A report the store cannot take yet is handed back to the broker, not acknowledged and lost (novox/hq issue 082)

internal/inventory/reachable.go

@@ -0,0 +1,35 @@
+package inventory
+
+import (
+	"errors"
+	"strings"
+
+	"github.com/jackc/pgx/v5/pgconn"
+)
+
+// Unreachable says whether an error means the store could not be asked right now, rather than
+// that it answered no.
+//
+// The difference decides whether something worth keeping is kept or lost. The store is recreated
+// when the foundation is adopted (ADR 0078), and for those seconds every write fails with "the
+// database system is starting up" or a refused connection; a caller that treats that like a
+// refusal throws away what it was writing — a node's report of the apply that caused the restart
+// was lost exactly so, and the mesh never heard from the node again (novox/hq issue 082).
+//
+// Unreachable is a connection that failed, and the server classes that mean "not now": 08
+// (connection exception) and 57 (operator intervention — starting up, shutting down, admin
+// shutdown, cannot connect now). Everything else is an answer, and asking again gets the same one.
+func Unreachable(err error) bool {
+	if err == nil {
+		return false
+	}
+	var connect *pgconn.ConnectError
+	if errors.As(err, &connect) {
+		return true
+	}
+	var pg *pgconn.PgError
+	if errors.As(err, &pg) {
+		return strings.HasPrefix(pg.Code, "08") || strings.HasPrefix(pg.Code, "57")
+	}
+	return false
+}

internal/inventory/reachable_test.go

@@ -0,0 +1,28 @@
+package inventory
+
+import (
+	"errors"
+	"fmt"
+	"testing"
+
+	"github.com/jackc/pgx/v5/pgconn"
+)
+
+// The store restarting is "not now"; a constraint the store enforced is an answer (issue 082).
+func TestAStoreThatIsStartingIsUnreachableAndARefusalIsNot(t *testing.T) {
+	for _, c := range []struct {
+		err  error
+		want bool
+	}{
+		{&pgconn.PgError{Code: "57P03", Message: "the database system is starting up"}, true},
+		{fmt.Errorf("recording: %w", &pgconn.PgError{Code: "57P01"}), true},
+		{&pgconn.PgError{Code: "08006"}, true},
+		{&pgconn.PgError{Code: "23503", Message: "violates foreign key constraint"}, false},
+		{errors.New("a report named no node"), false},
+		{nil, false},
+	} {
+		if got := Unreachable(c.err); got != c.want {
+			t.Errorf("Unreachable(%v) = %v, want %v", c.err, got, c.want)
+		}
+	}
+}