mesh-controller/internal/link/heard_test.go

package link_test

import (
	"context"
	"strings"
	"testing"

	"github.com/novox/mesh-controller/internal/inventory"
	"github.com/novox/mesh-controller/internal/link"
)

// Turning what a node said into what the mesh keeps.
//
// This is where "refused" and "failed" become different things. They are different situations
// with different remedies — one is fixed in what was sent and the other on the machine — and the
// mapping deciding which is which had no test at all.

func heardFrom(t *testing.T, report link.Report) (*inventory.Inventory, inventory.Doing, bool) {
	t.Helper()
	inv := inventory.ForTest(t)
	ctx := context.Background()
	if _, err := inv.AddNode(ctx, report.Node); err != nil {
		t.Fatal(err)
	}
	if _, err := (link.Enrolment{Inventory: inv}).Heard(ctx, report); err != nil {
		t.Fatal(err)
	}
	doing, said, err := inv.DoingOf(ctx, report.Node)
	if err != nil {
		t.Fatal(err)
	}
	return inv, doing, said
}

func TestARefusalIsKeptAsARefusalWithTheNodesOwnWords(t *testing.T) {
	_, doing, said := heardFrom(t, link.Report{
		Node: "workstation", Refused: `resource "conf": a file needs a path`,
	})
	if !said {
		t.Fatal("a refusal was not recorded")
	}
	if doing.Outcome != inventory.OutcomeRefused {
		t.Fatalf("a refusal was recorded as %q", doing.Outcome)
	}
	if !strings.Contains(doing.Refused, "needs a path") {
		// The host says exactly what it could not accept. Anything this end wrote instead would
		// be a second, worse explanation of the same thing.
		t.Fatalf("the node's own words were not kept: %q", doing.Refused)
	}
}

func TestSomeOfItFailingIsNotARefusal(t *testing.T) {
	// Refused means the machine is exactly as it was. Failed means it is in a state nobody
	// declared. Reporting one as the other sends somebody to the wrong place.
	_, doing, _ := heardFrom(t, link.Report{
		Node:    "workstation",
		Applied: []string{"a", "b", "c"},
		Failed:  map[string]string{"svc": "unit not found", "pkg": "no such package"},
	})
	if doing.Outcome != inventory.OutcomeFailed {
		t.Fatalf("a partial failure was recorded as %q", doing.Outcome)
	}
	if doing.Applied != 3 {
		t.Fatalf("what did apply was not kept: %d", doing.Applied)
	}
	if len(doing.Failed) != 2 {
		t.Fatalf("got %+v", doing.Failed)
	}
	// Ordered, so two readings of one failure are the same reading.
	if doing.Failed[0].ID != "pkg" || doing.Failed[1].ID != "svc" {
		t.Fatalf("failures came back unordered: %+v", doing.Failed)
	}
}

func TestACleanApplyIsRecordedAsOne(t *testing.T) {
	inv, doing, _ := heardFrom(t, link.Report{Node: "workstation", Applied: []string{"a", "b"}})
	if doing.Outcome != inventory.OutcomeApplied || doing.Applied != 2 {
		t.Fatalf("got %+v", doing)
	}
	wrong, err := inv.NotDoingWhatTheyWereTold(context.Background())
	if err != nil {
		t.Fatal(err)
	}
	if len(wrong) != 0 {
		t.Fatalf("a clean apply is listed as wrong: %+v", wrong)
	}
}

func TestABareAliveDoesNotWipeTheDeclarationThatSaysANodeIsCurrent(t *testing.T) {
	// A node reports it is alive every minute and describes what it applied rarely. If a bare
	// alive were written down as a report it would replace the last real apply with an empty one
	// — clearing the declaration digest current is measured against — so a node that had just
	// caught up would read as behind within the minute, and never converge. The lab saw exactly
	// this: a heavy wave whose apply outran the first heartbeat never reached `current`.
	inv := inventory.ForTest(t)
	ctx := context.Background()
	node, err := inv.AddNode(ctx, "anchor")
	if err != nil {
		t.Fatal(err)
	}
	// The mesh sent this node a declaration, and the node applied it and named which by digest.
	const digest = "d640d1b6a1b2c3d4e5f60718293a4b5c6d7e8f90a1b2c3d4e5f6071829304152"
	if err := inv.RecordSent(ctx, node.ID, digest); err != nil {
		t.Fatal(err)
	}
	if _, err := (link.Enrolment{Inventory: inv}).Heard(ctx, link.Report{
		Node: "anchor", Applied: []string{"a", "b"}, Declared: digest, Carried: []int{5432},
	}); err != nil {
		t.Fatal(err)
	}
	currentOf := func(name string) bool {
		reports, err := inv.LastReports(ctx)
		if err != nil {
			t.Fatal(err)
		}
		for _, r := range reports {
			if r.Node == name {
				return r.Current
			}
		}
		t.Fatalf("no report for %s", name)
		return false
	}
	if !currentOf("anchor") {
		t.Fatal("a node that applied exactly what it was sent does not read as current")
	}

	// Now the node says only that it is there, as it does every minute.
	if _, err := (link.Enrolment{Inventory: inv}).Heard(ctx, link.Report{Node: "anchor"}); err != nil {
		t.Fatal(err)
	}
	if !currentOf("anchor") {
		t.Fatal("a bare alive wiped the declaration digest, so a current node now reads as behind")
	}
	// And the last real account of what it did and holds is untouched.
	doing, said, err := inv.DoingOf(ctx, "anchor")
	if err != nil {
		t.Fatal(err)
	}
	if !said || doing.Outcome != inventory.OutcomeApplied || doing.Applied != 2 {
		t.Fatalf("a bare alive overwrote the last real report: said=%v %+v", said, doing)
	}
	owned, _, err := inv.Owned(ctx, node.ID)
	if err != nil {
		t.Fatal(err)
	}
	if len(owned) != 2 {
		t.Fatalf("a bare alive replaced the account of what the machine holds: %v", owned)
	}
}

func TestAFailureDoesNotBecomeTheAccountOfWhatTheMachineHolds(t *testing.T) {
	// A partial list is not an account of what the machine holds. Recording one as though it
	// were would tell a rebuilding node to remove what it still has — which is the fault that
	// destroyed a foundation once (novox/hq 04-ISSUES/010).
	inv := inventory.ForTest(t)
	ctx := context.Background()
	node, err := inv.AddNode(ctx, "workstation")
	if err != nil {
		t.Fatal(err)
	}
	if err := inv.RecordOwned(ctx, node.ID, []string{"one", "two", "three"}); err != nil {
		t.Fatal(err)
	}
	if _, err := (link.Enrolment{Inventory: inv}).Heard(ctx, link.Report{
		Node: "workstation", Applied: []string{"one"}, Failed: map[string]string{"two": "no"},
	}); err != nil {
		t.Fatal(err)
	}
	owned, _, err := inv.Owned(ctx, node.ID)
	if err != nil {
		t.Fatal(err)
	}
	if len(owned) != 3 {
		t.Fatalf("a partial report replaced the account of what the machine holds: %v", owned)
	}
}

// novox/hq ADR 0100: what an adopted node holds, the firewall it found and what is reachable on it
// are kept from the report that carries them, and a bare word that the node is there wipes none.
func TestWhatAnAdoptedNodeHoldsIsKeptAndAnAliveWordDoesNotWipeIt(t *testing.T) {
	inv, _, _ := heardFrom(t, link.Report{
		Node: "anchor", Applied: []string{"hello-web.served"}, Firewall: "ufw",
		Held: []link.Held{{ID: "hello-web.page", Module: "hello-web", Kind: "file",
			Target: "/var/lib/hello-web/index.html", Changed: "rewritten", Kept: "/var/lib/mesh/kept/x"}},
		Reachable: []link.Reach{{Protocol: "tcp", Address: "0.0.0.0", Port: 8080, By: "hello-web",
			Published: true, ContainerPort: 80}},
	})
	ctx := context.Background()
	check := func(when string) {
		t.Helper()
		got, err := inv.AdoptionOf(ctx, "anchor")
		if err != nil {
			t.Fatal(err)
		}
		if got.Firewall != "ufw" || len(got.Held) != 1 || got.Held[0].Changed != "rewritten" ||
			len(got.Reachable) != 1 || got.Reachable[0].ContainerPort != 80 || got.At.IsZero() {
			t.Fatalf("%s: what the node said is not what was kept: %+v", when, got)
		}
	}
	check("after the report")

	if _, err := (link.Enrolment{Inventory: inv}).Heard(ctx, link.Report{Node: "anchor"}); err != nil {
		t.Fatal(err)
	}
	check("after an alive word")

	// A reconcile report carrying only adoption is recorded, though it applied nothing.
	if _, err := (link.Enrolment{Inventory: inv}).Heard(ctx, link.Report{Node: "anchor",
		Firewall: "ufw"}); err != nil {
		t.Fatal(err)
	}
	got, err := inv.AdoptionOf(ctx, "anchor")
	if err != nil {
		t.Fatal(err)
	}
	if len(got.Held) != 0 || got.Firewall != "ufw" {
		t.Fatalf("a report from an adopted node holding nothing did not empty held: %+v", got)
	}
}

// What filters a machine, and the state of its found firewall, are kept from every report that
// carries them and never cleared by one that does not (novox/hq ADR 0168).
func TestWhatFiltersAMachineIsKeptFromItsReport(t *testing.T) {
	inv, _, _ := heardFrom(t, link.Report{
		Node: "home-server", Applied: []string{"a"},
		Filters: []link.Filter{
			{Where: "table inet mesh, chain forward", Owner: "mesh", Refuses: "policy drop"},
			{Where: "chain HAL-MESH-ONLY (iptables-legacy)", Owner: "other", Refuses: "-j DROP"},
		},
		FoundFirewall: &link.FoundFirewall{Kind: "ufw", Active: false, RetiredBy: "found-inactive"},
	})
	ctx := context.Background()
	f, err := inv.FilteringOf(ctx, "home-server")
	if err != nil {
		t.Fatal(err)
	}
	if len(f.Filters) != 2 || f.Filters[1].Owner != inventory.FilterOther || f.Alone() {
		t.Fatalf("recorded %+v", f)
	}
	if f.FoundFirewall == nil || f.FoundFirewall.RetiredBy != "found-inactive" || f.FoundFirewall.Active {
		t.Fatalf("the found firewall's state: %+v", f.FoundFirewall)
	}
	if len(f.Others()) != 1 || f.Others()[0].Where != "chain HAL-MESH-ONLY (iptables-legacy)" {
		t.Fatalf("others: %+v", f.Others())
	}
	// A bare word that the node is there clears nothing.
	if _, err := (link.Enrolment{Inventory: inv}).Heard(ctx, link.Report{Node: "home-server"}); err != nil {
		t.Fatal(err)
	}
	if again, _ := inv.FilteringOf(ctx, "home-server"); len(again.Filters) != 2 {
		t.Fatalf("a bare report cleared what filters the machine: %+v", again)
	}
	// The next full report replaces it: the chain removed by hand is gone from the record.
	if _, err := (link.Enrolment{Inventory: inv}).Heard(ctx, link.Report{Node: "home-server", Applied: []string{"a"},
		Filters: []link.Filter{{Where: "table inet mesh, chain forward", Owner: "mesh", Refuses: "policy drop"}}}); err != nil {
		t.Fatal(err)
	}
	if again, _ := inv.FilteringOf(ctx, "home-server"); len(again.Filters) != 1 || !again.Alone() {
		t.Fatalf("the next report did not replace what filters the machine: %+v", again)
	}
}