Files
cairnobs/enterprise/internal/audit/integration_test.go
T
jcoffey-dev 7a86008062 Complete the low-risk half of the Sentry -> Cairn OBS rebrand
Sweeps the references that carry no runtime coupling, and fixes one that
turned out to be a real bug rather than stale branding.

Docker network: sentry_default -> cairnobs_default across 23 runbook and
test-header `docker run` commands. Compose derives the network from the
directory name, so this lands together with renaming the working copy to
cairnobs/ -- the two are only correct as one change.

Stale references corrected: four Dockerfile "repo root (sentry/)"
headers; .env pointing at the long-renamed deploy/helm/sentry/ chart;
five Helm comments describing the topic as sentry.logs.raw when all four
code paths have defaulted to cairnobs.logs.raw for some time; an
absolute /home/john/Projects/sentry/ path in the operator's package doc,
now repo-relative; the hand-written Tenant CRD description in both of
its identical copies, whose Go source already said Cairn OBS.

Migration 0043 repoints the default tenant's data source. 0026 seeded it
with ('sentry', '/var/lib/sentry-search') to match what
api/internal/config then defaulted to; the rebrand later moved those
defaults to "cairnobs" and /var/lib/cairnobs-search without moving the
already-applied row, leaving the default tenant naming a ClickHouse
database nothing writes to. Scoped to the exact stale values so it is a
no-op on any deployment that set them deliberately. 0026's comment is
annotated as superseded; its applied SQL is untouched.

Deliberately not included: the gRPC wire packages (sentry.logs.v1,
sentry.agent.v1) and proto/sentry/ import paths, which cannot change
without a lockstep agent/server upgrade; the Helm chart's
sentry_metadata database and sentry role, which need a real Postgres
migration on existing deployments; and the compliance audit records in
docs/compliance/, which are a dated historical record.

go build, go vet, and go test pass for ingest and deploy/operator.
2026-08-22 18:39:25 -07:00

394 lines
15 KiB
Go

// Integration tests against a real Postgres, authenticated as the real
// audit_writer role -- this package's whole point is a set of guarantees
// (grants, the trigger, hash-chain correctness under concurrency) that a
// mocked pgxpool can't actually exercise. Skipped unless
// AUDIT_TEST_POSTGRES_ADDR is set; run via:
//
// docker run --rm --network cairnobs_default -v $(pwd)/../../..:/src -w /src/enterprise \
// -e AUDIT_TEST_POSTGRES_ADDR=metadata-postgres:5432 \
// -e AUDIT_TEST_POSTGRES_PASSWORD=audit-writer-dev-only \
// -e AUDIT_TEST_ADMIN_PASSWORD=cairnobs-dev-only \
// golang:1.25-alpine go test ./internal/audit/... -v
package audit
import (
"context"
"encoding/json"
"fmt"
"os"
"path/filepath"
"sync"
"testing"
"time"
"github.com/jackc/pgx/v5/pgxpool"
"github.com/cairnobs/cairnobs/api/ai/aiapi"
"github.com/cairnobs/cairnobs/api/authz"
"github.com/cairnobs/cairnobs/api/queryapi"
)
func testPool(t *testing.T, user, password string) *pgxpool.Pool {
t.Helper()
addr := os.Getenv("AUDIT_TEST_POSTGRES_ADDR")
if addr == "" {
t.Skip("AUDIT_TEST_POSTGRES_ADDR not set -- skipping live-Postgres integration test")
}
dsn := fmt.Sprintf("postgres://%s:%s@%s/cairnobs_metadata", user, password, addr)
pool, err := pgxpool.New(context.Background(), dsn)
if err != nil {
t.Fatalf("opening pool: %v", err)
}
t.Cleanup(pool.Close)
return pool
}
func cleanupAuditLog(t *testing.T, adminPool *pgxpool.Pool) {
t.Helper()
ctx := context.Background()
// Errors here were previously swallowed (_, _ =) -- that hid the
// real cause of a test failure (rows accumulating across test runs)
// behind what looked like a row-count/ID-assumption bug instead.
// Surface them.
if _, err := adminPool.Exec(ctx, "ALTER TABLE audit_log DISABLE TRIGGER audit_log_immutable"); err != nil {
t.Fatalf("cleanup: disabling trigger: %v", err)
}
tag, err := adminPool.Exec(ctx, "DELETE FROM audit_log")
if err != nil {
t.Fatalf("cleanup: deleting rows: %v", err)
}
t.Logf("cleanup: deleted %d pre-existing rows", tag.RowsAffected())
if _, err := adminPool.Exec(ctx, "ALTER TABLE audit_log ENABLE TRIGGER audit_log_immutable"); err != nil {
t.Fatalf("cleanup: re-enabling trigger: %v", err)
}
}
func TestAppendAndVerifyChainRealPostgres(t *testing.T) {
writerPool := testPool(t, "audit_writer", os.Getenv("AUDIT_TEST_POSTGRES_PASSWORD"))
adminPool := testPool(t, "cairnobs", os.Getenv("AUDIT_TEST_ADMIN_PASSWORD"))
cleanupAuditLog(t, adminPool)
defer cleanupAuditLog(t, adminPool)
store := NewStore(writerPool)
ctx := context.Background()
for i := 0; i < 5; i++ {
q := fmt.Sprintf("service=api | stats count %d", i)
rec, err := store.Append(ctx, Entry{
TenantID: "default", Source: SourceAPI, EventType: EventQuery,
QueryText: &q, Status: StatusSuccess,
})
if err != nil {
t.Fatalf("Append %d: %v", i, err)
}
if rec.RowHash == "" {
t.Fatalf("expected a non-empty row hash")
}
}
result, err := store.VerifyChain(ctx)
if err != nil {
t.Fatalf("VerifyChain: %v", err)
}
if !result.OK {
t.Fatalf("expected an intact chain, got broken at id=%d after %d rows checked", result.FirstBadID, result.RowsChecked)
}
if result.RowsChecked != 5 {
t.Fatalf("RowsChecked = %d, want 5", result.RowsChecked)
}
}
// TestQueryAPILoggerWritesAttributedToContextIdentity proves the
// adapter queryapi.Handler actually calls in production (via
// enterprise-api's wiring) reads tenant/user from context, not from any
// field on QueryAuditEntry -- matching that type's own doc comment.
func TestQueryAPILoggerWritesAttributedToContextIdentity(t *testing.T) {
writerPool := testPool(t, "audit_writer", os.Getenv("AUDIT_TEST_POSTGRES_PASSWORD"))
adminPool := testPool(t, "cairnobs", os.Getenv("AUDIT_TEST_ADMIN_PASSWORD"))
cleanupAuditLog(t, adminPool)
defer cleanupAuditLog(t, adminPool)
logger := NewQueryAPILogger(NewStore(writerPool), SourceAPI)
ctx := authz.WithIdentity(context.Background(), authz.Identity{TenantID: "acme", UserID: "11111111-1111-1111-1111-111111111111", Role: authz.RoleViewer})
err := logger.LogQuery(ctx, queryapi.QueryAuditEntry{
Query: "stats count", Language: "spl", RowCount: 3, Duration: 42 * time.Millisecond, Success: true,
})
if err != nil {
t.Fatalf("LogQuery: %v", err)
}
var tenantID, userID, queryText string
row := adminPool.QueryRow(context.Background(),
`SELECT tenant_id, user_id, query_text FROM audit_log ORDER BY id DESC LIMIT 1`)
if err := row.Scan(&tenantID, &userID, &queryText); err != nil {
t.Fatalf("reading back the written row: %v", err)
}
if tenantID != "acme" || userID != "11111111-1111-1111-1111-111111111111" || queryText != "stats count" {
t.Fatalf("got tenant_id=%q user_id=%q query_text=%q, want acme/11111111-.../\"stats count\"", tenantID, userID, queryText)
}
}
func TestQueryAPILoggerRefusesWithoutIdentity(t *testing.T) {
writerPool := testPool(t, "audit_writer", os.Getenv("AUDIT_TEST_POSTGRES_PASSWORD"))
adminPool := testPool(t, "cairnobs", os.Getenv("AUDIT_TEST_ADMIN_PASSWORD"))
cleanupAuditLog(t, adminPool)
defer cleanupAuditLog(t, adminPool)
logger := NewQueryAPILogger(NewStore(writerPool), SourceAPI)
err := logger.LogQuery(context.Background(), queryapi.QueryAuditEntry{Query: "stats count", Success: true})
if err == nil {
t.Fatal("expected LogQuery to refuse writing an entry with no tenant identity in context")
}
}
// TestAIInteractionLoggerWritesAttributedToContextIdentity is
// AIInteractionLogger's counterpart to TestQueryAPILoggerWritesAttributedToContextIdentity
// above (Phase 7 task 12) -- same "reads identity from ctx" contract,
// plus a check that Detail actually round-trips the operation/
// confidence/accepted/edited fields that don't have dedicated columns.
func TestAIInteractionLoggerWritesAttributedToContextIdentity(t *testing.T) {
writerPool := testPool(t, "audit_writer", os.Getenv("AUDIT_TEST_POSTGRES_PASSWORD"))
adminPool := testPool(t, "cairnobs", os.Getenv("AUDIT_TEST_ADMIN_PASSWORD"))
cleanupAuditLog(t, adminPool)
defer cleanupAuditLog(t, adminPool)
logger := NewAIInteractionLogger(NewStore(writerPool), SourceAPI)
ctx := authz.WithIdentity(context.Background(), authz.Identity{TenantID: "acme", UserID: "22222222-2222-2222-2222-222222222222", Role: authz.RoleViewer})
err := logger.LogInteraction(ctx, aiapi.InteractionEntry{
Operation: "translate",
Input: "errors in the last hour",
Output: "earliest=-1h severity=ERROR",
Confidence: "high",
Accepted: true,
Edited: false,
FinalQuery: "earliest=-1h severity=ERROR",
})
if err != nil {
t.Fatalf("LogInteraction: %v", err)
}
var tenantID, userID, eventType, queryText string
var detail []byte
row := adminPool.QueryRow(context.Background(),
`SELECT tenant_id, user_id, event_type, query_text, detail FROM audit_log ORDER BY id DESC LIMIT 1`)
if err := row.Scan(&tenantID, &userID, &eventType, &queryText, &detail); err != nil {
t.Fatalf("reading back the written row: %v", err)
}
if tenantID != "acme" || userID != "22222222-2222-2222-2222-222222222222" {
t.Fatalf("got tenant_id=%q user_id=%q, want acme/22222222-...", tenantID, userID)
}
if eventType != "ai_interaction" {
t.Fatalf("event_type = %q, want ai_interaction", eventType)
}
if queryText != "earliest=-1h severity=ERROR" {
t.Fatalf("query_text = %q, want the final query", queryText)
}
var parsed struct {
Operation string `json:"operation"`
Accepted bool `json:"accepted"`
}
if err := json.Unmarshal(detail, &parsed); err != nil {
t.Fatalf("unmarshaling detail: %v", err)
}
if parsed.Operation != "translate" || !parsed.Accepted {
t.Fatalf("detail = %+v, want operation=translate accepted=true", parsed)
}
}
func TestAIInteractionLoggerRefusesWithoutIdentity(t *testing.T) {
writerPool := testPool(t, "audit_writer", os.Getenv("AUDIT_TEST_POSTGRES_PASSWORD"))
adminPool := testPool(t, "cairnobs", os.Getenv("AUDIT_TEST_ADMIN_PASSWORD"))
cleanupAuditLog(t, adminPool)
defer cleanupAuditLog(t, adminPool)
logger := NewAIInteractionLogger(NewStore(writerPool), SourceAPI)
err := logger.LogInteraction(context.Background(), aiapi.InteractionEntry{Operation: "fix", Accepted: false})
if err == nil {
t.Fatal("expected LogInteraction to refuse writing an entry with no tenant identity in context")
}
}
// TestVerifyChainDetectsTampering proves the chain actually catches an
// in-place row modification -- not just that VerifyChain runs without
// erroring on untampered data, which a bug returning OK unconditionally
// would also pass.
func TestVerifyChainDetectsTampering(t *testing.T) {
writerPool := testPool(t, "audit_writer", os.Getenv("AUDIT_TEST_POSTGRES_PASSWORD"))
adminPool := testPool(t, "cairnobs", os.Getenv("AUDIT_TEST_ADMIN_PASSWORD"))
cleanupAuditLog(t, adminPool)
defer cleanupAuditLog(t, adminPool)
store := NewStore(writerPool)
ctx := context.Background()
var lastID int64
for i := 0; i < 3; i++ {
q := "service=api"
rec, err := store.Append(ctx, Entry{TenantID: "default", Source: SourceAPI, EventType: EventQuery, QueryText: &q, Status: StatusSuccess})
if err != nil {
t.Fatalf("Append: %v", err)
}
lastID = rec.ID
}
before, err := store.VerifyChain(ctx)
if err != nil || !before.OK {
t.Fatalf("expected chain to verify before tampering: ok=%v err=%v", before.OK, err)
}
// Simulate tampering: a privileged actor disables the trigger (the
// same escape hatch confirmed live in the design doc's verification
// -- this is the "even the trigger doesn't stop a superuser" case)
// and rewrites a row's status without recomputing the hash chain.
if _, err := adminPool.Exec(ctx, "ALTER TABLE audit_log DISABLE TRIGGER audit_log_immutable"); err != nil {
t.Fatalf("disabling trigger for the tamper simulation: %v", err)
}
if _, err := adminPool.Exec(ctx, "UPDATE audit_log SET status = 'error' WHERE id = $1", lastID); err != nil {
t.Fatalf("simulated tamper UPDATE: %v", err)
}
if _, err := adminPool.Exec(ctx, "ALTER TABLE audit_log ENABLE TRIGGER audit_log_immutable"); err != nil {
t.Fatalf("re-enabling trigger: %v", err)
}
after, err := store.VerifyChain(ctx)
if err != nil {
t.Fatalf("VerifyChain after tampering: %v", err)
}
if after.OK {
t.Fatalf("expected VerifyChain to detect the tampered row, got OK")
}
if after.FirstBadID != lastID {
t.Fatalf("FirstBadID = %d, want %d", after.FirstBadID, lastID)
}
}
// TestAppendConcurrentWritesProduceAValidChain exercises the advisory
// lock: without it, concurrent Append calls could read the same
// prev_hash and fork the chain. Real concurrency, real Postgres, not a
// unit test of the Go code alone.
func TestAppendConcurrentWritesProduceAValidChain(t *testing.T) {
writerPool := testPool(t, "audit_writer", os.Getenv("AUDIT_TEST_POSTGRES_PASSWORD"))
adminPool := testPool(t, "cairnobs", os.Getenv("AUDIT_TEST_ADMIN_PASSWORD"))
cleanupAuditLog(t, adminPool)
defer cleanupAuditLog(t, adminPool)
store := NewStore(writerPool)
ctx := context.Background()
const n = 20
var wg sync.WaitGroup
errs := make(chan error, n)
for i := 0; i < n; i++ {
wg.Add(1)
go func(i int) {
defer wg.Done()
q := fmt.Sprintf("query-%d", i)
_, err := store.Append(ctx, Entry{TenantID: "default", Source: SourceAPI, EventType: EventQuery, QueryText: &q, Status: StatusSuccess})
errs <- err
}(i)
}
wg.Wait()
close(errs)
for err := range errs {
if err != nil {
t.Fatalf("concurrent Append failed: %v", err)
}
}
result, err := store.VerifyChain(ctx)
if err != nil {
t.Fatalf("VerifyChain: %v", err)
}
if !result.OK {
t.Fatalf("expected an intact chain after %d concurrent appends, got broken at id=%d", n, result.FirstBadID)
}
if result.RowsChecked != n {
t.Fatalf("RowsChecked = %d, want %d", result.RowsChecked, n)
}
}
// TestCheckpointerRun ties Store + FileSink together against real
// audit_log data: writes some rows, checkpoints, writes more, checkpoints
// again, and confirms the second checkpoint picks up exactly where the
// first left off (FromID = previous ToID + 1) with a hash chained off
// the previous checkpoint's hash.
func TestCheckpointerRun(t *testing.T) {
writerPool := testPool(t, "audit_writer", os.Getenv("AUDIT_TEST_POSTGRES_PASSWORD"))
adminPool := testPool(t, "cairnobs", os.Getenv("AUDIT_TEST_ADMIN_PASSWORD"))
cleanupAuditLog(t, adminPool)
defer cleanupAuditLog(t, adminPool)
store := NewStore(writerPool)
sink := NewFileSink(filepath.Join(t.TempDir(), "checkpoints.jsonl"))
checkpointer := NewCheckpointer(store, sink)
ctx := context.Background()
// DELETE doesn't reset the BIGSERIAL sequence, so IDs are not
// guaranteed to start at 1 -- but Checkpoint.FromID is a *cursor
// position* (1, or the previous checkpoint's ToID+1), not "the
// lowest row ID that happens to still exist." In real usage audit_log
// never has gaps (append-only, protected by the immutability
// trigger), so those always coincide; here, this test's own
// destructive cleanupAuditLog between test functions creates a gap
// (rows from earlier tests were deleted, advancing the sequence)
// that real usage never produces -- so FromID is asserted against
// the cursor's own logic (1, since no prior checkpoint exists for
// this fresh FileSink), and ToID against the actual last ID Append
// returned.
var firstBatchLastID int64
for i := 0; i < 3; i++ {
q := "first batch"
rec, err := store.Append(ctx, Entry{TenantID: "default", Source: SourceAPI, EventType: EventQuery, QueryText: &q, Status: StatusSuccess})
if err != nil {
t.Fatalf("Append: %v", err)
}
firstBatchLastID = rec.ID
}
cp1, err := checkpointer.Run(ctx)
if err != nil {
t.Fatalf("first Run: %v", err)
}
if cp1 == nil {
t.Fatalf("expected a checkpoint after 3 rows, got nil")
}
if cp1.FromID != 1 || cp1.ToID != firstBatchLastID {
t.Fatalf("cp1 = %+v, want FromID=1 ToID=%d", cp1, firstBatchLastID)
}
// Nothing new since the last checkpoint -- Run should be a no-op.
noop, err := checkpointer.Run(ctx)
if err != nil {
t.Fatalf("no-op Run: %v", err)
}
if noop != nil {
t.Fatalf("expected nil (nothing new to checkpoint), got %+v", noop)
}
var secondBatchLastID int64
for i := 0; i < 2; i++ {
q := "second batch"
rec, err := store.Append(ctx, Entry{TenantID: "default", Source: SourceAPI, EventType: EventQuery, QueryText: &q, Status: StatusSuccess})
if err != nil {
t.Fatalf("Append: %v", err)
}
secondBatchLastID = rec.ID
}
cp2, err := checkpointer.Run(ctx)
if err != nil {
t.Fatalf("second Run: %v", err)
}
if cp2 == nil {
t.Fatalf("expected a second checkpoint, got nil")
}
if cp2.FromID != cp1.ToID+1 || cp2.ToID != secondBatchLastID {
t.Fatalf("cp2 = %+v, want FromID=%d ToID=%d", cp2, cp1.ToID+1, secondBatchLastID)
}
if cp2.PrevCheckpointHash != cp1.Hash {
t.Fatalf("cp2.PrevCheckpointHash = %q, want %q (chained to cp1)", cp2.PrevCheckpointHash, cp1.Hash)
}
}