luxolus
/
open-consul
2
0
Fork 0
Code Issues 1 Pull Requests Packages Releases Wiki Activity
open-consul/agent/proxycfg/state_test.go

683 lines
24 KiB
Go
Raw Normal View History

Proxy Config Manager (#4729) * Proxy Config Manager This component watches for local state changes on the agent and ensures that each service registered locally with Kind == connect-proxy has it's state being actively populated in the cache. This serves two purposes: 1. For the built-in proxy, it ensures that the state needed to accept connections is available in RAM shortly after registration and likely before the proxy actually starts accepting traffic. 2. For (future - next PR) xDS server and other possible future proxies that require _push_ based config discovery, this provides a mechanism to subscribe and be notified about updates to a proxy instance's config including upstream service discovery results. * Address review comments * Better comments; Better delivery of latest snapshot for slow watchers; Embed Config * Comment typos * Add upstream Stringer for funsies
2018-10-03 12:36:38 +00:00
package proxycfg
import (
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
"context"
"fmt"
"sync"
Proxy Config Manager (#4729) * Proxy Config Manager This component watches for local state changes on the agent and ensures that each service registered locally with Kind == connect-proxy has it's state being actively populated in the cache. This serves two purposes: 1. For the built-in proxy, it ensures that the state needed to accept connections is available in RAM shortly after registration and likely before the proxy actually starts accepting traffic. 2. For (future - next PR) xDS server and other possible future proxies that require _push_ based config discovery, this provides a mechanism to subscribe and be notified about updates to a proxy instance's config including upstream service discovery results. * Address review comments * Better comments; Better delivery of latest snapshot for slow watchers; Embed Config * Comment typos * Add upstream Stringer for funsies
2018-10-03 12:36:38 +00:00
"testing"
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
"github.com/hashicorp/consul/agent/cache"
cachetype "github.com/hashicorp/consul/agent/cache-types"
add test for discovery chain agent cache-type (#6130)
2019-07-15 15:09:52 +00:00
"github.com/hashicorp/consul/agent/consul/discoverychain"
Proxy Config Manager (#4729) * Proxy Config Manager This component watches for local state changes on the agent and ensures that each service registered locally with Kind == connect-proxy has it's state being actively populated in the cache. This serves two purposes: 1. For the built-in proxy, it ensures that the state needed to accept connections is available in RAM shortly after registration and likely before the proxy actually starts accepting traffic. 2. For (future - next PR) xDS server and other possible future proxies that require _push_ based config discovery, this provides a mechanism to subscribe and be notified about updates to a proxy instance's config including upstream service discovery results. * Address review comments * Better comments; Better delivery of latest snapshot for slow watchers; Embed Config * Comment typos * Add upstream Stringer for funsies
2018-10-03 12:36:38 +00:00
"github.com/hashicorp/consul/agent/structs"
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
"github.com/hashicorp/consul/sdk/testutil"
"github.com/stretchr/testify/require"
Proxy Config Manager (#4729) * Proxy Config Manager This component watches for local state changes on the agent and ensures that each service registered locally with Kind == connect-proxy has it's state being actively populated in the cache. This serves two purposes: 1. For the built-in proxy, it ensures that the state needed to accept connections is available in RAM shortly after registration and likely before the proxy actually starts accepting traffic. 2. For (future - next PR) xDS server and other possible future proxies that require _push_ based config discovery, this provides a mechanism to subscribe and be notified about updates to a proxy instance's config including upstream service discovery results. * Address review comments * Better comments; Better delivery of latest snapshot for slow watchers; Embed Config * Comment typos * Add upstream Stringer for funsies
2018-10-03 12:36:38 +00:00
)
func TestStateChanged(t *testing.T) {
tests := []struct {
name string
ns *structs.NodeService
token string
mutate func(ns structs.NodeService, token string) (*structs.NodeService, string)
want bool
}{
{
name: "nil node service",
ns: structs.TestNodeServiceProxy(t),
mutate: func(ns structs.NodeService, token string) (*structs.NodeService, string) {
return nil, token
},
want: true,
},
{
name: "same service",
ns: structs.TestNodeServiceProxy(t),
mutate: func(ns structs.NodeService, token string) (*structs.NodeService, string) {
return &ns, token
}, want: false,
},
{
name: "same service, different token",
ns: structs.TestNodeServiceProxy(t),
token: "foo",
mutate: func(ns structs.NodeService, token string) (*structs.NodeService, string) {
return &ns, "bar"
},
want: true,
},
{
name: "different service ID",
ns: structs.TestNodeServiceProxy(t),
token: "foo",
mutate: func(ns structs.NodeService, token string) (*structs.NodeService, string) {
ns.ID = "badger"
return &ns, token
},
want: true,
},
{
name: "different address",
ns: structs.TestNodeServiceProxy(t),
token: "foo",
mutate: func(ns structs.NodeService, token string) (*structs.NodeService, string) {
ns.Address = "10.10.10.10"
return &ns, token
},
want: true,
},
{
name: "different port",
ns: structs.TestNodeServiceProxy(t),
token: "foo",
mutate: func(ns structs.NodeService, token string) (*structs.NodeService, string) {
ns.Port = 12345
return &ns, token
},
want: true,
},
{
name: "different service kind",
ns: structs.TestNodeServiceProxy(t),
token: "foo",
mutate: func(ns structs.NodeService, token string) (*structs.NodeService, string) {
ns.Kind = ""
return &ns, token
},
want: true,
},
{
name: "different proxy target",
ns: structs.TestNodeServiceProxy(t),
token: "foo",
mutate: func(ns structs.NodeService, token string) (*structs.NodeService, string) {
ns.Proxy.DestinationServiceName = "badger"
return &ns, token
},
want: true,
},
{
name: "different proxy upstreams",
ns: structs.TestNodeServiceProxy(t),
token: "foo",
mutate: func(ns structs.NodeService, token string) (*structs.NodeService, string) {
ns.Proxy.Upstreams = nil
return &ns, token
},
want: true,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
require := require.New(t)
state, err := newState(tt.ns, tt.token)
require.NoError(err)
otherNS, otherToken := tt.mutate(*tt.ns, tt.token)
require.Equal(tt.want, state.Changed(otherNS, otherToken))
})
}
}
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
type testCacheNotifierRequest struct {
cacheType string
request cache.Request
ch chan<- cache.UpdateEvent
}
type testCacheNotifier struct {
lock sync.RWMutex
notifiers map[string]testCacheNotifierRequest
}
func newTestCacheNotifier() *testCacheNotifier {
return &testCacheNotifier{
notifiers: make(map[string]testCacheNotifierRequest),
}
}
func (cn *testCacheNotifier) Notify(ctx context.Context, t string, r cache.Request, correlationId string, ch chan<- cache.UpdateEvent) error {
cn.lock.Lock()
cn.notifiers[correlationId] = testCacheNotifierRequest{t, r, ch}
cn.lock.Unlock()
return nil
}
func (cn *testCacheNotifier) getNotifierRequest(t testing.TB, correlationId string) testCacheNotifierRequest {
cn.lock.RLock()
req, ok := cn.notifiers[correlationId]
cn.lock.RUnlock()
require.True(t, ok)
return req
}
func (cn *testCacheNotifier) getChanForCorrelationId(t testing.TB, correlationId string) chan<- cache.UpdateEvent {
req := cn.getNotifierRequest(t, correlationId)
require.NotNil(t, req.ch)
return req.ch
}
func (cn *testCacheNotifier) sendNotification(t testing.TB, correlationId string, event cache.UpdateEvent) {
cn.getChanForCorrelationId(t, correlationId) <- event
}
func (cn *testCacheNotifier) verifyWatch(t testing.TB, correlationId string) (string, cache.Request) {
// t.Logf("Watches: %+v", cn.notifiers)
req := cn.getNotifierRequest(t, correlationId)
require.NotNil(t, req.ch)
return req.cacheType, req.request
}
type verifyWatchRequest func(t testing.TB, cacheType string, request cache.Request)
func genVerifyDCSpecificWatch(expectedCacheType string, expectedDatacenter string) verifyWatchRequest {
return func(t testing.TB, cacheType string, request cache.Request) {
require.Equal(t, expectedCacheType, cacheType)
reqReal, ok := request.(*structs.DCSpecificRequest)
require.True(t, ok)
require.Equal(t, expectedDatacenter, reqReal.Datacenter)
}
}
func genVerifyRootsWatch(expectedDatacenter string) verifyWatchRequest {
return genVerifyDCSpecificWatch(cachetype.ConnectCARootName, expectedDatacenter)
}
func genVerifyListServicesWatch(expectedDatacenter string) verifyWatchRequest {
Updates to Config Entries and Connect for Namespaces (#7116)
2020-01-24 15:04:58 +00:00
return genVerifyDCSpecificWatch(cachetype.CatalogServiceListName, expectedDatacenter)
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
}
func verifyDatacentersWatch(t testing.TB, cacheType string, request cache.Request) {
require.Equal(t, cachetype.CatalogDatacentersName, cacheType)
_, ok := request.(*structs.DatacentersRequest)
require.True(t, ok)
}
func genVerifyLeafWatch(expectedService string, expectedDatacenter string) verifyWatchRequest {
return func(t testing.TB, cacheType string, request cache.Request) {
require.Equal(t, cachetype.ConnectCALeafName, cacheType)
reqReal, ok := request.(*cachetype.ConnectCALeafRequest)
require.True(t, ok)
require.Equal(t, expectedDatacenter, reqReal.Datacenter)
require.Equal(t, expectedService, reqReal.Service)
}
}
func genVerifyIntentionWatch(expectedService string, expectedDatacenter string) verifyWatchRequest {
return func(t testing.TB, cacheType string, request cache.Request) {
require.Equal(t, cachetype.IntentionMatchName, cacheType)
reqReal, ok := request.(*structs.IntentionQueryRequest)
require.True(t, ok)
require.Equal(t, expectedDatacenter, reqReal.Datacenter)
require.NotNil(t, reqReal.Match)
require.Equal(t, structs.IntentionMatchDestination, reqReal.Match.Type)
require.Len(t, reqReal.Match.Entries, 1)
require.Equal(t, structs.IntentionDefaultNamespace, reqReal.Match.Entries[0].Namespace)
require.Equal(t, expectedService, reqReal.Match.Entries[0].Name)
}
}
func genVerifyPreparedQueryWatch(expectedName string, expectedDatacenter string) verifyWatchRequest {
return func(t testing.TB, cacheType string, request cache.Request) {
require.Equal(t, cachetype.PreparedQueryName, cacheType)
reqReal, ok := request.(*structs.PreparedQueryExecuteRequest)
require.True(t, ok)
require.Equal(t, expectedDatacenter, reqReal.Datacenter)
require.Equal(t, expectedName, reqReal.QueryIDOrName)
require.Equal(t, true, reqReal.Connect)
}
}
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
func genVerifyDiscoveryChainWatch(expected *structs.DiscoveryChainRequest) verifyWatchRequest {
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
return func(t testing.TB, cacheType string, request cache.Request) {
require.Equal(t, cachetype.CompiledDiscoveryChainName, cacheType)
reqReal, ok := request.(*structs.DiscoveryChainRequest)
require.True(t, ok)
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
require.Equal(t, expected, reqReal)
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
}
}
func genVerifyGatewayWatch(expectedDatacenter string) verifyWatchRequest {
return func(t testing.TB, cacheType string, request cache.Request) {
require.Equal(t, cachetype.InternalServiceDumpName, cacheType)
reqReal, ok := request.(*structs.ServiceDumpRequest)
require.True(t, ok)
require.Equal(t, expectedDatacenter, reqReal.Datacenter)
require.True(t, reqReal.UseServiceKind)
require.Equal(t, structs.ServiceKindMeshGateway, reqReal.ServiceKind)
}
}
func genVerifyServiceSpecificRequest(expectedCacheType, expectedService, expectedFilter, expectedDatacenter string, connect bool) verifyWatchRequest {
return func(t testing.TB, cacheType string, request cache.Request) {
require.Equal(t, expectedCacheType, cacheType)
reqReal, ok := request.(*structs.ServiceSpecificRequest)
require.True(t, ok)
require.Equal(t, expectedDatacenter, reqReal.Datacenter)
require.Equal(t, expectedService, reqReal.ServiceName)
require.Equal(t, expectedFilter, reqReal.QueryOptions.Filter)
require.Equal(t, connect, reqReal.Connect)
}
}
func genVerifyServiceWatch(expectedService, expectedFilter, expectedDatacenter string, connect bool) verifyWatchRequest {
return genVerifyServiceSpecificRequest(cachetype.HealthServicesName, expectedService, expectedFilter, expectedDatacenter, connect)
}
// This test is meant to exercise the various parts of the cache watching done by the state as
// well as its management of the ConfigSnapshot
//
// This test is expressly not calling Watch which in turn would execute the run function in a go
// routine. This allows the test to be fully synchronous and deterministic while still being able
// to validate the logic of most of the watching and state updating.
//
// The general strategy here is to
//
// 1. Initialize a state with a call to newState + setting some of the extra stuff like the CacheNotifier
// We will not be using the CacheNotifier to send notifications but calling handleUpdate ourselves
// 2. Iterate through a list of verification stages performing validation and updates for each.
// a. Ensure that the required watches are in place and validate they are correct
// b. Process a bunch of UpdateEvents by calling handleUpdate
// c. Validate that the ConfigSnapshot has been updated appropriately
func TestState_WatchesAndUpdates(t *testing.T) {
t.Parallel()
agent: allow mesh gateways to initialize even if there are no connect services registered yet (#6576) Fixes #6543 Also improved some of the proxycfg tests to cover snapshot validity better.
2019-10-17 21:46:49 +00:00
indexedRoots, issuedCert := TestCerts(t)
rootWatchEvent := func() cache.UpdateEvent {
return cache.UpdateEvent{
CorrelationID: rootsWatchID,
Result: indexedRoots,
Err: nil,
}
}
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
type verificationStage struct {
requiredWatches map[string]verifyWatchRequest
events []cache.UpdateEvent
verifySnapshot func(t testing.TB, snap *ConfigSnapshot)
}
type testCase struct {
// the state to operate on. the logger, source, cache,
// ctx and cancel fields will be filled in by the test
ns structs.NodeService
sourceDC string
stages []verificationStage
}
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
newConnectProxyCase := func(meshGatewayProxyConfigValue structs.MeshGatewayMode) testCase {
ns := structs.NodeService{
Kind: structs.ServiceKindConnectProxy,
ID: "web-sidecar-proxy",
Service: "web-sidecar-proxy",
Address: "10.0.1.1",
Port: 443,
Proxy: structs.ConnectProxyConfig{
DestinationServiceName: "web",
Upstreams: structs.Upstreams{
structs.Upstream{
DestinationType: structs.UpstreamDestTypePreparedQuery,
DestinationName: "query",
LocalBindPort: 10001,
},
structs.Upstream{
DestinationType: structs.UpstreamDestTypeService,
DestinationName: "api",
LocalBindPort: 10002,
},
structs.Upstream{
DestinationType: structs.UpstreamDestTypeService,
DestinationName: "api-failover-remote",
Datacenter: "dc2",
LocalBindPort: 10003,
MeshGateway: structs.MeshGatewayConfig{
Mode: structs.MeshGatewayModeRemote,
},
},
structs.Upstream{
DestinationType: structs.UpstreamDestTypeService,
DestinationName: "api-failover-local",
Datacenter: "dc2",
LocalBindPort: 10004,
MeshGateway: structs.MeshGatewayConfig{
Mode: structs.MeshGatewayModeLocal,
},
},
structs.Upstream{
DestinationType: structs.UpstreamDestTypeService,
DestinationName: "api-failover-direct",
Datacenter: "dc2",
LocalBindPort: 10005,
MeshGateway: structs.MeshGatewayConfig{
Mode: structs.MeshGatewayModeNone,
},
},
structs.Upstream{
DestinationType: structs.UpstreamDestTypeService,
DestinationName: "api-dc2",
LocalBindPort: 10006,
},
},
},
}
if meshGatewayProxyConfigValue != structs.MeshGatewayModeDefault {
ns.Proxy.MeshGateway.Mode = meshGatewayProxyConfigValue
}
stage0 := verificationStage{
requiredWatches: map[string]verifyWatchRequest{
rootsWatchID: genVerifyRootsWatch("dc1"),
leafWatchID: genVerifyLeafWatch("web", "dc1"),
intentionsWatchID: genVerifyIntentionWatch("web", "dc1"),
"upstream:prepared_query:query": genVerifyPreparedQueryWatch("query", "dc1"),
"discovery-chain:api": genVerifyDiscoveryChainWatch(&structs.DiscoveryChainRequest{
Name: "api",
EvaluateInDatacenter: "dc1",
Updates to Config Entries and Connect for Namespaces (#7116)
2020-01-24 15:04:58 +00:00
EvaluateInNamespace: "",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
Datacenter: "dc1",
OverrideMeshGateway: structs.MeshGatewayConfig{
Mode: meshGatewayProxyConfigValue,
},
}),
"discovery-chain:api-failover-remote?dc=dc2": genVerifyDiscoveryChainWatch(&structs.DiscoveryChainRequest{
Name: "api-failover-remote",
EvaluateInDatacenter: "dc2",
Updates to Config Entries and Connect for Namespaces (#7116)
2020-01-24 15:04:58 +00:00
EvaluateInNamespace: "",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
Datacenter: "dc1",
OverrideMeshGateway: structs.MeshGatewayConfig{
Mode: structs.MeshGatewayModeRemote,
},
}),
"discovery-chain:api-failover-local?dc=dc2": genVerifyDiscoveryChainWatch(&structs.DiscoveryChainRequest{
Name: "api-failover-local",
EvaluateInDatacenter: "dc2",
Updates to Config Entries and Connect for Namespaces (#7116)
2020-01-24 15:04:58 +00:00
EvaluateInNamespace: "",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
Datacenter: "dc1",
OverrideMeshGateway: structs.MeshGatewayConfig{
Mode: structs.MeshGatewayModeLocal,
},
}),
"discovery-chain:api-failover-direct?dc=dc2": genVerifyDiscoveryChainWatch(&structs.DiscoveryChainRequest{
Name: "api-failover-direct",
EvaluateInDatacenter: "dc2",
Updates to Config Entries and Connect for Namespaces (#7116)
2020-01-24 15:04:58 +00:00
EvaluateInNamespace: "",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
Datacenter: "dc1",
OverrideMeshGateway: structs.MeshGatewayConfig{
Mode: structs.MeshGatewayModeNone,
},
}),
"discovery-chain:api-dc2": genVerifyDiscoveryChainWatch(&structs.DiscoveryChainRequest{
Name: "api-dc2",
EvaluateInDatacenter: "dc1",
Updates to Config Entries and Connect for Namespaces (#7116)
2020-01-24 15:04:58 +00:00
EvaluateInNamespace: "",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
Datacenter: "dc1",
OverrideMeshGateway: structs.MeshGatewayConfig{
Mode: meshGatewayProxyConfigValue,
},
}),
},
events: []cache.UpdateEvent{
agent: allow mesh gateways to initialize even if there are no connect services registered yet (#6576) Fixes #6543 Also improved some of the proxycfg tests to cover snapshot validity better.
2019-10-17 21:46:49 +00:00
rootWatchEvent(),
cache.UpdateEvent{
CorrelationID: leafWatchID,
Result: issuedCert,
Err: nil,
},
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
cache.UpdateEvent{
CorrelationID: "discovery-chain:api",
Result: &structs.DiscoveryChainResponse{
connect: generate the full SNI names for discovery targets in the compiler rather than in the xds package (#6340)
2019-08-19 18:03:03 +00:00
Chain: discoverychain.TestCompileConfigEntries(t, "api", "default", "dc1", "trustdomain.consul", "dc1",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
func(req *discoverychain.CompileRequest) {
req.OverrideMeshGateway.Mode = meshGatewayProxyConfigValue
}),
},
Err: nil,
},
cache.UpdateEvent{
CorrelationID: "discovery-chain:api-failover-remote?dc=dc2",
Result: &structs.DiscoveryChainResponse{
connect: generate the full SNI names for discovery targets in the compiler rather than in the xds package (#6340)
2019-08-19 18:03:03 +00:00
Chain: discoverychain.TestCompileConfigEntries(t, "api-failover-remote", "default", "dc2", "trustdomain.consul", "dc1",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
func(req *discoverychain.CompileRequest) {
req.OverrideMeshGateway.Mode = structs.MeshGatewayModeRemote
}),
},
Err: nil,
},
cache.UpdateEvent{
CorrelationID: "discovery-chain:api-failover-local?dc=dc2",
Result: &structs.DiscoveryChainResponse{
connect: generate the full SNI names for discovery targets in the compiler rather than in the xds package (#6340)
2019-08-19 18:03:03 +00:00
Chain: discoverychain.TestCompileConfigEntries(t, "api-failover-local", "default", "dc2", "trustdomain.consul", "dc1",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
func(req *discoverychain.CompileRequest) {
req.OverrideMeshGateway.Mode = structs.MeshGatewayModeLocal
}),
},
Err: nil,
},
cache.UpdateEvent{
CorrelationID: "discovery-chain:api-failover-direct?dc=dc2",
Result: &structs.DiscoveryChainResponse{
connect: generate the full SNI names for discovery targets in the compiler rather than in the xds package (#6340)
2019-08-19 18:03:03 +00:00
Chain: discoverychain.TestCompileConfigEntries(t, "api-failover-direct", "default", "dc2", "trustdomain.consul", "dc1",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
func(req *discoverychain.CompileRequest) {
req.OverrideMeshGateway.Mode = structs.MeshGatewayModeNone
}),
},
Err: nil,
},
cache.UpdateEvent{
CorrelationID: "discovery-chain:api-dc2",
Result: &structs.DiscoveryChainResponse{
connect: generate the full SNI names for discovery targets in the compiler rather than in the xds package (#6340)
2019-08-19 18:03:03 +00:00
Chain: discoverychain.TestCompileConfigEntries(t, "api-dc2", "default", "dc1", "trustdomain.consul", "dc1",
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
func(req *discoverychain.CompileRequest) {
req.OverrideMeshGateway.Mode = meshGatewayProxyConfigValue
},
&structs.ServiceResolverConfigEntry{
Kind: structs.ServiceResolver,
Name: "api-dc2",
Redirect: &structs.ServiceResolverRedirect{
Service: "api",
Datacenter: "dc2",
},
},
),
},
Err: nil,
},
},
agent: allow mesh gateways to initialize even if there are no connect services registered yet (#6576) Fixes #6543 Also improved some of the proxycfg tests to cover snapshot validity better.
2019-10-17 21:46:49 +00:00
verifySnapshot: func(t testing.TB, snap *ConfigSnapshot) {
require.True(t, snap.Valid())
require.True(t, snap.MeshGateway.IsEmpty())
require.Equal(t, indexedRoots, snap.Roots)
require.Equal(t, issuedCert, snap.ConnectProxy.Leaf)
require.Len(t, snap.ConnectProxy.DiscoveryChain, 5, "%+v", snap.ConnectProxy.DiscoveryChain)
require.Len(t, snap.ConnectProxy.WatchedUpstreams, 5, "%+v", snap.ConnectProxy.WatchedUpstreams)
require.Len(t, snap.ConnectProxy.WatchedUpstreamEndpoints, 5, "%+v", snap.ConnectProxy.WatchedUpstreamEndpoints)
require.Len(t, snap.ConnectProxy.WatchedGateways, 5, "%+v", snap.ConnectProxy.WatchedGateways)
require.Len(t, snap.ConnectProxy.WatchedGatewayEndpoints, 5, "%+v", snap.ConnectProxy.WatchedGatewayEndpoints)
require.Len(t, snap.ConnectProxy.WatchedServiceChecks, 0, "%+v", snap.ConnectProxy.WatchedServiceChecks)
Updates to Config Entries and Connect for Namespaces (#7116)
2020-01-24 15:04:58 +00:00
require.Len(t, snap.ConnectProxy.PreparedQueryEndpoints, 0, "%+v", snap.ConnectProxy.PreparedQueryEndpoints)
agent: allow mesh gateways to initialize even if there are no connect services registered yet (#6576) Fixes #6543 Also improved some of the proxycfg tests to cover snapshot validity better.
2019-10-17 21:46:49 +00:00
},
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
}
stage1 := verificationStage{
requiredWatches: map[string]verifyWatchRequest{
connect: expose an API endpoint to compile the discovery chain (#6248) In addition to exposing compilation over the API cleaned up the structures that would be exchanged to be cleaner and easier to support and understand. Also removed ability to configure the envoy OverprovisioningFactor.
2019-08-02 20:34:54 +00:00
"upstream-target:api.default.dc1:api": genVerifyServiceWatch("api", "", "dc1", true),
connect: fix failover through a mesh gateway to a remote datacenter (#6259) Failover is pushed entirely down to the data plane by creating envoy clusters and putting each successive destination in a different load assignment priority band. For example this shows that normally requests go to 1.2.3.4:8080 but when that fails they go to 6.7.8.9:8080: - name: foo load_assignment: cluster_name: foo policy: overprovisioning_factor: 100000 endpoints: - priority: 0 lb_endpoints: - endpoint: address: socket_address: address: 1.2.3.4 port_value: 8080 - priority: 1 lb_endpoints: - endpoint: address: socket_address: address: 6.7.8.9 port_value: 8080 Mesh gateways route requests based solely on the SNI header tacked onto the TLS layer. Envoy currently only lets you configure the outbound SNI header at the cluster layer. If you try to failover through a mesh gateway you ideally would configure the SNI value per endpoint, but that's not possible in envoy today. This PR introduces a simpler way around the problem for now: 1. We identify any target of failover that will use mesh gateway mode local or remote and then further isolate any resolver node in the compiled discovery chain that has a failover destination set to one of those targets. 2. For each of these resolvers we will perform a small measurement of comparative healths of the endpoints that come back from the health API for the set of primary target and serial failover targets. We walk the list of targets in order and if any endpoint is healthy we return that target, otherwise we move on to the next target. 3. The CDS and EDS endpoints both perform the measurements in (2) for the affected resolver nodes. 4. For CDS this measurement selects which TLS SNI field to use for the cluster (note the cluster is always going to be named for the primary target) 5. For EDS this measurement selects which set of endpoints will populate the cluster. Priority tiered failover is ignored. One of the big downsides to this approach to failover is that the failover detection and correction is going to be controlled by consul rather than deferring that entirely to the data plane as with the prior version. This also means that we are bound to only failover using official health signals and cannot make use of data plane signals like outlier detection to affect failover. In this specific scenario the lack of data plane signals is ok because the effectiveness is already muted by the fact that the ultimate destination endpoints will have their data plane signals scrambled when they pass through the mesh gateway wrapper anyway so we're not losing much. Another related fix is that we now use the endpoint health from the underlying service, not the health of the gateway (regardless of failover mode).
2019-08-05 18:30:35 +00:00
"upstream-target:api-failover-remote.default.dc2:api-failover-remote?dc=dc2": genVerifyServiceWatch("api-failover-remote", "", "dc2", true),
"upstream-target:api-failover-local.default.dc2:api-failover-local?dc=dc2": genVerifyServiceWatch("api-failover-local", "", "dc2", true),
connect: expose an API endpoint to compile the discovery chain (#6248) In addition to exposing compilation over the API cleaned up the structures that would be exchanged to be cleaner and easier to support and understand. Also removed ability to configure the envoy OverprovisioningFactor.
2019-08-02 20:34:54 +00:00
"upstream-target:api-failover-direct.default.dc2:api-failover-direct?dc=dc2": genVerifyServiceWatch("api-failover-direct", "", "dc2", true),
connect: fix failover through a mesh gateway to a remote datacenter (#6259) Failover is pushed entirely down to the data plane by creating envoy clusters and putting each successive destination in a different load assignment priority band. For example this shows that normally requests go to 1.2.3.4:8080 but when that fails they go to 6.7.8.9:8080: - name: foo load_assignment: cluster_name: foo policy: overprovisioning_factor: 100000 endpoints: - priority: 0 lb_endpoints: - endpoint: address: socket_address: address: 1.2.3.4 port_value: 8080 - priority: 1 lb_endpoints: - endpoint: address: socket_address: address: 6.7.8.9 port_value: 8080 Mesh gateways route requests based solely on the SNI header tacked onto the TLS layer. Envoy currently only lets you configure the outbound SNI header at the cluster layer. If you try to failover through a mesh gateway you ideally would configure the SNI value per endpoint, but that's not possible in envoy today. This PR introduces a simpler way around the problem for now: 1. We identify any target of failover that will use mesh gateway mode local or remote and then further isolate any resolver node in the compiled discovery chain that has a failover destination set to one of those targets. 2. For each of these resolvers we will perform a small measurement of comparative healths of the endpoints that come back from the health API for the set of primary target and serial failover targets. We walk the list of targets in order and if any endpoint is healthy we return that target, otherwise we move on to the next target. 3. The CDS and EDS endpoints both perform the measurements in (2) for the affected resolver nodes. 4. For CDS this measurement selects which TLS SNI field to use for the cluster (note the cluster is always going to be named for the primary target) 5. For EDS this measurement selects which set of endpoints will populate the cluster. Priority tiered failover is ignored. One of the big downsides to this approach to failover is that the failover detection and correction is going to be controlled by consul rather than deferring that entirely to the data plane as with the prior version. This also means that we are bound to only failover using official health signals and cannot make use of data plane signals like outlier detection to affect failover. In this specific scenario the lack of data plane signals is ok because the effectiveness is already muted by the fact that the ultimate destination endpoints will have their data plane signals scrambled when they pass through the mesh gateway wrapper anyway so we're not losing much. Another related fix is that we now use the endpoint health from the underlying service, not the health of the gateway (regardless of failover mode).
2019-08-05 18:30:35 +00:00
"mesh-gateway:dc2:api-failover-remote?dc=dc2": genVerifyGatewayWatch("dc2"),
"mesh-gateway:dc1:api-failover-local?dc=dc2": genVerifyGatewayWatch("dc1"),
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
},
agent: allow mesh gateways to initialize even if there are no connect services registered yet (#6576) Fixes #6543 Also improved some of the proxycfg tests to cover snapshot validity better.
2019-10-17 21:46:49 +00:00
verifySnapshot: func(t testing.TB, snap *ConfigSnapshot) {
require.True(t, snap.Valid())
require.True(t, snap.MeshGateway.IsEmpty())
require.Equal(t, indexedRoots, snap.Roots)
require.Equal(t, issuedCert, snap.ConnectProxy.Leaf)
require.Len(t, snap.ConnectProxy.DiscoveryChain, 5, "%+v", snap.ConnectProxy.DiscoveryChain)
require.Len(t, snap.ConnectProxy.WatchedUpstreams, 5, "%+v", snap.ConnectProxy.WatchedUpstreams)
require.Len(t, snap.ConnectProxy.WatchedUpstreamEndpoints, 5, "%+v", snap.ConnectProxy.WatchedUpstreamEndpoints)
require.Len(t, snap.ConnectProxy.WatchedGateways, 5, "%+v", snap.ConnectProxy.WatchedGateways)
require.Len(t, snap.ConnectProxy.WatchedGatewayEndpoints, 5, "%+v", snap.ConnectProxy.WatchedGatewayEndpoints)
require.Len(t, snap.ConnectProxy.WatchedServiceChecks, 0, "%+v", snap.ConnectProxy.WatchedServiceChecks)
Updates to Config Entries and Connect for Namespaces (#7116)
2020-01-24 15:04:58 +00:00
require.Len(t, snap.ConnectProxy.PreparedQueryEndpoints, 0, "%+v", snap.ConnectProxy.PreparedQueryEndpoints)
agent: allow mesh gateways to initialize even if there are no connect services registered yet (#6576) Fixes #6543 Also improved some of the proxycfg tests to cover snapshot validity better.
2019-10-17 21:46:49 +00:00
},
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
}
connect: fix failover through a mesh gateway to a remote datacenter (#6259) Failover is pushed entirely down to the data plane by creating envoy clusters and putting each successive destination in a different load assignment priority band. For example this shows that normally requests go to 1.2.3.4:8080 but when that fails they go to 6.7.8.9:8080: - name: foo load_assignment: cluster_name: foo policy: overprovisioning_factor: 100000 endpoints: - priority: 0 lb_endpoints: - endpoint: address: socket_address: address: 1.2.3.4 port_value: 8080 - priority: 1 lb_endpoints: - endpoint: address: socket_address: address: 6.7.8.9 port_value: 8080 Mesh gateways route requests based solely on the SNI header tacked onto the TLS layer. Envoy currently only lets you configure the outbound SNI header at the cluster layer. If you try to failover through a mesh gateway you ideally would configure the SNI value per endpoint, but that's not possible in envoy today. This PR introduces a simpler way around the problem for now: 1. We identify any target of failover that will use mesh gateway mode local or remote and then further isolate any resolver node in the compiled discovery chain that has a failover destination set to one of those targets. 2. For each of these resolvers we will perform a small measurement of comparative healths of the endpoints that come back from the health API for the set of primary target and serial failover targets. We walk the list of targets in order and if any endpoint is healthy we return that target, otherwise we move on to the next target. 3. The CDS and EDS endpoints both perform the measurements in (2) for the affected resolver nodes. 4. For CDS this measurement selects which TLS SNI field to use for the cluster (note the cluster is always going to be named for the primary target) 5. For EDS this measurement selects which set of endpoints will populate the cluster. Priority tiered failover is ignored. One of the big downsides to this approach to failover is that the failover detection and correction is going to be controlled by consul rather than deferring that entirely to the data plane as with the prior version. This also means that we are bound to only failover using official health signals and cannot make use of data plane signals like outlier detection to affect failover. In this specific scenario the lack of data plane signals is ok because the effectiveness is already muted by the fact that the ultimate destination endpoints will have their data plane signals scrambled when they pass through the mesh gateway wrapper anyway so we're not losing much. Another related fix is that we now use the endpoint health from the underlying service, not the health of the gateway (regardless of failover mode).
2019-08-05 18:30:35 +00:00
if meshGatewayProxyConfigValue == structs.MeshGatewayModeLocal {
stage1.requiredWatches["mesh-gateway:dc1:api-dc2"] = genVerifyGatewayWatch("dc1")
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
}
return testCase{
ns: ns,
sourceDC: "dc1",
stages: []verificationStage{stage0, stage1},
}
}
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
cases := map[string]testCase{
"initial-gateway": testCase{
ns: structs.NodeService{
Kind: structs.ServiceKindMeshGateway,
ID: "mesh-gateway",
Service: "mesh-gateway",
Address: "10.0.1.1",
Port: 443,
},
sourceDC: "dc1",
stages: []verificationStage{
verificationStage{
requiredWatches: map[string]verifyWatchRequest{
rootsWatchID: genVerifyRootsWatch("dc1"),
serviceListWatchID: genVerifyListServicesWatch("dc1"),
datacentersWatchID: verifyDatacentersWatch,
},
agent: allow mesh gateways to initialize even if there are no connect services registered yet (#6576) Fixes #6543 Also improved some of the proxycfg tests to cover snapshot validity better.
2019-10-17 21:46:49 +00:00
verifySnapshot: func(t testing.TB, snap *ConfigSnapshot) {
require.False(t, snap.Valid(), "gateway without root is not valid")
require.True(t, snap.ConnectProxy.IsEmpty())
},
},
verificationStage{
events: []cache.UpdateEvent{
rootWatchEvent(),
},
verifySnapshot: func(t testing.TB, snap *ConfigSnapshot) {
require.False(t, snap.Valid(), "gateway without services is valid")
require.True(t, snap.ConnectProxy.IsEmpty())
require.Equal(t, indexedRoots, snap.Roots)
require.Empty(t, snap.MeshGateway.WatchedServices)
require.False(t, snap.MeshGateway.WatchedServicesSet)
require.Empty(t, snap.MeshGateway.WatchedDatacenters)
require.Empty(t, snap.MeshGateway.ServiceGroups)
require.Empty(t, snap.MeshGateway.ServiceResolvers)
require.Empty(t, snap.MeshGateway.GatewayGroups)
},
},
verificationStage{
events: []cache.UpdateEvent{
cache.UpdateEvent{
CorrelationID: serviceListWatchID,
Updates to Config Entries and Connect for Namespaces (#7116)
2020-01-24 15:04:58 +00:00
Result: &structs.IndexedServiceList{
Services: make(structs.ServiceList, 0),
agent: allow mesh gateways to initialize even if there are no connect services registered yet (#6576) Fixes #6543 Also improved some of the proxycfg tests to cover snapshot validity better.
2019-10-17 21:46:49 +00:00
},
Err: nil,
},
},
verifySnapshot: func(t testing.TB, snap *ConfigSnapshot) {
require.True(t, snap.Valid(), "gateway with empty service list is vaild")
require.True(t, snap.ConnectProxy.IsEmpty())
require.Equal(t, indexedRoots, snap.Roots)
require.Empty(t, snap.MeshGateway.WatchedServices)
require.True(t, snap.MeshGateway.WatchedServicesSet)
require.Empty(t, snap.MeshGateway.WatchedDatacenters)
require.Empty(t, snap.MeshGateway.ServiceGroups)
require.Empty(t, snap.MeshGateway.ServiceResolvers)
require.Empty(t, snap.MeshGateway.GatewayGroups)
},
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
},
},
},
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
"connect-proxy": newConnectProxyCase(structs.MeshGatewayModeDefault),
"connect-proxy-mesh-gateway-local": newConnectProxyCase(structs.MeshGatewayModeLocal),
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
}
for name, tc := range cases {
t.Run(name, func(t *testing.T) {
state, err := newState(&tc.ns, "")
// verify building the initial state worked
require.NoError(t, err)
require.NotNil(t, state)
// setup the test logger to use the t.Log
Allow users to configure either unstructured or JSON logging (#7130) * hclog Allow users to choose between unstructured and JSON logging
2020-01-28 23:50:41 +00:00
state.logger = testutil.Logger(t)
Various Gateway Fixes (#6093) * Ensure the mesh gateway configuration comes back in the api within each upstream * Add a test for the MeshGatewayConfig in the ToAPI functions * Ensure we don’t use gateways for dc local connections * Update the svc kind index for deletions * Replace the proxycfg.state cache with an interface for testing Also start implementing proxycfg state testing. * Update the state tests to verify some gateway watches for upstream-targets of a discovery chain.
2019-07-12 21:19:37 +00:00
// setup a new testing cache notifier
cn := newTestCacheNotifier()
state.cache = cn
// setup the local datacenter information
state.source = &structs.QuerySource{
Datacenter: tc.sourceDC,
}
// setup the ctx as initWatches expects this to be there
state.ctx, state.cancel = context.WithCancel(context.Background())
// ensure the initial watch setup did not error
require.NoError(t, state.initWatches())
// get the initial configuration snapshot
snap := state.initialConfigSnapshot()
//--------------------------------------------------------------------
//
// All the nested subtests here are to make failures easier to
// correlate back with the test table
//
//--------------------------------------------------------------------
for idx, stage := range tc.stages {
require.True(t, t.Run(fmt.Sprintf("stage-%d", idx), func(t *testing.T) {
for correlationId, verifier := range stage.requiredWatches {
require.True(t, t.Run(correlationId, func(t *testing.T) {
// verify that the watch was initiated
cacheType, request := cn.verifyWatch(t, correlationId)
// run the verifier if any
if verifier != nil {
verifier(t, cacheType, request)
}
}))
}
// the state is not currently executing the run method in a goroutine
// therefore we just tell it about the updates
for eveIdx, event := range stage.events {
require.True(t, t.Run(fmt.Sprintf("update-%d", eveIdx), func(t *testing.T) {
require.NoError(t, state.handleUpdate(event, &snap))
}))
}
// verify the snapshot
if stage.verifySnapshot != nil {
stage.verifySnapshot(t, &snap)
}
}))
}
})
}
}