}
+			_, err = os.Stat(absca)
+			if err != nil {
+				return fmt.Errorf("Cannot stat %v: please specify a valid and readable cert authority file", absca)
+			}
+			s.certAuthority = absca
+		}
+		return nil
+	}
+}
+
 func WithDatastore(ds models.Datastore) ServerOption {
 	return func(ctx context.Context, s *Server) error {
 		s.datastore = ds
@@ -228,6 +337,120 @@ func WithAgent(agent agent.Agent) ServerOption {
 	}
 }
 
+func WithLogstoreFromDatastore() ServerOption {
+	return func(ctx context.Context, s *Server) error {
+		if s.datastore == nil {
+			return errors.New("Need a datastore in order to use it as a logstore")
+		}
+		if s.logstore == nil {
+			s.logstore = s.datastore
+		}
+		return nil
+	}
+}
+
+// WithFullAgent is a shorthand for WithAgent(... create a full agent here ...)
+func WithFullAgent() ServerOption {
+	return func(ctx context.Context, s *Server) error {
+		s.nodeType = ServerTypeFull
+		if s.logstore == nil {
+			s.logstore = s.datastore
+		}
+		if s.datastore == nil || s.logstore == nil || s.mq == nil {
+			return errors.New("Full nodes must configure FN_DB_URL, FN_LOG_URL, FN_MQ_URL.")
+		}
+		s.agent = agent.New(agent.NewCachedDataAccess(agent.NewDirectDataAccess(s.datastore, s.logstore, s.mq)))
+		return nil
+	}
+}
+
+// WithAgentFromEnv must be provided as the last server option because it relies
+// on all other options being set first.
+func WithAgentFromEnv() ServerOption {
+	return func(ctx context.Context, s *Server) error {
+		switch s.nodeType {
+		case ServerTypeAPI:
+			return errors.New("Should not initialize an agent for an Fn API node.")
+		case ServerTypeRunner:
+			runnerURL := getEnv(EnvRunnerURL, "")
+			if runnerURL == "" {
+				return errors.New("No FN_RUNNER_API_URL provided for an Fn Runner node.")
+			}
+			cl, err := hybrid.NewClient(runnerURL)
+			if err != nil {
+				return err
+			}
+			s.agent = agent.New(agent.NewCachedDataAccess(cl))
+		case ServerTypePureRunner:
+			if s.datastore != nil {
+				return errors.New("Pure runner nodes must not be configured with a datastore (FN_DB_URL).")
+			}
+			if s.mq != nil {
+				return errors.New("Pure runner nodes must not be configured with a message queue (FN_MQ_URL).")
+			}
+			ds, err := hybrid.NewNopDataStore()
+			if err != nil {
+				return err
+			}
+			s.agent = agent.NewSyncOnly(agent.NewCachedDataAccess(ds))
+		case ServerTypeLB:
+			s.nodeType = ServerTypeLB
+			runnerURL := getEnv(EnvRunnerURL, "")
+			if runnerURL == "" {
+				return errors.New("No FN_RUNNER_API_URL provided for an Fn NuLB node.")
+			}
+			if s.datastore != nil {
+				return errors.New("NuLB nodes must not be configured with a datastore (FN_DB_URL).")
+			}
+			if s.mq != nil {
+				return errors.New("NuLB nodes must not be configured with a message queue (FN_MQ_URL).")
+			}
+
+			cl, err := hybrid.NewClient(runnerURL)
+			if err != nil {
+				return err
+			}
+			delegatedAgent := agent.New(agent.NewCachedDataAccess(cl))
+
+			npmAddress := getEnv(EnvNPMAddress, "")
+			runnerAddresses := getEnv(EnvRunnerAddresses, "")
+
+			var nodePool agent.NodePool
+			if npmAddress != "" {
+				// TODO refactor DefaultgRPCNodePool as an extension
+				nodePool = agent_grpc.DefaultgRPCNodePool(npmAddress, s.cert, s.certKey, s.certAuthority)
+			} else if runnerAddresses != "" {
+				nodePool = agent_grpc.DefaultStaticNodePool(strings.Split(runnerAddresses, ","))
+			} else {
+				return errors.New("Must provide either FN_NPM_ADDRESS or FN_RUNNER_ADDRESSES for an Fn NuLB node.")
+			}
+
+			// Select the placement algorithm
+			var placer agent.Placer
+			switch getEnv(EnvLBPlacementAlg, "") {
+			case "ch":
+				placer = agent.NewCHPlacer()
+			default:
+				placer = agent.NewNaivePlacer()
+			}
+			s.agent, err = agent.NewLBAgent(delegatedAgent, nodePool, placer)
+			if err != nil {
+				return errors.New("LBAgent creation failed")
+			}
+		default:
+			s.nodeType = ServerTypeFull
+			if s.logstore == nil { // TODO seems weird?
+				s.logstore = s.datastore
+			}
+			if s.datastore == nil || s.logstore == nil || s.mq == nil {
+				return errors.New("Full nodes must configure FN_DB_URL, FN_LOG_URL, FN_MQ_URL.")
+			}
+			s.agent = agent.New(agent.NewCachedDataAccess(agent.NewDirectDataAccess(s.datastore, s.logstore, s.mq)))
+		}
+		return nil
+	}
+}
+
 // New creates a new Functions server with the opts given. For convenience, users may
 // prefer to use NewFromEnv but New is more flexible if needed.
 func New(ctx context.Context, opts ...ServerOption) *Server {
@@ -237,6 +460,9 @@ func New(ctx context.Context, opts ...ServerOption) *Server {
 	log := common.Logger(ctx)
 	s := &Server{
 		Router: gin.New(),
+		// Add default ports
+		webListenPort:  DefaultPort,
+		grpcListenPort: DefaultGRPCPort,
 		// Almost everything else is configured through opts (see NewFromEnv for ex.) or below
 	}
 
@@ -250,33 +476,15 @@ func New(ctx context.Context, opts ...ServerOption) *Server {
 		}
 	}
 
-	if s.logstore == nil { // TODO seems weird?
-		s.logstore = s.datastore
-	}
-
-	// TODO we maybe should use the agent.DirectDataAccess in the /runner endpoints server side?
-
+	// Check that WithAgent options have been processed correctly.
 	switch s.nodeType {
 	case ServerTypeAPI:
 		if s.agent != nil {
-			log.Info("shutting down agent configured for api node")
-			s.agent.Close()
-		}
-		s.agent = nil
-	case ServerTypeRunner:
-		if s.agent == nil {
-			log.Fatal("No agent started for a runner node, add FN_RUNNER_API_URL to configuration.")
+			log.Fatal("Incorrect configuration, API nodes must not have an agent initialized.")
 		}
 	default:
-		s.nodeType = ServerTypeFull
-		if s.datastore == nil || s.logstore == nil || s.mq == nil {
-			log.Fatal("Full nodes must configure FN_DB_URL, FN_LOG_URL, FN_MQ_URL.")
-		}
-
-		// TODO force caller to use WithAgent option? at this point we need to use the ds/ls/mq configured after all opts are run.
 		if s.agent == nil {
-			// TODO for tests we don't want cache, really, if we force WithAgent this can be fixed. cache needs to be moved anyway so that runner nodes can use it...
-			s.agent = agent.New(agent.NewCachedDataAccess(agent.NewDirectDataAccess(s.datastore, s.logstore, s.mq)))
+			log.Fatal("Incorrect configuration, non-API nodes must have an agent initialized.")
 		}
 	}
 
@@ -377,7 +585,7 @@ func (s *Server) Start(ctx context.Context) {
 func (s *Server) startGears(ctx context.Context, cancel context.CancelFunc) {
 	// By default it serves on :8080 unless a
 	// FN_PORT environment variable was defined.
-	listen := fmt.Sprintf(":%d", getEnvInt(EnvPort, DefaultPort))
+	listen := fmt.Sprintf(":%d", s.webListenPort)
 
 	const runHeader = `
         ______
@@ -392,6 +600,24 @@ func (s *Server) startGears(ctx context.Context, cancel context.CancelFunc) {
 
 	installChildReaper()
 
+	if s.nodeType == ServerTypePureRunner {
+		// Run grpc too
+		grpcAddr := fmt.Sprintf(":%d", s.grpcListenPort)
+		pr, err := agent.CreatePureRunner(grpcAddr, s.agent, s.cert, s.certKey, s.certAuthority)
+		if err != nil {
+			logrus.WithError(err).Error("Pure runner server creation error")
+			cancel()
+		} else {
+			go func() {
+				err := pr.Start()
+				if err != nil {
+					logrus.WithError(err).Error("fail to start pure runner")
+					cancel()
+				}
+			}()
+		}
+	}
+
 	server := http.Server{
 		Addr:    listen,
 		Handler: &ochttp.Handler{Handler: s.Router},
@@ -418,7 +644,10 @@ func (s *Server) startGears(ctx context.Context, cancel context.CancelFunc) {
 	}
 
 	if s.agent != nil {
-		s.agent.Close() // after we stop taking requests, wait for all tasks to finish
+		err := s.agent.Close() // after we stop taking requests, wait for all tasks to finish
+		if err != nil {
+			logrus.WithError(err).Error("Fail to close the agent")
+		}
 	}
 }
 
@@ -437,51 +666,54 @@ func (s *Server) bindHandlers(ctx context.Context) {
 
 	profilerSetup(engine, "/debug")
 
-	if s.nodeType != ServerTypeRunner {
-		v1 := engine.Group("/v1")
-		v1.Use(setAppNameInCtx)
-		v1.Use(s.apiMiddlewareWrapper())
-		v1.GET("/apps", s.handleAppList)
-		v1.POST("/apps", s.handleAppCreate)
+	// Pure runners don't have any route, they have grpc
+	if s.nodeType != ServerTypePureRunner {
+		if s.nodeType != ServerTypeRunner {
+			v1 := engine.Group("/v1")
+			v1.Use(setAppNameInCtx)
+			v1.Use(s.apiMiddlewareWrapper())
+			v1.GET("/apps", s.handleAppList)
+			v1.POST("/apps", s.handleAppCreate)
 
-		{
-			apps := v1.Group("/apps/:app")
-			apps.Use(appNameCheck)
+			{
+				apps := v1.Group("/apps/:app")
+				apps.Use(appNameCheck)
 
-			apps.GET("", s.handleAppGet)
-			apps.PATCH("", s.handleAppUpdate)
-			apps.DELETE("", s.handleAppDelete)
+				apps.GET("", s.handleAppGet)
+				apps.PATCH("", s.handleAppUpdate)
+				apps.DELETE("", s.handleAppDelete)
 
-			apps.GET("/routes", s.handleRouteList)
-			apps.POST("/routes", s.handleRoutesPostPutPatch)
-			apps.GET("/routes/:route", s.handleRouteGet)
-			apps.PATCH("/routes/*route", s.handleRoutesPostPutPatch)
-			apps.PUT("/routes/*route", s.handleRoutesPostPutPatch)
-			apps.DELETE("/routes/*route", s.handleRouteDelete)
+				apps.GET("/routes", s.handleRouteList)
+				apps.POST("/routes", s.handleRoutesPostPutPatch)
+				apps.GET("/routes/:route", s.handleRouteGet)
+				apps.PATCH("/routes/*route", s.handleRoutesPostPutPatch)
+				apps.PUT("/routes/*route", s.handleRoutesPostPutPatch)
+				apps.DELETE("/routes/*route", s.handleRouteDelete)
 
-			apps.GET("/calls", s.handleCallList)
+				apps.GET("/calls", s.handleCallList)
 
-			apps.GET("/calls/:call", s.handleCallGet)
-			apps.GET("/calls/:call/log", s.handleCallLogGet)
+				apps.GET("/calls/:call", s.handleCallGet)
+				apps.GET("/calls/:call/log", s.handleCallLogGet)
+			}
+
+			{
+				runner := v1.Group("/runner")
+				runner.PUT("/async", s.handleRunnerEnqueue)
+				runner.GET("/async", s.handleRunnerDequeue)
+
+				runner.POST("/start", s.handleRunnerStart)
+				runner.POST("/finish", s.handleRunnerFinish)
+			}
 		}
 
-		{
-			runner := v1.Group("/runner")
-			runner.PUT("/async", s.handleRunnerEnqueue)
-			runner.GET("/async", s.handleRunnerDequeue)
-
-			runner.POST("/start", s.handleRunnerStart)
-			runner.POST("/finish", s.handleRunnerFinish)
+		if s.nodeType != ServerTypeAPI {
+			runner := engine.Group("/r")
+			runner.Use(appNameCheck)
+			runner.Any("/:app", s.handleFunctionCall)
+			runner.Any("/:app/*route", s.handleFunctionCall)
 		}
 	}
 
-	if s.nodeType != ServerTypeAPI {
-		runner := engine.Group("/r")
-		runner.Use(appNameCheck)
-		runner.Any("/:app", s.handleFunctionCall)
-		runner.Any("/:app/*route", s.handleFunctionCall)
-	}
-
 	engine.NoRoute(func(c *gin.Context) {
 		var err error
 		switch {

api/server/server_test.go

@@ -220,10 +220,7 @@ func TestApiNode(t *testing.T) {
 	ds, logDB, close := prepareDB(ctx, t)
 	defer close()
 
-	rnr, rnrcancel := testRunner(t, ds)
-	defer rnrcancel()
-
-	srv := testServer(ds, &mqs.Mock{}, logDB, rnr, ServerTypeAPI)
+	srv := testServer(ds, &mqs.Mock{}, logDB, nil, ServerTypeAPI)
 
 	for _, test := range []struct {
 		name              string