open-consul/agent/local/state_test.go
Matt Keeler 2c7844d220
Implement Client Agent Auto Config
There are a couple of things in here.

First, just like auto encrypt, any Cluster.AutoConfig RPC will implicitly use the less secure RPC mechanism.

This drastically modifies how the Consul Agent starts up and moves most of the responsibilities (other than signal handling) from the cli command and into the Agent.
2020-06-17 16:49:46 -04:00

2145 lines
55 KiB
Go

package local_test
import (
"errors"
"fmt"
"os"
"testing"
"time"
"github.com/hashicorp/consul/testrpc"
"github.com/hashicorp/go-hclog"
"github.com/hashicorp/consul/agent"
"github.com/hashicorp/consul/agent/config"
"github.com/hashicorp/consul/agent/local"
"github.com/hashicorp/consul/agent/structs"
"github.com/hashicorp/consul/agent/token"
"github.com/hashicorp/consul/api"
"github.com/hashicorp/consul/sdk/testutil/retry"
"github.com/hashicorp/consul/types"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func unNilMap(in map[string]string) map[string]string {
if in == nil {
return make(map[string]string)
}
return in
}
func TestAgentAntiEntropy_Services(t *testing.T) {
t.Parallel()
a := agent.NewTestAgent(t, "")
defer a.Shutdown()
testrpc.WaitForTestAgent(t, a.RPC, "dc1")
// Register info
args := &structs.RegisterRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
Address: "127.0.0.1",
}
// Exists both, same (noop)
var out struct{}
srv1 := &structs.NodeService{
ID: "mysql",
Service: "mysql",
Tags: []string{"master"},
Port: 5000,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
assert.False(t, a.State.ServiceExists(structs.ServiceID{ID: srv1.ID}))
a.State.AddService(srv1, "")
assert.True(t, a.State.ServiceExists(structs.ServiceID{ID: srv1.ID}))
args.Service = srv1
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
// Exists both, different (update)
srv2 := &structs.NodeService{
ID: "redis",
Service: "redis",
Tags: []string{},
Port: 8000,
Weights: &structs.Weights{
Passing: 1,
Warning: 0,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddService(srv2, "")
srv2_mod := new(structs.NodeService)
*srv2_mod = *srv2
srv2_mod.Port = 9000
args.Service = srv2_mod
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
// Exists local (create)
srv3 := &structs.NodeService{
ID: "web",
Service: "web",
Tags: []string{},
Port: 80,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddService(srv3, "")
// Exists remote (delete)
srv4 := &structs.NodeService{
ID: "lb",
Service: "lb",
Tags: []string{},
Port: 443,
Weights: &structs.Weights{
Passing: 1,
Warning: 0,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
args.Service = srv4
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
// Exists both, different address (update)
srv5 := &structs.NodeService{
ID: "api",
Service: "api",
Tags: []string{},
Address: "127.0.0.10",
Port: 8000,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddService(srv5, "")
srv5_mod := new(structs.NodeService)
*srv5_mod = *srv5
srv5_mod.Address = "127.0.0.1"
args.Service = srv5_mod
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
// Exists local, in sync, remote missing (create)
srv6 := &structs.NodeService{
ID: "cache",
Service: "cache",
Tags: []string{},
Port: 11211,
Weights: &structs.Weights{
Passing: 1,
Warning: 0,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.SetServiceState(&local.ServiceState{
Service: srv6,
InSync: true,
})
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
var services structs.IndexedNodeServices
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
t.Fatalf("err: %v", err)
}
// Make sure we sent along our node info when we synced.
id := services.NodeServices.Node.ID
addrs := services.NodeServices.Node.TaggedAddresses
meta := services.NodeServices.Node.Meta
delete(meta, structs.MetaSegmentKey) // Added later, not in config.
assert.Equal(t, a.Config.NodeID, id)
assert.Equal(t, a.Config.TaggedAddresses, addrs)
assert.Equal(t, unNilMap(a.Config.NodeMeta), meta)
// We should have 6 services (consul included)
if len(services.NodeServices.Services) != 6 {
t.Fatalf("bad: %v", services.NodeServices.Services)
}
// All the services should match
for id, serv := range services.NodeServices.Services {
serv.CreateIndex, serv.ModifyIndex = 0, 0
switch id {
case "mysql":
require.Equal(t, srv1, serv)
case "redis":
require.Equal(t, srv2, serv)
case "web":
require.Equal(t, srv3, serv)
case "api":
require.Equal(t, srv5, serv)
case "cache":
require.Equal(t, srv6, serv)
case structs.ConsulServiceID:
// ignore
default:
t.Fatalf("unexpected service: %v", id)
}
}
if err := servicesInSync(a.State, 5, structs.DefaultEnterpriseMeta()); err != nil {
t.Fatal(err)
}
// Remove one of the services
a.State.RemoveService(structs.NewServiceID("api", nil))
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
t.Fatalf("err: %v", err)
}
// We should have 5 services (consul included)
if len(services.NodeServices.Services) != 5 {
t.Fatalf("bad: %v", services.NodeServices.Services)
}
// All the services should match
for id, serv := range services.NodeServices.Services {
serv.CreateIndex, serv.ModifyIndex = 0, 0
switch id {
case "mysql":
require.Equal(t, srv1, serv)
case "redis":
require.Equal(t, srv2, serv)
case "web":
require.Equal(t, srv3, serv)
case "cache":
require.Equal(t, srv6, serv)
case structs.ConsulServiceID:
// ignore
default:
t.Fatalf("unexpected service: %v", id)
}
}
if err := servicesInSync(a.State, 4, structs.DefaultEnterpriseMeta()); err != nil {
t.Fatal(err)
}
}
func TestAgentAntiEntropy_Services_ConnectProxy(t *testing.T) {
t.Parallel()
assert := assert.New(t)
a := agent.NewTestAgent(t, "")
defer a.Shutdown()
testrpc.WaitForTestAgent(t, a.RPC, "dc1")
// Register node info
var out struct{}
args := &structs.RegisterRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
Address: "127.0.0.1",
}
// Exists both same (noop)
srv1 := &structs.NodeService{
Kind: structs.ServiceKindConnectProxy,
ID: "mysql-proxy",
Service: "mysql-proxy",
Port: 5000,
Proxy: structs.ConnectProxyConfig{DestinationServiceName: "db"},
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddService(srv1, "")
args.Service = srv1
assert.Nil(a.RPC("Catalog.Register", args, &out))
// Exists both, different (update)
srv2 := &structs.NodeService{
ID: "redis-proxy",
Service: "redis-proxy",
Port: 8000,
Kind: structs.ServiceKindConnectProxy,
Proxy: structs.ConnectProxyConfig{DestinationServiceName: "redis"},
Weights: &structs.Weights{
Passing: 1,
Warning: 0,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddService(srv2, "")
srv2_mod := new(structs.NodeService)
*srv2_mod = *srv2
srv2_mod.Port = 9000
args.Service = srv2_mod
assert.Nil(a.RPC("Catalog.Register", args, &out))
// Exists local (create)
srv3 := &structs.NodeService{
ID: "web-proxy",
Service: "web-proxy",
Port: 80,
Kind: structs.ServiceKindConnectProxy,
Proxy: structs.ConnectProxyConfig{DestinationServiceName: "web"},
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddService(srv3, "")
// Exists remote (delete)
srv4 := &structs.NodeService{
ID: "lb-proxy",
Service: "lb-proxy",
Port: 443,
Kind: structs.ServiceKindConnectProxy,
Proxy: structs.ConnectProxyConfig{DestinationServiceName: "lb"},
Weights: &structs.Weights{
Passing: 1,
Warning: 0,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
args.Service = srv4
assert.Nil(a.RPC("Catalog.Register", args, &out))
// Exists local, in sync, remote missing (create)
srv5 := &structs.NodeService{
ID: "cache-proxy",
Service: "cache-proxy",
Port: 11211,
Kind: structs.ServiceKindConnectProxy,
Proxy: structs.ConnectProxyConfig{DestinationServiceName: "cache-proxy"},
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.SetServiceState(&local.ServiceState{
Service: srv5,
InSync: true,
})
assert.Nil(a.State.SyncFull())
var services structs.IndexedNodeServices
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
assert.Nil(a.RPC("Catalog.NodeServices", &req, &services))
// We should have 5 services (consul included)
assert.Len(services.NodeServices.Services, 5)
// All the services should match
for id, serv := range services.NodeServices.Services {
serv.CreateIndex, serv.ModifyIndex = 0, 0
switch id {
case "mysql-proxy":
assert.Equal(srv1, serv)
case "redis-proxy":
assert.Equal(srv2, serv)
case "web-proxy":
assert.Equal(srv3, serv)
case "cache-proxy":
assert.Equal(srv5, serv)
case structs.ConsulServiceID:
// ignore
default:
t.Fatalf("unexpected service: %v", id)
}
}
assert.Nil(servicesInSync(a.State, 4, structs.DefaultEnterpriseMeta()))
// Remove one of the services
a.State.RemoveService(structs.NewServiceID("cache-proxy", nil))
assert.Nil(a.State.SyncFull())
assert.Nil(a.RPC("Catalog.NodeServices", &req, &services))
// We should have 4 services (consul included)
assert.Len(services.NodeServices.Services, 4)
// All the services should match
for id, serv := range services.NodeServices.Services {
serv.CreateIndex, serv.ModifyIndex = 0, 0
switch id {
case "mysql-proxy":
assert.Equal(srv1, serv)
case "redis-proxy":
assert.Equal(srv2, serv)
case "web-proxy":
assert.Equal(srv3, serv)
case structs.ConsulServiceID:
// ignore
default:
t.Fatalf("unexpected service: %v", id)
}
}
assert.Nil(servicesInSync(a.State, 3, structs.DefaultEnterpriseMeta()))
}
func TestAgent_ServiceWatchCh(t *testing.T) {
t.Parallel()
a := agent.NewTestAgent(t, "")
defer a.Shutdown()
testrpc.WaitForTestAgent(t, a.RPC, "dc1")
require := require.New(t)
// register a local service
srv1 := &structs.NodeService{
ID: "svc_id1",
Service: "svc1",
Tags: []string{"tag1"},
Port: 6100,
}
require.NoError(a.State.AddService(srv1, ""))
verifyState := func(ss *local.ServiceState) {
require.NotNil(ss)
require.NotNil(ss.WatchCh)
// Sanity check WatchCh blocks
select {
case <-ss.WatchCh:
t.Fatal("should block until service changes")
default:
}
}
// Should be able to get a ServiceState
ss := a.State.ServiceState(srv1.CompoundServiceID())
verifyState(ss)
// Update service in another go routine
go func() {
srv2 := srv1
srv2.Port = 6200
require.NoError(a.State.AddService(srv2, ""))
}()
// We should observe WatchCh close
select {
case <-ss.WatchCh:
// OK!
case <-time.After(500 * time.Millisecond):
t.Fatal("timeout waiting for WatchCh to close")
}
// Should also fire for state being set explicitly
ss = a.State.ServiceState(srv1.CompoundServiceID())
verifyState(ss)
go func() {
a.State.SetServiceState(&local.ServiceState{
Service: ss.Service,
Token: "foo",
})
}()
// We should observe WatchCh close
select {
case <-ss.WatchCh:
// OK!
case <-time.After(500 * time.Millisecond):
t.Fatal("timeout waiting for WatchCh to close")
}
// Should also fire for service being removed
ss = a.State.ServiceState(srv1.CompoundServiceID())
verifyState(ss)
go func() {
require.NoError(a.State.RemoveService(srv1.CompoundServiceID()))
}()
// We should observe WatchCh close
select {
case <-ss.WatchCh:
// OK!
case <-time.After(500 * time.Millisecond):
t.Fatal("timeout waiting for WatchCh to close")
}
}
func TestAgentAntiEntropy_EnableTagOverride(t *testing.T) {
t.Parallel()
a := agent.NewTestAgent(t, "")
defer a.Shutdown()
testrpc.WaitForTestAgent(t, a.RPC, "dc1")
args := &structs.RegisterRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
Address: "127.0.0.1",
}
var out struct{}
// register a local service with tag override enabled
srv1 := &structs.NodeService{
ID: "svc_id1",
Service: "svc1",
Tags: []string{"tag1"},
Port: 6100,
EnableTagOverride: true,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
}
a.State.AddService(srv1, "")
// register a local service with tag override disabled
srv2 := &structs.NodeService{
ID: "svc_id2",
Service: "svc2",
Tags: []string{"tag2"},
Port: 6200,
EnableTagOverride: false,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
}
a.State.AddService(srv2, "")
// make sure they are both in the catalog
if err := a.State.SyncChanges(); err != nil {
t.Fatalf("err: %v", err)
}
// update the services in the catalog and change the tags and port.
// Only tag changes should be propagated for services where tag
// override is enabled.
args.Service = &structs.NodeService{
ID: srv1.ID,
Service: srv1.Service,
Tags: []string{"tag1_mod"},
Port: 7100,
EnableTagOverride: true,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
}
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
args.Service = &structs.NodeService{
ID: srv2.ID,
Service: srv2.Service,
Tags: []string{"tag2_mod"},
Port: 7200,
EnableTagOverride: false,
Weights: &structs.Weights{
Passing: 1,
Warning: 0,
},
}
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
// sync catalog and local state
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
var services structs.IndexedNodeServices
retry.Run(t, func(r *retry.R) {
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
r.Fatalf("err: %v", err)
}
// All the services should match
for id, serv := range services.NodeServices.Services {
serv.CreateIndex, serv.ModifyIndex = 0, 0
switch id {
case "svc_id1":
// tags should be modified but not the port
got := serv
want := &structs.NodeService{
ID: "svc_id1",
Service: "svc1",
Tags: []string{"tag1_mod"},
Port: 6100,
EnableTagOverride: true,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
assert.Equal(r, want, got)
case "svc_id2":
got, want := serv, srv2
assert.Equal(r, want, got)
case structs.ConsulServiceID:
// ignore
default:
r.Fatalf("unexpected service: %v", id)
}
}
if err := servicesInSync(a.State, 2, structs.DefaultEnterpriseMeta()); err != nil {
r.Fatal(err)
}
})
}
func TestAgentAntiEntropy_Services_WithChecks(t *testing.T) {
t.Parallel()
a := agent.NewTestAgent(t, "")
defer a.Shutdown()
testrpc.WaitForTestAgent(t, a.RPC, "dc1")
{
// Single check
srv := &structs.NodeService{
ID: "mysql",
Service: "mysql",
Tags: []string{"master"},
Port: 5000,
}
a.State.AddService(srv, "")
chk := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "mysql",
Name: "mysql",
ServiceID: "mysql",
Status: api.HealthPassing,
}
a.State.AddCheck(chk, "")
if err := a.State.SyncFull(); err != nil {
t.Fatal("sync failed: ", err)
}
// We should have 2 services (consul included)
svcReq := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
var services structs.IndexedNodeServices
if err := a.RPC("Catalog.NodeServices", &svcReq, &services); err != nil {
t.Fatalf("err: %v", err)
}
if len(services.NodeServices.Services) != 2 {
t.Fatalf("bad: %v", services.NodeServices.Services)
}
// We should have one health check
chkReq := structs.ServiceSpecificRequest{
Datacenter: "dc1",
ServiceName: "mysql",
}
var checks structs.IndexedHealthChecks
if err := a.RPC("Health.ServiceChecks", &chkReq, &checks); err != nil {
t.Fatalf("err: %v", err)
}
if len(checks.HealthChecks) != 1 {
t.Fatalf("bad: %v", checks)
}
}
{
// Multiple checks
srv := &structs.NodeService{
ID: "redis",
Service: "redis",
Tags: []string{"master"},
Port: 5000,
}
a.State.AddService(srv, "")
chk1 := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "redis:1",
Name: "redis:1",
ServiceID: "redis",
Status: api.HealthPassing,
}
a.State.AddCheck(chk1, "")
chk2 := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "redis:2",
Name: "redis:2",
ServiceID: "redis",
Status: api.HealthPassing,
}
a.State.AddCheck(chk2, "")
if err := a.State.SyncFull(); err != nil {
t.Fatal("sync failed: ", err)
}
// We should have 3 services (consul included)
svcReq := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
var services structs.IndexedNodeServices
if err := a.RPC("Catalog.NodeServices", &svcReq, &services); err != nil {
t.Fatalf("err: %v", err)
}
if len(services.NodeServices.Services) != 3 {
t.Fatalf("bad: %v", services.NodeServices.Services)
}
// We should have two health checks
chkReq := structs.ServiceSpecificRequest{
Datacenter: "dc1",
ServiceName: "redis",
}
var checks structs.IndexedHealthChecks
if err := a.RPC("Health.ServiceChecks", &chkReq, &checks); err != nil {
t.Fatalf("err: %v", err)
}
if len(checks.HealthChecks) != 2 {
t.Fatalf("bad: %v", checks)
}
}
}
var testRegisterRules = `
node "" {
policy = "write"
}
service "api" {
policy = "write"
}
service "consul" {
policy = "write"
}
`
func TestAgentAntiEntropy_Services_ACLDeny(t *testing.T) {
t.Parallel()
a := agent.NewTestAgent(t, `
acl_datacenter = "dc1"
acl_master_token = "root"
acl_default_policy = "deny" `)
defer a.Shutdown()
testrpc.WaitForLeader(t, a.RPC, "dc1")
// Create the ACL
arg := structs.ACLRequest{
Datacenter: "dc1",
Op: structs.ACLSet,
ACL: structs.ACL{
Name: "User token",
Type: structs.ACLTokenTypeClient,
Rules: testRegisterRules,
},
WriteRequest: structs.WriteRequest{
Token: "root",
},
}
var token string
if err := a.RPC("ACL.Apply", &arg, &token); err != nil {
t.Fatalf("err: %v", err)
}
// Create service (disallowed)
srv1 := &structs.NodeService{
ID: "mysql",
Service: "mysql",
Tags: []string{"master"},
Port: 5000,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
}
a.State.AddService(srv1, token)
// Create service (allowed)
srv2 := &structs.NodeService{
ID: "api",
Service: "api",
Tags: []string{"foo"},
Port: 5001,
Weights: &structs.Weights{
Passing: 1,
Warning: 0,
},
}
a.State.AddService(srv2, token)
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
// Verify that we are in sync
{
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
QueryOptions: structs.QueryOptions{
Token: "root",
},
}
var services structs.IndexedNodeServices
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
t.Fatalf("err: %v", err)
}
// We should have 2 services (consul included)
if len(services.NodeServices.Services) != 2 {
t.Fatalf("bad: %v", services.NodeServices.Services)
}
// All the services should match
for id, serv := range services.NodeServices.Services {
serv.CreateIndex, serv.ModifyIndex = 0, 0
switch id {
case "mysql":
t.Fatalf("should not be permitted")
case "api":
require.Equal(t, srv2, serv)
case structs.ConsulServiceID:
// ignore
default:
t.Fatalf("unexpected service: %v", id)
}
}
if err := servicesInSync(a.State, 2, structs.DefaultEnterpriseMeta()); err != nil {
t.Fatal(err)
}
}
// Now remove the service and re-sync
a.State.RemoveService(structs.NewServiceID("api", nil))
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
// Verify that we are in sync
{
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
QueryOptions: structs.QueryOptions{
Token: "root",
},
}
var services structs.IndexedNodeServices
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
t.Fatalf("err: %v", err)
}
// We should have 1 service (just consul)
if len(services.NodeServices.Services) != 1 {
t.Fatalf("bad: %v", services.NodeServices.Services)
}
// All the services should match
for id, serv := range services.NodeServices.Services {
serv.CreateIndex, serv.ModifyIndex = 0, 0
switch id {
case "mysql":
t.Fatalf("should not be permitted")
case "api":
t.Fatalf("should be deleted")
case structs.ConsulServiceID:
// ignore
default:
t.Fatalf("unexpected service: %v", id)
}
}
if err := servicesInSync(a.State, 1, structs.DefaultEnterpriseMeta()); err != nil {
t.Fatal(err)
}
}
// Make sure the token got cleaned up.
if token := a.State.ServiceToken(structs.NewServiceID("api", nil)); token != "" {
t.Fatalf("bad: %s", token)
}
}
func TestAgentAntiEntropy_Checks(t *testing.T) {
t.Parallel()
a := agent.NewTestAgent(t, "")
defer a.Shutdown()
testrpc.WaitForTestAgent(t, a.RPC, "dc1")
// Register info
args := &structs.RegisterRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
Address: "127.0.0.1",
}
// Exists both, same (noop)
var out struct{}
chk1 := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "mysql",
Name: "mysql",
Status: api.HealthPassing,
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddCheck(chk1, "")
args.Check = chk1
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
// Exists both, different (update)
chk2 := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "redis",
Name: "redis",
Status: api.HealthPassing,
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddCheck(chk2, "")
chk2_mod := new(structs.HealthCheck)
*chk2_mod = *chk2
chk2_mod.Status = api.HealthCritical
args.Check = chk2_mod
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
// Exists local (create)
chk3 := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "web",
Name: "web",
Status: api.HealthPassing,
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddCheck(chk3, "")
// Exists remote (delete)
chk4 := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "lb",
Name: "lb",
Status: api.HealthPassing,
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
args.Check = chk4
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
// Exists local, in sync, remote missing (create)
chk5 := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "cache",
Name: "cache",
Status: api.HealthPassing,
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.SetCheckState(&local.CheckState{
Check: chk5,
InSync: true,
})
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
var checks structs.IndexedHealthChecks
retry.Run(t, func(r *retry.R) {
// Verify that we are in sync
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
r.Fatalf("err: %v", err)
}
// We should have 5 checks (serf included)
if len(checks.HealthChecks) != 5 {
r.Fatalf("bad: %v", checks)
}
// All the checks should match
for _, chk := range checks.HealthChecks {
chk.CreateIndex, chk.ModifyIndex = 0, 0
switch chk.CheckID {
case "mysql":
require.Equal(t, chk, chk1)
case "redis":
require.Equal(t, chk, chk2)
case "web":
require.Equal(t, chk, chk3)
case "cache":
require.Equal(t, chk, chk5)
case "serfHealth":
// ignore
default:
r.Fatalf("unexpected check: %v", chk)
}
}
if err := checksInSync(a.State, 4, structs.DefaultEnterpriseMeta()); err != nil {
r.Fatal(err)
}
})
retry.Run(t, func(r *retry.R) {
// Make sure we sent along our node info addresses when we synced.
{
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
var services structs.IndexedNodeServices
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
r.Fatalf("err: %v", err)
}
id := services.NodeServices.Node.ID
addrs := services.NodeServices.Node.TaggedAddresses
meta := services.NodeServices.Node.Meta
delete(meta, structs.MetaSegmentKey) // Added later, not in config.
assert.Equal(t, a.Config.NodeID, id)
assert.Equal(t, a.Config.TaggedAddresses, addrs)
assert.Equal(t, unNilMap(a.Config.NodeMeta), meta)
}
})
retry.Run(t, func(r *retry.R) {
// Remove one of the checks
a.State.RemoveCheck(structs.NewCheckID("redis", nil))
if err := a.State.SyncFull(); err != nil {
r.Fatalf("err: %v", err)
}
// Verify that we are in sync
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
r.Fatalf("err: %v", err)
}
// We should have 5 checks (serf included)
if len(checks.HealthChecks) != 4 {
r.Fatalf("bad: %v", checks)
}
// All the checks should match
for _, chk := range checks.HealthChecks {
chk.CreateIndex, chk.ModifyIndex = 0, 0
switch chk.CheckID {
case "mysql":
require.Equal(t, chk1, chk)
case "web":
require.Equal(t, chk3, chk)
case "cache":
require.Equal(t, chk5, chk)
case "serfHealth":
// ignore
default:
r.Fatalf("unexpected check: %v", chk)
}
}
if err := checksInSync(a.State, 3, structs.DefaultEnterpriseMeta()); err != nil {
r.Fatal(err)
}
})
}
func TestAgentAntiEntropy_RemovingServiceAndCheck(t *testing.T) {
t.Parallel()
a := agent.NewTestAgent(t, "")
defer a.Shutdown()
testrpc.WaitForTestAgent(t, a.RPC, "dc1")
// Register info
args := &structs.RegisterRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
Address: "127.0.0.1",
}
var out struct{}
// Exists remote (delete)
svcID := "deleted-check-service"
srv := &structs.NodeService{
ID: svcID,
Service: "echo",
Tags: []string{},
Address: "127.0.0.1",
Port: 8080,
}
args.Service = srv
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
// Exists remote (delete)
chk := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "lb",
Name: "lb",
ServiceID: svcID,
Status: api.HealthPassing,
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
args.Check = chk
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
var services structs.IndexedNodeServices
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
t.Fatalf("err: %v", err)
}
// The consul service will still be registered
if len(services.NodeServices.Services) != 1 {
t.Fatalf("Expected all services to be deleted, got: %#v", services.NodeServices.Services)
}
var checks structs.IndexedHealthChecks
// Verify that we are in sync
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
t.Fatalf("err: %v", err)
}
// The serfHealth check will still be here
if len(checks.HealthChecks) != 1 {
t.Fatalf("Expected the health check to be deleted, got: %#v", checks.HealthChecks)
}
}
func TestAgentAntiEntropy_Checks_ACLDeny(t *testing.T) {
t.Parallel()
dc := "dc1"
a := &agent.TestAgent{HCL: `
acl_datacenter = "` + dc + `"
acl_master_token = "root"
acl_default_policy = "deny" `}
if err := a.Start(t); err != nil {
t.Fatal(err)
}
defer a.Shutdown()
testrpc.WaitForLeader(t, a.RPC, dc)
// Create the ACL
arg := structs.ACLRequest{
Datacenter: dc,
Op: structs.ACLSet,
ACL: structs.ACL{
Name: "User token",
Type: structs.ACLTokenTypeClient,
Rules: testRegisterRules,
},
WriteRequest: structs.WriteRequest{
Token: "root",
},
}
var token string
if err := a.RPC("ACL.Apply", &arg, &token); err != nil {
t.Fatalf("err: %v", err)
}
// Create services using the root token
srv1 := &structs.NodeService{
ID: "mysql",
Service: "mysql",
Tags: []string{"master"},
Port: 5000,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddService(srv1, "root")
srv2 := &structs.NodeService{
ID: "api",
Service: "api",
Tags: []string{"foo"},
Port: 5001,
Weights: &structs.Weights{
Passing: 1,
Warning: 1,
},
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddService(srv2, "root")
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
// Verify that we are in sync
{
req := structs.NodeSpecificRequest{
Datacenter: dc,
Node: a.Config.NodeName,
QueryOptions: structs.QueryOptions{
Token: "root",
},
}
var services structs.IndexedNodeServices
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
t.Fatalf("err: %v", err)
}
// We should have 3 services (consul included)
if len(services.NodeServices.Services) != 3 {
t.Fatalf("bad: %v", services.NodeServices.Services)
}
// All the services should match
for id, serv := range services.NodeServices.Services {
serv.CreateIndex, serv.ModifyIndex = 0, 0
switch id {
case "mysql":
require.Equal(t, srv1, serv)
case "api":
require.Equal(t, srv2, serv)
case structs.ConsulServiceID:
// ignore
default:
t.Fatalf("unexpected service: %v", id)
}
}
if err := servicesInSync(a.State, 2, structs.DefaultEnterpriseMeta()); err != nil {
t.Fatal(err)
}
}
// This check won't be allowed.
chk1 := &structs.HealthCheck{
Node: a.Config.NodeName,
ServiceID: "mysql",
ServiceName: "mysql",
ServiceTags: []string{"master"},
CheckID: "mysql-check",
Name: "mysql",
Status: api.HealthPassing,
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddCheck(chk1, token)
// This one will be allowed.
chk2 := &structs.HealthCheck{
Node: a.Config.NodeName,
ServiceID: "api",
ServiceName: "api",
ServiceTags: []string{"foo"},
CheckID: "api-check",
Name: "api",
Status: api.HealthPassing,
EnterpriseMeta: *structs.DefaultEnterpriseMeta(),
}
a.State.AddCheck(chk2, token)
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
// Verify that we are in sync
req := structs.NodeSpecificRequest{
Datacenter: dc,
Node: a.Config.NodeName,
QueryOptions: structs.QueryOptions{
Token: "root",
},
}
var checks structs.IndexedHealthChecks
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
t.Fatalf("err: %v", err)
}
// We should have 2 checks (serf included)
if len(checks.HealthChecks) != 2 {
t.Fatalf("bad: %v", checks)
}
// All the checks should match
for _, chk := range checks.HealthChecks {
chk.CreateIndex, chk.ModifyIndex = 0, 0
switch chk.CheckID {
case "mysql-check":
t.Fatalf("should not be permitted")
case "api-check":
require.Equal(t, chk, chk2)
case "serfHealth":
// ignore
default:
t.Fatalf("unexpected check: %v", chk)
}
}
if err := checksInSync(a.State, 2, structs.DefaultEnterpriseMeta()); err != nil {
t.Fatal(err)
}
// Now delete the check and wait for sync.
a.State.RemoveCheck(structs.NewCheckID("api-check", nil))
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
// Verify that we are in sync
{
req := structs.NodeSpecificRequest{
Datacenter: dc,
Node: a.Config.NodeName,
QueryOptions: structs.QueryOptions{
Token: "root",
},
}
var checks structs.IndexedHealthChecks
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
t.Fatalf("err: %v", err)
}
// We should have 1 check (just serf)
if len(checks.HealthChecks) != 1 {
t.Fatalf("bad: %v", checks)
}
// All the checks should match
for _, chk := range checks.HealthChecks {
chk.CreateIndex, chk.ModifyIndex = 0, 0
switch chk.CheckID {
case "mysql-check":
t.Fatalf("should not be permitted")
case "api-check":
t.Fatalf("should be deleted")
case "serfHealth":
// ignore
default:
t.Fatalf("unexpected check: %v", chk)
}
}
}
if err := checksInSync(a.State, 1, structs.DefaultEnterpriseMeta()); err != nil {
t.Fatal(err)
}
// Make sure the token got cleaned up.
if token := a.State.CheckToken(structs.NewCheckID("api-check", nil)); token != "" {
t.Fatalf("bad: %s", token)
}
}
func TestAgent_UpdateCheck_DiscardOutput(t *testing.T) {
t.Parallel()
a := agent.NewTestAgent(t, `
discard_check_output = true
check_update_interval = "0s" # set to "0s" since otherwise output checks are deferred
`)
defer a.Shutdown()
testrpc.WaitForLeader(t, a.RPC, "dc1")
inSync := func(id string) bool {
s := a.State.CheckState(structs.NewCheckID(types.CheckID(id), nil))
if s == nil {
return false
}
return s.InSync
}
// register a check
check := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "web",
Name: "web",
Status: api.HealthPassing,
Output: "first output",
}
if err := a.State.AddCheck(check, ""); err != nil {
t.Fatalf("bad: %s", err)
}
if err := a.State.SyncFull(); err != nil {
t.Fatalf("bad: %s", err)
}
if !inSync("web") {
t.Fatal("check should be in sync")
}
// update the check with the same status but different output
// and the check should still be in sync.
a.State.UpdateCheck(check.CompoundCheckID(), api.HealthPassing, "second output")
if !inSync("web") {
t.Fatal("check should be in sync")
}
// disable discarding of check output and update the check again with different
// output. Then the check should be out of sync.
a.State.SetDiscardCheckOutput(false)
a.State.UpdateCheck(check.CompoundCheckID(), api.HealthPassing, "third output")
if inSync("web") {
t.Fatal("check should be out of sync")
}
}
func TestAgentAntiEntropy_Check_DeferSync(t *testing.T) {
t.Parallel()
a := &agent.TestAgent{HCL: `
check_update_interval = "500ms"
`}
if err := a.Start(t); err != nil {
t.Fatal(err)
}
defer a.Shutdown()
testrpc.WaitForTestAgent(t, a.RPC, "dc1")
// Create a check
check := &structs.HealthCheck{
Node: a.Config.NodeName,
CheckID: "web",
Name: "web",
Status: api.HealthPassing,
Output: "",
}
a.State.AddCheck(check, "")
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
// Verify that we are in sync
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
var checks structs.IndexedHealthChecks
retry.Run(t, func(r *retry.R) {
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
r.Fatalf("err: %v", err)
}
if got, want := len(checks.HealthChecks), 2; got != want {
r.Fatalf("got %d health checks want %d", got, want)
}
})
// Update the check output! Should be deferred
a.State.UpdateCheck(structs.NewCheckID("web", nil), api.HealthPassing, "output")
// We are going to wait up to 850ms for the deferred check update to run. The update
// can happen any time within: check_update_interval / 2 + random(min: 0, max: check_update_interval)
// For this test that means it will get deferred for 250ms - 750ms. We add up to 100ms on top of that to
// account for potentially slow tests on a overloaded system.
timer := &retry.Timer{Timeout: 850 * time.Millisecond, Wait: 50 * time.Millisecond}
start := time.Now()
retry.RunWith(timer, t, func(r *retry.R) {
cs := a.State.CheckState(structs.NewCheckID("web", nil))
if cs == nil {
r.Fatalf("check is not registered")
}
if cs.DeferCheck != nil {
r.Fatalf("Deferred Check timeout not removed yet")
}
})
elapsed := time.Since(start)
// ensure the check deferral didn't update too fast
if elapsed < 240*time.Millisecond {
t.Fatalf("early update: elapsed %v\n\n%+v", elapsed, checks)
}
// ensure the check deferral didn't update too late
if elapsed > 850*time.Millisecond {
t.Fatalf("late update: elapsed: %v\n\n%+v", elapsed, checks)
}
// Wait for a deferred update. TODO (slackpad) This isn't a great test
// because we might be stuck in the random stagger from the full sync
// after the leader election (~3 seconds) so it's easy to exceed the
// default retry timeout here. Extending this makes the test a little
// less flaky, but this isn't very clean for this first deferred update
// since the full sync might pick it up, not the timer trigger. The
// good news is that the later update below should be well past the full
// sync so we are getting some coverage. We should rethink this a bit and
// rework the deferred update stuff to be more testable.
//
// TODO - figure out why after the deferred check calls TriggerSyncChanges that this
// takes so long to happen. I have seen it take upwards of 1.5s before the check gets
// synced.
timer = &retry.Timer{Timeout: 6 * time.Second, Wait: 100 * time.Millisecond}
retry.RunWith(timer, t, func(r *retry.R) {
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
r.Fatalf("err: %v", err)
}
// Verify updated
for _, chk := range checks.HealthChecks {
switch chk.CheckID {
case "web":
if chk.Output != "output" {
r.Fatalf("no update: %v", chk)
}
}
}
})
// Change the output in the catalog to force it out of sync.
eCopy := check.Clone()
eCopy.Output = "changed"
reg := structs.RegisterRequest{
Datacenter: a.Config.Datacenter,
Node: a.Config.NodeName,
Address: a.Config.AdvertiseAddrLAN.IP.String(),
TaggedAddresses: a.Config.TaggedAddresses,
Check: eCopy,
WriteRequest: structs.WriteRequest{},
}
var out struct{}
if err := a.RPC("Catalog.Register", &reg, &out); err != nil {
t.Fatalf("err: %s", err)
}
// Verify that the output is out of sync.
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
t.Fatalf("err: %v", err)
}
for _, chk := range checks.HealthChecks {
switch chk.CheckID {
case "web":
if chk.Output != "changed" {
t.Fatalf("unexpected update: %v", chk)
}
}
}
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
// Verify that the output was synced back to the agent's value.
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
t.Fatalf("err: %v", err)
}
for _, chk := range checks.HealthChecks {
switch chk.CheckID {
case "web":
if chk.Output != "output" {
t.Fatalf("missed update: %v", chk)
}
}
}
// Reset the catalog again.
if err := a.RPC("Catalog.Register", &reg, &out); err != nil {
t.Fatalf("err: %s", err)
}
// Verify that the output is out of sync.
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
t.Fatalf("err: %v", err)
}
for _, chk := range checks.HealthChecks {
switch chk.CheckID {
case "web":
if chk.Output != "changed" {
t.Fatalf("unexpected update: %v", chk)
}
}
}
// Now make an update that should be deferred.
a.State.UpdateCheck(structs.NewCheckID("web", nil), api.HealthPassing, "deferred")
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
// Verify that the output is still out of sync since there's a deferred
// update pending.
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
t.Fatalf("err: %v", err)
}
for _, chk := range checks.HealthChecks {
switch chk.CheckID {
case "web":
if chk.Output != "changed" {
t.Fatalf("unexpected update: %v", chk)
}
}
}
// Wait for the deferred update.
retry.Run(t, func(r *retry.R) {
if err := a.RPC("Health.NodeChecks", &req, &checks); err != nil {
r.Fatal(err)
}
// Verify updated
for _, chk := range checks.HealthChecks {
switch chk.CheckID {
case "web":
if chk.Output != "deferred" {
r.Fatalf("no update: %v", chk)
}
}
}
})
}
func TestAgentAntiEntropy_NodeInfo(t *testing.T) {
t.Parallel()
nodeID := types.NodeID("40e4a748-2192-161a-0510-9bf59fe950b5")
nodeMeta := map[string]string{
"somekey": "somevalue",
}
a := &agent.TestAgent{HCL: `
node_id = "40e4a748-2192-161a-0510-9bf59fe950b5"
node_meta {
somekey = "somevalue"
}`}
if err := a.Start(t); err != nil {
t.Fatal(err)
}
defer a.Shutdown()
testrpc.WaitForLeader(t, a.RPC, "dc1")
// Register info
args := &structs.RegisterRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
Address: "127.0.0.1",
}
var out struct{}
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
req := structs.NodeSpecificRequest{
Datacenter: "dc1",
Node: a.Config.NodeName,
}
var services structs.IndexedNodeServices
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
t.Fatalf("err: %v", err)
}
id := services.NodeServices.Node.ID
addrs := services.NodeServices.Node.TaggedAddresses
meta := services.NodeServices.Node.Meta
delete(meta, structs.MetaSegmentKey) // Added later, not in config.
require.Equal(t, a.Config.NodeID, id)
require.Equal(t, a.Config.TaggedAddresses, addrs)
assert.Equal(t, unNilMap(a.Config.NodeMeta), meta)
// Blow away the catalog version of the node info
if err := a.RPC("Catalog.Register", args, &out); err != nil {
t.Fatalf("err: %v", err)
}
if err := a.State.SyncFull(); err != nil {
t.Fatalf("err: %v", err)
}
// Wait for the sync - this should have been a sync of just the node info
if err := a.RPC("Catalog.NodeServices", &req, &services); err != nil {
t.Fatalf("err: %v", err)
}
{
id := services.NodeServices.Node.ID
addrs := services.NodeServices.Node.TaggedAddresses
meta := services.NodeServices.Node.Meta
delete(meta, structs.MetaSegmentKey) // Added later, not in config.
require.Equal(t, nodeID, id)
require.Equal(t, a.Config.TaggedAddresses, addrs)
require.Equal(t, nodeMeta, meta)
}
}
func TestAgent_ServiceTokens(t *testing.T) {
t.Parallel()
tokens := new(token.Store)
tokens.UpdateUserToken("default", token.TokenSourceConfig)
cfg := config.DefaultRuntimeConfig(`bind_addr = "127.0.0.1" data_dir = "dummy"`)
l := local.NewState(agent.LocalConfig(cfg), nil, tokens)
l.TriggerSyncChanges = func() {}
l.AddService(&structs.NodeService{ID: "redis"}, "")
// Returns default when no token is set
if token := l.ServiceToken(structs.NewServiceID("redis", nil)); token != "default" {
t.Fatalf("bad: %s", token)
}
// Returns configured token
l.AddService(&structs.NodeService{ID: "redis"}, "abc123")
if token := l.ServiceToken(structs.NewServiceID("redis", nil)); token != "abc123" {
t.Fatalf("bad: %s", token)
}
// Keeps token around for the delete
l.RemoveService(structs.NewServiceID("redis", nil))
if token := l.ServiceToken(structs.NewServiceID("redis", nil)); token != "abc123" {
t.Fatalf("bad: %s", token)
}
}
func TestAgent_CheckTokens(t *testing.T) {
t.Parallel()
tokens := new(token.Store)
tokens.UpdateUserToken("default", token.TokenSourceConfig)
cfg := config.DefaultRuntimeConfig(`bind_addr = "127.0.0.1" data_dir = "dummy"`)
l := local.NewState(agent.LocalConfig(cfg), nil, tokens)
l.TriggerSyncChanges = func() {}
// Returns default when no token is set
l.AddCheck(&structs.HealthCheck{CheckID: types.CheckID("mem")}, "")
if token := l.CheckToken(structs.NewCheckID("mem", nil)); token != "default" {
t.Fatalf("bad: %s", token)
}
// Returns configured token
l.AddCheck(&structs.HealthCheck{CheckID: types.CheckID("mem")}, "abc123")
if token := l.CheckToken(structs.NewCheckID("mem", nil)); token != "abc123" {
t.Fatalf("bad: %s", token)
}
// Keeps token around for the delete
l.RemoveCheck(structs.NewCheckID("mem", nil))
if token := l.CheckToken(structs.NewCheckID("mem", nil)); token != "abc123" {
t.Fatalf("bad: %s", token)
}
}
func TestAgent_CheckCriticalTime(t *testing.T) {
t.Parallel()
cfg := config.DefaultRuntimeConfig(`bind_addr = "127.0.0.1" data_dir = "dummy"`)
l := local.NewState(agent.LocalConfig(cfg), nil, new(token.Store))
l.TriggerSyncChanges = func() {}
svc := &structs.NodeService{ID: "redis", Service: "redis", Port: 8000}
l.AddService(svc, "")
// Add a passing check and make sure it's not critical.
checkID := types.CheckID("redis:1")
chk := &structs.HealthCheck{
Node: "node",
CheckID: checkID,
Name: "redis:1",
ServiceID: "redis",
Status: api.HealthPassing,
}
l.AddCheck(chk, "")
if checks := l.CriticalCheckStates(structs.DefaultEnterpriseMeta()); len(checks) > 0 {
t.Fatalf("should not have any critical checks")
}
// Set it to warning and make sure that doesn't show up as critical.
l.UpdateCheck(structs.NewCheckID(checkID, nil), api.HealthWarning, "")
if checks := l.CriticalCheckStates(structs.DefaultEnterpriseMeta()); len(checks) > 0 {
t.Fatalf("should not have any critical checks")
}
// Fail the check and make sure the time looks reasonable.
l.UpdateCheck(structs.NewCheckID(checkID, nil), api.HealthCritical, "")
if c, ok := l.CriticalCheckStates(structs.DefaultEnterpriseMeta())[structs.NewCheckID(checkID, nil)]; !ok {
t.Fatalf("should have a critical check")
} else if c.CriticalFor() > time.Millisecond {
t.Fatalf("bad: %#v, check was critical for %v", c, c.CriticalFor())
}
// Wait a while, then fail it again and make sure the time keeps track
// of the initial failure, and doesn't reset here. Since we are sleeping for
// 50ms the check should not be any less than that.
time.Sleep(50 * time.Millisecond)
l.UpdateCheck(chk.CompoundCheckID(), api.HealthCritical, "")
if c, ok := l.CriticalCheckStates(structs.DefaultEnterpriseMeta())[structs.NewCheckID(checkID, nil)]; !ok {
t.Fatalf("should have a critical check")
} else if c.CriticalFor() < 50*time.Millisecond {
t.Fatalf("bad: %#v, check was critical for %v", c, c.CriticalFor())
}
// Set it passing again.
l.UpdateCheck(structs.NewCheckID(checkID, nil), api.HealthPassing, "")
if checks := l.CriticalCheckStates(structs.DefaultEnterpriseMeta()); len(checks) > 0 {
t.Fatalf("should not have any critical checks")
}
// Fail the check and make sure the time looks like it started again
// from the latest failure, not the original one.
l.UpdateCheck(structs.NewCheckID(checkID, nil), api.HealthCritical, "")
if c, ok := l.CriticalCheckStates(structs.DefaultEnterpriseMeta())[structs.NewCheckID(checkID, nil)]; !ok {
t.Fatalf("should have a critical check")
} else if c.CriticalFor() > time.Millisecond {
t.Fatalf("bad: %#v, check was critical for %v", c, c.CriticalFor())
}
}
func TestAgent_AddCheckFailure(t *testing.T) {
t.Parallel()
cfg := config.DefaultRuntimeConfig(`bind_addr = "127.0.0.1" data_dir = "dummy"`)
l := local.NewState(agent.LocalConfig(cfg), nil, new(token.Store))
l.TriggerSyncChanges = func() {}
// Add a check for a service that does not exist and verify that it fails
checkID := types.CheckID("redis:1")
chk := &structs.HealthCheck{
Node: "node",
CheckID: checkID,
Name: "redis:1",
ServiceID: "redis",
Status: api.HealthPassing,
}
wantErr := errors.New(`Check "redis:1" refers to non-existent service "redis"`)
got := l.AddCheck(chk, "")
require.Equal(t, wantErr, got)
}
func TestAgent_AliasCheck(t *testing.T) {
t.Parallel()
require := require.New(t)
cfg := config.DefaultRuntimeConfig(`bind_addr = "127.0.0.1" data_dir = "dummy"`)
l := local.NewState(agent.LocalConfig(cfg), nil, new(token.Store))
l.TriggerSyncChanges = func() {}
// Add checks
require.NoError(l.AddService(&structs.NodeService{Service: "s1"}, ""))
require.NoError(l.AddService(&structs.NodeService{Service: "s2"}, ""))
require.NoError(l.AddCheck(&structs.HealthCheck{CheckID: types.CheckID("c1"), ServiceID: "s1"}, ""))
require.NoError(l.AddCheck(&structs.HealthCheck{CheckID: types.CheckID("c2"), ServiceID: "s2"}, ""))
// Add an alias
notifyCh := make(chan struct{}, 1)
require.NoError(l.AddAliasCheck(structs.NewCheckID(types.CheckID("a1"), nil), structs.NewServiceID("s1", nil), notifyCh))
// Update and verify we get notified
l.UpdateCheck(structs.NewCheckID(types.CheckID("c1"), nil), api.HealthCritical, "")
select {
case <-notifyCh:
default:
t.Fatal("notify not received")
}
// Update again and verify we do not get notified
l.UpdateCheck(structs.NewCheckID(types.CheckID("c1"), nil), api.HealthCritical, "")
select {
case <-notifyCh:
t.Fatal("notify received")
default:
}
// Update other check and verify we do not get notified
l.UpdateCheck(structs.NewCheckID(types.CheckID("c2"), nil), api.HealthCritical, "")
select {
case <-notifyCh:
t.Fatal("notify received")
default:
}
// Update change and verify we get notified
l.UpdateCheck(structs.NewCheckID(types.CheckID("c1"), nil), api.HealthPassing, "")
select {
case <-notifyCh:
default:
t.Fatal("notify not received")
}
}
func TestAgent_sendCoordinate(t *testing.T) {
t.Parallel()
a := agent.StartTestAgent(t, agent.TestAgent{Overrides: `
sync_coordinate_interval_min = "1ms"
sync_coordinate_rate_target = 10.0
consul = {
coordinate = {
update_period = "100ms"
update_batch_size = 10
update_max_batches = 1
}
}
`})
defer a.Shutdown()
testrpc.WaitForLeader(t, a.RPC, "dc1")
t.Logf("%d %d %s",
a.Config.ConsulCoordinateUpdateBatchSize,
a.Config.ConsulCoordinateUpdateMaxBatches,
a.Config.ConsulCoordinateUpdatePeriod.String())
// Make sure the coordinate is present.
req := structs.DCSpecificRequest{
Datacenter: a.Config.Datacenter,
}
var reply structs.IndexedCoordinates
retry.Run(t, func(r *retry.R) {
if err := a.RPC("Coordinate.ListNodes", &req, &reply); err != nil {
r.Fatalf("err: %s", err)
}
if len(reply.Coordinates) != 1 {
r.Fatalf("expected a coordinate: %v", reply)
}
coord := reply.Coordinates[0]
if coord.Node != a.Config.NodeName || coord.Coord == nil {
r.Fatalf("bad: %v", coord)
}
})
}
func servicesInSync(state *local.State, wantServices int, entMeta *structs.EnterpriseMeta) error {
services := state.ServiceStates(entMeta)
if got, want := len(services), wantServices; got != want {
return fmt.Errorf("got %d services want %d", got, want)
}
for id, s := range services {
if !s.InSync {
return fmt.Errorf("service %q should be in sync %+v", id.String(), s)
}
}
return nil
}
func checksInSync(state *local.State, wantChecks int, entMeta *structs.EnterpriseMeta) error {
checks := state.CheckStates(entMeta)
if got, want := len(checks), wantChecks; got != want {
return fmt.Errorf("got %d checks want %d", got, want)
}
for id, c := range checks {
if !c.InSync {
return fmt.Errorf("check %q should be in sync", id.String())
}
}
return nil
}
func TestState_Notify(t *testing.T) {
t.Parallel()
logger := hclog.New(&hclog.LoggerOptions{
Output: os.Stderr,
})
state := local.NewState(local.Config{},
logger, &token.Store{})
// Stub state syncing
state.TriggerSyncChanges = func() {}
require := require.New(t)
assert := assert.New(t)
// Register a notifier
notifyCh := make(chan struct{}, 1)
state.Notify(notifyCh)
defer state.StopNotify(notifyCh)
assert.Empty(notifyCh)
drainCh(notifyCh)
// Add a service
err := state.AddService(&structs.NodeService{
Service: "web",
}, "fake-token-web")
require.NoError(err)
// Should have a notification
assert.NotEmpty(notifyCh)
drainCh(notifyCh)
// Re-Add same service
err = state.AddService(&structs.NodeService{
Service: "web",
Port: 4444,
}, "fake-token-web")
require.NoError(err)
// Should have a notification
assert.NotEmpty(notifyCh)
drainCh(notifyCh)
// Remove service
require.NoError(state.RemoveService(structs.NewServiceID("web", nil)))
// Should have a notification
assert.NotEmpty(notifyCh)
drainCh(notifyCh)
// Stopping should... stop
state.StopNotify(notifyCh)
// Add a service
err = state.AddService(&structs.NodeService{
Service: "web",
}, "fake-token-web")
require.NoError(err)
// Should NOT have a notification
assert.Empty(notifyCh)
drainCh(notifyCh)
}
// Test that alias check is updated after AddCheck, UpdateCheck, and RemoveCheck for the same service id
func TestAliasNotifications_local(t *testing.T) {
t.Parallel()
a := agent.NewTestAgent(t, "")
defer a.Shutdown()
testrpc.WaitForTestAgent(t, a.RPC, "dc1")
// Register service with a failing TCP check
svcID := "socat"
srv := &structs.NodeService{
ID: svcID,
Service: "echo",
Tags: []string{},
Address: "127.0.0.10",
Port: 8080,
}
a.State.AddService(srv, "")
scID := "socat-sidecar-proxy"
sc := &structs.NodeService{
ID: scID,
Service: scID,
Tags: []string{},
Address: "127.0.0.10",
Port: 9090,
}
a.State.AddService(sc, "")
tcpID := types.CheckID("service:socat-tcp")
chk0 := &structs.HealthCheck{
Node: "",
CheckID: tcpID,
Name: "tcp check",
Status: api.HealthPassing,
ServiceID: svcID,
}
a.State.AddCheck(chk0, "")
// Register an alias for the service
proxyID := types.CheckID("service:socat-sidecar-proxy:2")
chk1 := &structs.HealthCheck{
Node: "",
CheckID: proxyID,
Name: "Connect Sidecar Aliasing socat",
Status: api.HealthPassing,
ServiceID: scID,
}
chkt := &structs.CheckType{
AliasService: svcID,
}
require.NoError(t, a.AddCheck(chk1, chkt, true, "", agent.ConfigSourceLocal))
// Add a failing check to the same service ID, alias should also fail
maintID := types.CheckID("service:socat-maintenance")
chk2 := &structs.HealthCheck{
Node: "",
CheckID: maintID,
Name: "socat:Service Maintenance Mode",
Status: api.HealthCritical,
ServiceID: svcID,
}
a.State.AddCheck(chk2, "")
retry.Run(t, func(r *retry.R) {
check := a.State.Check(structs.NewCheckID(proxyID, nil))
require.NotNil(r, check)
require.Equal(r, api.HealthCritical, check.Status)
})
// Remove the failing check, alias should pass
a.State.RemoveCheck(structs.NewCheckID(maintID, nil))
retry.Run(t, func(r *retry.R) {
check := a.State.Check(structs.NewCheckID(proxyID, nil))
require.NotNil(r, check)
require.Equal(r, api.HealthPassing, check.Status)
})
// Update TCP check to failing, alias should fail
a.State.UpdateCheck(structs.NewCheckID(tcpID, nil), api.HealthCritical, "")
retry.Run(t, func(r *retry.R) {
check := a.State.Check(structs.NewCheckID(proxyID, nil))
require.NotNil(r, check)
require.Equal(r, api.HealthCritical, check.Status)
})
}
// drainCh drains a channel by reading messages until it would block.
func drainCh(ch chan struct{}) {
for {
select {
case <-ch:
default:
return
}
}
}