consul/test-integ/connect/snapshot_test.go

213 lines
6.1 KiB
Go

// Copyright (c) HashiCorp, Inc.
// SPDX-License-Identifier: BUSL-1.1
package connect
import (
"fmt"
"io"
"net/http"
"net/url"
"testing"
"time"
"github.com/stretchr/testify/require"
"github.com/hashicorp/consul/api"
"github.com/hashicorp/consul/sdk/testutil/retry"
"github.com/hashicorp/consul/test/integration/consul-container/libs/utils"
"github.com/hashicorp/consul/testing/deployer/sprawl/sprawltest"
"github.com/hashicorp/consul/testing/deployer/topology"
)
// Test_Snapshot_Restore_Agentless verifies consul agent can continue
// to push envoy confgi after restoring from a snapshot.
//
// - This test is to detect server agent frozen after restoring from a snapshot
// (https://github.com/hashicorp/consul/pull/18636)
//
// - This bug only appeared in agentless mode
//
// Steps:
// 1. The test spins up a one-server cluster with static-server and static-client.
// 2. A snapshot is taken and the cluster is restored from the snapshot
// 3. A new static-server replaces the old one
// 4. At the end, we assert the static-client's upstream is updated with the
// new static-server
func Test_Snapshot_Restore_Agentless(t *testing.T) {
t.Parallel()
staticServerSID := topology.NewServiceID("static-server", "default", "default")
staticClientSID := topology.NewServiceID("static-client", "default", "default")
clu := &topology.Config{
Images: utils.TargetImages(),
Networks: []*topology.Network{
{Name: "dc1"},
},
Clusters: []*topology.Cluster{
{
Name: "dc1",
Nodes: []*topology.Node{
{
Kind: topology.NodeKindServer,
// NOTE: uncomment the following lines to trigger the agent frozen bug
// Images: topology.Images{
// ConsulEnterprise: "hashicorp/consul-enterprise:1.16.1-ent",
// },
Name: "dc1-server1",
Addresses: []*topology.Address{
{Network: "dc1"},
},
},
{
Kind: topology.NodeKindDataplane,
Name: "dc1-client1",
Services: []*topology.Service{
{
ID: staticServerSID,
Image: "docker.mirror.hashicorp.services/fortio/fortio",
Port: 8080,
EnvoyAdminPort: 19000,
CheckTCP: "127.0.0.1:8080",
Command: []string{
"server",
"-http-port", "8080",
"-redirect-port", "-disabled",
},
},
},
},
{
Kind: topology.NodeKindDataplane,
Name: "dc1-client2",
Services: []*topology.Service{
{
ID: staticClientSID,
Image: "docker.mirror.hashicorp.services/fortio/fortio",
Port: 8080,
EnvoyAdminPort: 19000,
CheckTCP: "127.0.0.1:8080",
Command: []string{
"server",
"-http-port", "8080",
"-redirect-port", "-disabled",
},
Upstreams: []*topology.Upstream{
{
ID: staticServerSID,
LocalPort: 5000,
},
},
},
},
},
// Client3 for second static-server
{
Kind: topology.NodeKindDataplane,
Name: "dc1-client3",
Disabled: true,
Services: []*topology.Service{
{
ID: staticServerSID,
Image: "docker.mirror.hashicorp.services/fortio/fortio",
Port: 8080,
EnvoyAdminPort: 19000,
CheckTCP: "127.0.0.1:8080",
Command: []string{
"server",
"-http-port", "8080",
"-redirect-port", "-disabled",
},
},
},
},
},
Enterprise: utils.IsEnterprise(),
InitialConfigEntries: []api.ConfigEntry{
&api.ProxyConfigEntry{
Kind: api.ProxyDefaults,
Name: "global",
Config: map[string]any{
"protocol": "http",
},
},
&api.ServiceConfigEntry{
Kind: api.ServiceDefaults,
Name: "static-server",
},
&api.ServiceIntentionsConfigEntry{
Kind: api.ServiceIntentions,
Name: "static-server",
Sources: []*api.SourceIntention{
{
Name: "static-client",
Action: api.IntentionActionAllow,
},
},
},
},
},
},
}
sp := sprawltest.Launch(t, clu)
client, err := sp.HTTPClientForCluster("dc1")
require.NoError(t, err)
staticClient := sp.Topology().Clusters["dc1"].ServiceByID(
topology.NewNodeID("dc1-client2", "default"),
staticClientSID,
)
staticClientAddress := fmt.Sprintf("%s:%d", staticClient.Node.LocalAddress(), staticClient.Port)
// The following url causes the static-client's fortio server to
// fetch the ?url= param (the upstream static-server in our case).
url := fmt.Sprintf("http://%s/fortio/fetch2?url=%s", staticClientAddress,
url.QueryEscape("http://localhost:5000"),
)
// We retry the first request until we get 200 OK since it may take a while
// for the server to be available.
// Use a custom retry.Timer since the default one usually times out too early.
retrySendRequest := func(isSuccess bool) {
t.Log("static-client sending requests to static-server...")
retry.RunWith(&retry.Timer{Timeout: 60 * time.Second, Wait: time.Millisecond * 500}, t, func(r *retry.R) {
resp, err := client.Post(url, "text/plain", nil)
require.NoError(r, err)
defer resp.Body.Close()
if isSuccess {
require.Equal(r, http.StatusOK, resp.StatusCode)
} else {
require.NotEqual(r, http.StatusOK, resp.StatusCode)
}
body, err := io.ReadAll(resp.Body)
require.NoError(r, err)
fmt.Println("Body: ", string(body), resp.StatusCode)
})
}
retrySendRequest(true)
t.Log("...ok, got 200 responses")
t.Log("Take a snapshot of the cluster and restore ...")
err = sp.SnapshotSave("dc1")
require.NoError(t, err)
// Shutdown existing static-server
cfg := sp.Config()
cluster := cfg.Cluster("dc1")
cluster.Nodes[1].Disabled = true // client 1 -- static-server
require.NoError(t, sp.Relaunch(cfg))
retrySendRequest(false)
// Add a new static-server
cfg = sp.Config()
cluster = cfg.Cluster("dc1")
cluster.Nodes[3].Disabled = false // client 3 -- new static-server
require.NoError(t, sp.Relaunch(cfg))
// Ensure the static-client connected to static-server
retrySendRequest(true)
}