Files
cairnobs/enterprise/cmd/enterprise-auth/main.go
T
jcoffey-dev 08a90a27aa Build SAML login (enterprise/internal/loginhandler), mirroring OIDC
Adds GET /auth/saml/login + POST /auth/saml/acs alongside the existing
OIDC pair, both converging on the same upsert-user/resolve-tenant/
issue-session path. loginhandler.New now takes an optional
*saml.ServiceProvider, RegisterRoutes registers each protocol's routes
independently so either, both, or neither can be configured. SAML's
replay/unsolicited-response defense (InResponseTo, standing in for
OIDC's state) is carried via a SameSite=None sentry_saml_request cookie
-- None because the ACS endpoint receives a cross-site POST from the
IdP's origin, which SameSite=Lax cookies are never sent on.
enterprise-auth's main.go now fetches+parses SAML_IDP_METADATA_URL at
startup (samlsp.FetchMetadata) and wires the result through.

Verified to the same bar as OIDC: a real fake IdP
(crewjam/saml/samlidp, genuine XML signing/verification) drives the
full login->ACS->session-cookie round trip and negative paths (bad
InResponseTo, missing request cookie, missing email/NameID, no/multiple
tenant memberships), all in loginhandler/saml_test.go, no Docker
needed. The login-form HTML is bypassed by pre-seeding a saml.Session
directly into samlidp's session store and presenting the matching
`session` cookie -- an IdP-supported shortcut (confirmed by reading
GetSession), the same "skip the UI, keep the crypto real" approach
oidctest gave the OIDC tests.

Writing that test caught two real bugs in internal/saml.ParseResponse,
both fixed here: it never called r.ParseForm() before reading the
POSTed SAMLResponse field, so every real ACS POST would have silently
decoded an empty response; and its email-attribute matching missed
urn:oid:0.9.2342.19200300.100.1.3 (the standard LDAP "mail" OID), which
is what an IdP sends by default absent an explicit
AttributeConsumingService request for "email" -- exactly what
samlidp's own DefaultAssertionMaker does, and plausibly what real IdPs'
default SAML app templates do too.

Docs (CLAUDE.md, threat-model.md, architecture.md, enterprise/README.md,
phase-4-runbook.md, docker-compose.yml's enterprise-auth comment)
updated in lockstep: SAML login moves from "protocol mechanics only" to
"built, verified with a real fake IdP, not yet tried against a real
external IdP or a running enterprise-auth container" -- the same
disclosed gap OIDC already carried.
2026-08-14 06:39:36 -07:00

201 lines
6.9 KiB
Go

// Command enterprise-auth is Sentry's SSO/tenant-provisioning/RBAC
// service (commercial license, not AGPL) -- see
// /docs/phase-4-isolation-design.md and /docs/phase-4-rbac-design.md.
//
// Wires session issuance/validation (internal/session), the
// POST /internal/authorize endpoint api/authz.HTTPAuthorizer calls (the
// piece that turns on RBAC enforcement in /api), and --
// internal/loginhandler -- the real GET /auth/{oidc,saml}/login and
// GET /auth/oidc/callback + POST /auth/saml/acs handlers that issue a
// *human* session after an actual IdP round trip, resolving tenant/role
// via internal/rbacstore. Both protocols are now fully wired -- OIDC via
// discovery, SAML via fetching+parsing SAML_IDP_METADATA_URL at startup
// (crewjam/saml's samlsp.FetchMetadata; a trusted operator-supplied URL,
// same trust level as OIDC_ISSUER_URL's discovery fetch, not
// end-user-controlled input). Also fully wired: -mint-service-token
// (the RoleService credential /alerting presents).
package main
import (
"context"
"flag"
"fmt"
"log/slog"
"net/http"
"net/url"
"os"
"os/signal"
"strings"
"syscall"
"time"
"github.com/crewjam/saml/samlsp"
"github.com/jackc/pgx/v5/pgxpool"
"github.com/sentry/sentry/enterprise/internal/authhandler"
"github.com/sentry/sentry/enterprise/internal/config"
"github.com/sentry/sentry/enterprise/internal/loginhandler"
"github.com/sentry/sentry/enterprise/internal/oidc"
"github.com/sentry/sentry/enterprise/internal/rbacstore"
samlpkg "github.com/sentry/sentry/enterprise/internal/saml"
"github.com/sentry/sentry/enterprise/internal/session"
)
func main() {
logger := slog.New(slog.NewJSONHandler(os.Stdout, nil))
cfg, err := config.Load()
if err != nil {
logger.Error("loading config", "error", err)
os.Exit(1)
}
// -mint-service-token issues a RoleService credential and prints it
// to stdout, then exits -- an operator bootstrap step (run once,
// paste the output into /alerting's API_SERVICE_TOKEN), not an HTTP
// endpoint. Minting a service token has no session/cookie to check
// like a human login flow would, so this is deliberately an offline
// operator action gated by access to enterprise-auth's own
// environment/secrets, not a network-reachable endpoint.
mintServiceToken := flag.String("mint-service-token", "", "mint a RoleService credential for the named caller (e.g. \"alerting\") and exit")
// -healthcheck: same self-check mode as api/-healthcheck (see that
// binary's doc comment) -- enterprise-auth's image is distroless too.
healthcheck := flag.Bool("healthcheck", false, "self-check mode for Docker's HEALTHCHECK")
flag.Parse()
if *healthcheck {
os.Exit(runHealthcheck(cfg.HTTPListenAddr))
}
sessionManager, err := session.NewManager(cfg.SessionSigningKey)
if err != nil {
logger.Error("constructing session manager", "error", err)
os.Exit(1)
}
if *mintServiceToken != "" {
token, err := sessionManager.IssueServiceToken(*mintServiceToken)
if err != nil {
logger.Error("minting service token", "error", err)
os.Exit(1)
}
fmt.Println(token)
return
}
ctx, stop := signal.NotifyContext(context.Background(), syscall.SIGINT, syscall.SIGTERM)
defer stop()
pgDSN := fmt.Sprintf("postgres://%s:%s@%s/%s", cfg.Postgres.Username, cfg.Postgres.Password, cfg.Postgres.Addr, cfg.Postgres.Database)
pgPool, err := pgxpool.New(ctx, pgDSN)
if err != nil {
logger.Error("opening postgres pool", "error", err)
os.Exit(1)
}
defer pgPool.Close()
if err := pgPool.Ping(ctx); err != nil {
logger.Error("pinging postgres", "error", err)
os.Exit(1)
}
rbac := rbacstore.NewStore(pgPool)
// oidcProvider stays nil (loginhandler.RegisterRoutes then registers
// nothing) unless OIDC is actually configured -- matches every other
// optional-config path in this codebase.
var oidcProvider *oidc.Provider
if cfg.OIDC.IssuerURL != "" {
oidcProvider, err = oidc.New(ctx, oidc.Config{
IssuerURL: cfg.OIDC.IssuerURL, ClientID: cfg.OIDC.ClientID,
ClientSecret: cfg.OIDC.ClientSecret, RedirectURL: cfg.OIDC.RedirectURL,
Scopes: []string{"email", "profile"},
})
if err != nil {
logger.Error("discovering OIDC issuer", "error", err)
os.Exit(1)
}
logger.Info("OIDC provider configured", "issuer", cfg.OIDC.IssuerURL)
} else {
logger.Info("OIDC not configured (OIDC_ISSUER_URL unset) -- skipping discovery, /auth/oidc/* routes disabled")
}
// samlProvider stays nil (loginhandler.RegisterRoutes then registers
// nothing) unless SAML is actually configured -- same shape as OIDC
// above.
var samlProvider *samlpkg.ServiceProvider
if cfg.SAML.IDPMetadataURL != "" {
metadataURL, err := url.Parse(cfg.SAML.IDPMetadataURL)
if err != nil {
logger.Error("parsing SAML_IDP_METADATA_URL", "error", err)
os.Exit(1)
}
idpMetadata, err := samlsp.FetchMetadata(ctx, http.DefaultClient, *metadataURL)
if err != nil {
logger.Error("fetching SAML IdP metadata", "error", err)
os.Exit(1)
}
samlProvider, err = samlpkg.New(samlpkg.Config{
EntityID: cfg.SAML.EntityID, ACSURL: cfg.SAML.ACSURL, IDPMetadata: idpMetadata,
})
if err != nil {
logger.Error("constructing SAML service provider", "error", err)
os.Exit(1)
}
logger.Info("SAML provider configured", "idp_metadata_url", cfg.SAML.IDPMetadataURL)
} else {
logger.Info("SAML not configured (SAML_IDP_METADATA_URL unset) -- /auth/saml/* routes disabled")
}
mux := http.NewServeMux()
mux.HandleFunc("GET /healthz", func(w http.ResponseWriter, _ *http.Request) {
w.WriteHeader(http.StatusOK)
})
features := authhandler.Features{
OIDCEnabled: cfg.OIDC.IssuerURL != "",
SAMLEnabled: cfg.SAML.IDPMetadataURL != "",
}
authhandler.New(logger, sessionManager, features).RegisterRoutes(mux)
loginhandler.New(logger, oidcProvider, samlProvider, sessionManager, rbac, cfg.PostLoginRedirectURL).RegisterRoutes(mux)
srv := &http.Server{Addr: cfg.HTTPListenAddr, Handler: mux}
errCh := make(chan error, 1)
go func() {
logger.Info("enterprise-auth listening", "addr", cfg.HTTPListenAddr)
errCh <- srv.ListenAndServe()
}()
select {
case <-ctx.Done():
shutdownCtx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
defer cancel()
if err := srv.Shutdown(shutdownCtx); err != nil {
logger.Error("graceful shutdown failed", "error", err)
}
case err := <-errCh:
if err != nil && err != http.ErrServerClosed {
logger.Error("server exited with error", "error", err)
os.Exit(1)
}
}
}
// runHealthcheck mirrors api/cmd/api/main.go's runHealthcheck exactly --
// see that function's doc comment for why this execs the binary against
// itself rather than using an external tool.
func runHealthcheck(listenAddr string) int {
addr := listenAddr
if strings.HasPrefix(addr, ":") {
addr = "localhost" + addr
}
client := http.Client{Timeout: 3 * time.Second}
resp, err := client.Get("http://" + addr + "/healthz")
if err != nil {
return 1
}
defer resp.Body.Close()
if resp.StatusCode != http.StatusOK {
return 1
}
return 0
}