open-consul/agent/consul/discovery_chain_endpoint_te...

337 lines
9.0 KiB
Go
Raw Normal View History

package consul
import (
"fmt"
"os"
"testing"
"time"
"github.com/stretchr/testify/require"
msgpackrpc "github.com/hashicorp/consul-net-rpc/net-rpc-msgpackrpc"
"github.com/hashicorp/consul/acl"
"github.com/hashicorp/consul/agent/connect"
"github.com/hashicorp/consul/agent/structs"
"github.com/hashicorp/consul/sdk/testutil"
"github.com/hashicorp/consul/testrpc"
)
func TestDiscoveryChainEndpoint_Get(t *testing.T) {
if testing.Short() {
t.Skip("too slow for testing.Short")
}
t.Parallel()
dir1, s1 := testServerWithConfig(t, func(c *Config) {
c.PrimaryDatacenter = "dc1"
c.PrimaryDatacenter = "dc1"
c.ACLsEnabled = true
c.ACLInitialManagementToken = "root"
c.ACLResolverSettings.ACLDefaultPolicy = "deny"
})
defer os.RemoveAll(dir1)
defer s1.Shutdown()
codec := rpcClient(t, s1)
defer codec.Close()
waitForLeaderEstablishment(t, s1)
testrpc.WaitForTestAgent(t, s1.RPC, "dc1", testrpc.WithToken("root"))
denyToken, err := upsertTestTokenWithPolicyRules(codec, "root", "dc1", "")
require.NoError(t, err)
allowToken, err := upsertTestTokenWithPolicyRules(codec, "root", "dc1", `service "web" { policy = "read" }`)
require.NoError(t, err)
getChain := func(args *structs.DiscoveryChainRequest) (*structs.DiscoveryChainResponse, error) {
resp := structs.DiscoveryChainResponse{}
err := msgpackrpc.CallWithCodec(codec, "DiscoveryChain.Get", &args, &resp)
if err != nil {
return nil, err
}
// clear fields that we don't care about
resp.QueryMeta = structs.QueryMeta{}
return &resp, nil
}
newTarget := func(service, serviceSubset, namespace, partition, datacenter string) *structs.DiscoveryTarget {
t := structs.NewDiscoveryTarget(service, serviceSubset, namespace, partition, datacenter)
t.SNI = connect.TargetSNI(t, connect.TestClusterID+".consul")
t.Name = t.SNI
t.ConnectTimeout = 5 * time.Second // default
return t
}
targetWithConnectTimeout := func(t *structs.DiscoveryTarget, connectTimeout time.Duration) *structs.DiscoveryTarget {
t.ConnectTimeout = connectTimeout
return t
}
// ==== compiling the default chain (no config entries)
{ // no token
_, err := getChain(&structs.DiscoveryChainRequest{
Name: "web",
EvaluateInDatacenter: "dc1",
EvaluateInNamespace: "default",
EvaluateInPartition: "default",
Datacenter: "dc1",
})
if !acl.IsErrPermissionDenied(err) {
t.Fatalf("err: %v", err)
}
}
{ // wrong token
_, err := getChain(&structs.DiscoveryChainRequest{
Name: "web",
EvaluateInDatacenter: "dc1",
EvaluateInNamespace: "default",
EvaluateInPartition: "default",
Datacenter: "dc1",
QueryOptions: structs.QueryOptions{Token: denyToken.SecretID},
})
if !acl.IsErrPermissionDenied(err) {
t.Fatalf("err: %v", err)
}
}
expectDefaultResponse_DC1_Default := &structs.DiscoveryChainResponse{
Chain: &structs.CompiledDiscoveryChain{
ServiceName: "web",
Namespace: "default",
Partition: "default",
Datacenter: "dc1",
Protocol: "tcp",
StartNode: "resolver:web.default.default.dc1",
Default: true,
Nodes: map[string]*structs.DiscoveryGraphNode{
"resolver:web.default.default.dc1": {
Type: structs.DiscoveryGraphNodeTypeResolver,
Name: "web.default.default.dc1",
Resolver: &structs.DiscoveryResolver{
Default: true,
ConnectTimeout: 5 * time.Second,
Target: "web.default.default.dc1",
},
},
},
Targets: map[string]*structs.DiscoveryTarget{
"web.default.default.dc1": newTarget("web", "", "default", "default", "dc1"),
},
},
}
// various ways with good token
for _, tc := range []struct {
evalDC string
evalNS string
evalPart string
expect *structs.DiscoveryChainResponse
}{
{
evalDC: "dc1",
evalNS: "default",
evalPart: "default",
expect: expectDefaultResponse_DC1_Default,
},
{
evalDC: "",
evalNS: "default",
evalPart: "default",
expect: expectDefaultResponse_DC1_Default,
},
{
evalDC: "dc1",
evalNS: "",
evalPart: "default",
expect: expectDefaultResponse_DC1_Default,
},
{
evalDC: "",
evalNS: "",
evalPart: "default",
expect: expectDefaultResponse_DC1_Default,
},
{
evalDC: "dc1",
evalNS: "default",
evalPart: "",
expect: expectDefaultResponse_DC1_Default,
},
{
evalDC: "",
evalNS: "default",
evalPart: "",
expect: expectDefaultResponse_DC1_Default,
},
{
evalDC: "dc1",
evalNS: "",
evalPart: "",
expect: expectDefaultResponse_DC1_Default,
},
{
evalDC: "",
evalNS: "",
evalPart: "",
expect: expectDefaultResponse_DC1_Default,
},
} {
tc := tc
name := fmt.Sprintf("dc=%q ns=%q", tc.evalDC, tc.evalNS)
require.True(t, t.Run(name, func(t *testing.T) {
resp, err := getChain(&structs.DiscoveryChainRequest{
Name: "web",
EvaluateInDatacenter: tc.evalDC,
EvaluateInNamespace: tc.evalNS,
EvaluateInPartition: tc.evalPart,
Datacenter: "dc1",
QueryOptions: structs.QueryOptions{Token: allowToken.SecretID},
})
require.NoError(t, err)
require.Equal(t, tc.expect, resp)
}))
}
{ // Now create one config entry.
out := false
require.NoError(t, msgpackrpc.CallWithCodec(codec, "ConfigEntry.Apply",
&structs.ConfigEntryRequest{
Datacenter: "dc1",
Entry: &structs.ServiceResolverConfigEntry{
Kind: structs.ServiceResolver,
Name: "web",
ConnectTimeout: 33 * time.Second,
},
WriteRequest: structs.WriteRequest{Token: "root"},
}, &out))
require.True(t, out)
}
// ==== compiling a chain with config entries
{ // good token
resp, err := getChain(&structs.DiscoveryChainRequest{
Name: "web",
EvaluateInDatacenter: "dc1",
EvaluateInNamespace: "default",
EvaluateInPartition: "default",
Datacenter: "dc1",
QueryOptions: structs.QueryOptions{Token: allowToken.SecretID},
})
require.NoError(t, err)
expect := &structs.DiscoveryChainResponse{
Chain: &structs.CompiledDiscoveryChain{
ServiceName: "web",
Namespace: "default",
Partition: "default",
Datacenter: "dc1",
Protocol: "tcp",
StartNode: "resolver:web.default.default.dc1",
Nodes: map[string]*structs.DiscoveryGraphNode{
"resolver:web.default.default.dc1": {
Type: structs.DiscoveryGraphNodeTypeResolver,
Name: "web.default.default.dc1",
Resolver: &structs.DiscoveryResolver{
ConnectTimeout: 33 * time.Second,
Target: "web.default.default.dc1",
},
},
},
Targets: map[string]*structs.DiscoveryTarget{
"web.default.default.dc1": targetWithConnectTimeout(
newTarget("web", "", "default", "default", "dc1"),
33*time.Second,
),
},
},
}
require.Equal(t, expect, resp)
}
}
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
func TestDiscoveryChainEndpoint_Get_BlockOnNoChange(t *testing.T) {
if testing.Short() {
t.Skip("too slow for testing.Short")
}
t.Parallel()
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
_, s1 := testServerWithConfig(t, func(c *Config) {
c.DevMode = true // keep it in ram to make it 10x faster on macos
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
c.PrimaryDatacenter = "dc1"
})
codec := rpcClient(t, s1)
waitForLeaderEstablishment(t, s1)
testrpc.WaitForTestAgent(t, s1.RPC, "dc1")
{ // create one unrelated entry
var out bool
require.NoError(t, msgpackrpc.CallWithCodec(codec, "ConfigEntry.Apply", &structs.ConfigEntryRequest{
Datacenter: "dc1",
Entry: &structs.ServiceResolverConfigEntry{
Kind: structs.ServiceResolver,
Name: "unrelated",
ConnectTimeout: 33 * time.Second,
},
}, &out))
require.True(t, out)
}
run := func(t *testing.T, dataPrefix string) {
rpcBlockingQueryTestHarness(t,
func(minQueryIndex uint64) (*structs.QueryMeta, <-chan error) {
args := &structs.DiscoveryChainRequest{
Name: "web",
EvaluateInDatacenter: "dc1",
EvaluateInNamespace: "default",
EvaluateInPartition: "default",
Datacenter: "dc1",
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
}
args.QueryOptions.MinQueryIndex = minQueryIndex
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
var out structs.DiscoveryChainResponse
errCh := channelCallRPC(s1, "DiscoveryChain.Get", &args, &out, nil)
return &out.QueryMeta, errCh
},
func(i int) <-chan error {
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
var out bool
return channelCallRPC(s1, "ConfigEntry.Apply", &structs.ConfigEntryRequest{
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
Datacenter: "dc1",
Entry: &structs.ServiceConfigEntry{
Kind: structs.ServiceDefaults,
Name: fmt.Sprintf(dataPrefix+"%d", i),
},
}, &out, nil)
},
)
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
}
testutil.RunStep(t, "test the errNotFound path", func(t *testing.T) {
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
run(t, "other")
})
{ // create one relevant entry
var out bool
require.NoError(t, msgpackrpc.CallWithCodec(codec, "ConfigEntry.Apply", &structs.ConfigEntryRequest{
Entry: &structs.ServiceConfigEntry{
Kind: structs.ServiceDefaults,
Name: "web",
Protocol: "grpc",
},
}, &out))
require.True(t, out)
}
testutil.RunStep(t, "test the errNotChanged path", func(t *testing.T) {
server: suppress spurious blocking query returns where multiple config entries are involved (#12362) Starting from and extending the mechanism introduced in #12110 we can specially handle the 3 main special Consul RPC endpoints that react to many config entries in a single blocking query in Connect: - `DiscoveryChain.Get` - `ConfigEntry.ResolveServiceConfig` - `Intentions.Match` All of these will internally watch for many config entries, and at least one of those will likely be not found in any given query. Because these are blends of multiple reads the exact solution from #12110 isn't perfectly aligned, but we can tweak the approach slightly and regain the utility of that mechanism. ### No Config Entries Found In this case, despite looking for many config entries none may be found at all. Unlike #12110 in this scenario we do not return an empty reply to the caller, but instead synthesize a struct from default values to return. This can be handled nearly identically to #12110 with the first 1-2 replies being non-empty payloads followed by the standard spurious wakeup suppression mechanism from #12110. ### No Change Since Last Wakeup Once a blocking query loop on the server has completed and slept at least once, there is a further optimization we can make here to detect if any of the config entries that were present at specific versions for the prior execution of the loop are identical for the loop we just woke up for. In that scenario we can return a slightly different internal sentinel error and basically externally handle it similar to #12110. This would mean that even if 20 discovery chain read RPC handling goroutines wakeup due to the creation of an unrelated config entry, the only ones that will terminate and reply with a blob of data are those that genuinely have new data to report. ### Extra Endpoints Since this pattern is pretty reusable, other key config-entry-adjacent endpoints used by `agent/proxycfg` also were updated: - `ConfigEntry.List` - `Internal.IntentionUpstreams` (tproxy)
2022-02-25 21:46:34 +00:00
run(t, "completely-different-other")
})
}