AGENTS.md: Update for hardened search, bind, and panic handling

This commit is contained in:
Michael Aldridge 2026-11-01 20:03:07 -06:00
parent 3cbc9652ea
commit e7b4c70bab

135
AGENTS.md
View file

@ -8,47 +8,53 @@ NetAuth RPC calls via the `netauth/protocol` protobuf types. It is not a full
LDAP server and never stores data; it only reshapes and serves live NetAuth
data. Single static binary (`ldapd`), intended to run bound to localhost.
There are **no tests anywhere** in this repo (no `*_test.go` files). Do not
assume `go test ./...` validates anything.
## Commands
```sh
go build ./... # build
go build ./...
go vet ./...
go run . # run locally (reads config like the real binary, see below)
go test ./... # or: go test -race ./internal/ldap/
go run . # needs a config file or NETAUTH_LDAP_DOMAIN
go mod vendor # repo vendors deps; vendor/ and dist/ are gitignored
# Production-style build (matches Dockerfile)
CGO_ENABLED=0 GOOS=linux go build -a -ldflags '-extldflags "-static"' -o ldapd .
```
Release is via [goreleaser](.goreleaser.yml) triggered by `v*` tags in
[.github/workflows/release.yml](.github/workflows/release.yml); the before-hooks
CI ([.github/workflows/test.yml](.github/workflows/test.yml)) runs build, vet
and `go test -race` on pushes to master and on PRs. Release is via
[goreleaser](.goreleaser.yml) triggered by `v*` tags in
[.github/workflows/release.yml](.github/workflows/release.yml); its before-hooks
run `go mod vendor` and `scripts/prepare.sh` (regenerates `NOTICE` from vendored
license files via `scripts/vendor-licenses`, which needs the vendor tree present).
Note `dist/config.yaml` is a generated/stale goreleaser artifact, not the config
actually used.
license files via `scripts/vendor-licenses`, which needs the vendor tree
present). Note `dist/config.yaml` is a stale generated artifact, not the config
actually used — the real one is `.goreleaser.yml` at the repo root.
## Layout
```
main.go # wiring only: config, logging, client, serve
internal/ldap/ # the LDAP server (package ldap)
ldap.go # New(), routing, SetDomain(), Serve/ServeTLS
ldap.go # New(), route (re)building, Serve/ServeTLS, SetDomain
handler.go # panic guard wrapper + per-op failure responders
conn.go # per-connection bind state (bound/markBound/markUnbound)
type.go # server struct + naClient interface (the seam for fakes)
option.go # functional options: WithLogger, WithNetAuth, WithAnonBind
bind.go # bind handler (auth against NetAuth)
search.go # RootDSE, entity & group search handlers
attributes.go # RFC 4511 attribute selection matching
bleve_mapper.go # LDAP filter -> Bleve query string translation
util.go # DN parsing (entityIDFromDN)
internal/buildinfo/ # Version var ("dev"), advertised in RootDSE
*_test.go # tests; see "Testing" below
internal/buildinfo/ # Version var, advertised in RootDSE; set by goreleaser
```
Architecture: `main` builds a `viper` config + `netauth` client, constructs the
server with functional options, and calls `Serve`/`ServeTLS`. The underlying
LDAP engine is a `RouteMux` (net/http-style routing on LDAP ops) with handlers
keyed by base DN and scope.
keyed by base DN and scope. Every handler is wrapped by `server.guard` so a
panic cannot take the process down (the vendored server dispatches requests in
goroutines with no recovery of its own).
## Non-obvious gotchas
@ -56,56 +62,77 @@ keyed by base DN and scope.
`github.com/ps78674/ldapserver` + `github.com/ps78674/goldap`, not the more
common `nwaples/ldapserver` / `go-ldap` packages. Import as
`ldap "github.com/ps78674/ldapserver"`. Check the vendored API before
assuming upstream signatures.
- **Routes depend on the domain**: `New()` only registers NotFound/Abandon/Bind
and the RootDSE route. The `ou=entities` / `ou=groups` search routes are
registered as a side effect of `SetDomain()`, which **must be called** after
`New()` or searches won't route at all. Calling it twice re-registers routes.
- **Base DN construction**: `SetDomain` takes a plain domain (`example.com`),
splits on `.`, prepends `dc=netauth`, yielding `dc=netauth,dc=example,dc=com`.
Tree layout under it is fixed: `uid=<id>,ou=entities,<base>` and
`cn=<name>,ou=groups,<base>`.
- **Supported LDAP filters are extremely limited** (`bleve_mapper.go`): only
equality match, OR, and attribute-present filters; only the attributes `uid`
(→ `ID:`) and `cn` (→ `Name:`); values are passed through as raw Bleve query
strings to NetAuth's search. Everything else returns
`unwillingToPerform`/"Filter type not supported".
- **Anonymous bind fall-through** (`bind.go`): when `allowAnon` is enabled and
the bind DN is empty, a success response is written but the handler **does
not return** — execution continues into `entityIDFromDN("")`, which errors
and writes a second response. This appears to be a bug; be careful not to
"fix" it blindly if behavior parity matters, and definitely add a response
write guard if you touch it.
- **Only simple bind** — no SASL; anything else is rejected outright.
- **DN parsing known defect**: `entityIDFromDN` splits on `,` without handling
escaped commas (documented as acceptable since NetAuth IDs can't contain
commas). Error strings there have typos ("is underneath", "must be rooted");
leave them alone unless asked.
- **Search fan-out**: each entity result triggers an extra `EntityGroups` RPC
for `memberOf`, and each group result triggers `GroupMembers` for `member`.
A broad search is N+1 RPCs against NetAuth. Group expansion is done
server-side by NetAuth (flattened, fully expanded groups).
assuming upstream signatures. The fork has real limits: TLS floor is
SSL 3.0, `Server.Handle` calls `log.Fatalln` on double registration, and
there is no panic recovery in the request path.
- **The server struct is its own Handler**: `New()` registers `*server` once
(the fork forbids re-registration), and `ServeLDAP` delegates to whatever
`s.routes` mux is current. `SetDomain` rebuilds the mux in place, which is
why it is safe to call repeatedly. Never call `x.Handle()` again.
- **SetDomain is required and validated**: returns an error for an empty
domain or empty labels; searches do not route until it succeeds. Naming
context is always `dc=netauth,dc=<label>,...`. Tree layout under it is fixed:
`uid=<id>,ou=entities,<base>` and `cn=<name>,ou=groups,<base>`.
- **Bind state is per connection**: searches (except RootDSE) are refused with
`insufficientAccessRights` until a successful bind on the same connection
(`conn.go`). A failed bind clears the state per RFC 4511. Anonymous binds
succeed only when `WithAnonBind(true)` and also mark the connection bound.
- **Filter support** (`bleve_mapper.go`): equality, present, AND, OR, NOT
(nested only) and simple substrings (`js*`, `*smith`, `a*th`) are supported;
only `uid` (→ `ID`) and `cn` (→ `Name`). Everything else errors rather than
returning wrong results. Rules worth knowing before touching this file:
- Plain values are quoted; wildcard values must match `^[\w.\-*]+$` to
prevent Bleve query injection.
- A top-level (or all-negative) filter errors instead of silently returning
zero results, because Bleve cannot anchor a query made only of `-` clauses.
- AND branches become `+term`; negated branches stay `-term` un-prefixed.
- **DN parsing**: `entityIDFromDN` splits on `,` without handling escaped
commas (accepted defect: NetAuth IDs cannot contain commas). It validates
the DN has enough components and that the naming context matches exactly.
- **Search fan-out**: entity results include `memberOf` and group results
include `member`, each costing one extra NetAuth RPC per entry. These are
only fetched when the client's attribute selection asks for them
(`attributes.go`); an empty selection or `*` means all.
- **Logging is dual-library**: app logging uses `hashicorp/go-hclog`; the
vendored ldapserver uses `logrus`, and `main.go` hijacks the global logrus
logger to funnel protocol traces into hclog at forced Trace level via
`NETAUTH_LOGLEVEL`. Use `s.l` (hclog, keyed with `"key", value` pairs) inside
the server package.
- **NetAuth dependency seam**: server code talks to NetAuth only through the
`naClient` interface in `type.go:12`. If you add a NetAuth call, add it to
that interface, not directly to the concrete client.
- **Version injection mismatch**: goreleaser ldflags set `main.version` etc.,
but the RootDSE reports `internal/buildinfo.Version`, which stays `"dev"` in
built binaries.
`naClient` interface in `type.go`. If you add a NetAuth call, add it to that
interface, not directly to the concrete client.
## Testing
Tests live in `internal/ldap/*_test.go`. Two constraints shape them:
- **goldap types are unconstructible from outside**: the message/filter types
have unexported fields and no exported constructors. Tests hand-encode BER
and run it through the real decoder; see the `ber*` helpers in
`testutil_test.go` (`berBindRequest`, `berSearchRequestWithAttrs`,
`berEqualityFilter`, `berSubstringFilter`, ...). Follow that pattern rather
than trying to build request structs directly.
- **Result codes have no getters**: the result types expose setters only, so
`resultCode()` in `result_test.go` re-encodes a response and parses the code
back out of the DER.
- **Fakes**: implement `naClient` (`fakeClient` in `bind_test.go` records
calls and returns canned protobuf data). Per-connection state is faked by
setting `s.conn` to return a `fakeConn` (`boundTestServer` in
`search_test.go`). Search tests must bind before searching.
- Handlers take the `ldap.ResponseWriter` interface, so assertions use
`recordingWriter` (handler_test.go), which captures the exact response
sequence.
## Configuration
viper-driven; config file `config.{yaml,...}` searched in `/etc/netauth/`,
`$HOME/.netauth/`, `.` — **a missing config file is fatal** (exit 5), even with
all values provided by env. Env vars use the `NETAUTH_` prefix with `_`
replacing `.` (e.g. `NETAUTH_LDAP_BIND`, `NETAUTH_LDAP_DOMAIN`,
`NETAUTH_LOGLEVEL`). Known keys (defaults in `main.go:16-22`):
`ldap.bind` (localhost:389), `ldap.tls` (false), `ldap.key`, `ldap.cert`,
`ldap.allow_anon` (false), `ldap.domain` (no default — required).
`$HOME/.netauth/`, `.`. A missing config file is tolerated (env-only
deployments); a corrupt one is fatal (exit 5). Env vars use the `NETAUTH_`
prefix with `_` replacing `.` (e.g. `NETAUTH_LDAP_BIND`,
`NETAUTH_LDAP_DOMAIN`, `NETAUTH_LOGLEVEL`). Known keys (defaults in
`main.go`): `ldap.bind` (localhost:389), `ldap.tls` (false), `ldap.key`,
`ldap.cert`, `ldap.allow_anon` (false), `ldap.domain` (no default — required;
startup exits with 3 if invalid).
## Style notes
@ -114,3 +141,5 @@ replacing `.` (e.g. `NETAUTH_LDAP_BIND`, `NETAUTH_LDAP_DOMAIN`,
- Handlers are methods on `*server` named `handle<X>`; response pattern is
construct response → `w.Write(res)` on every exit path.
- Loop style favors `for i := range x` throughout.
- Errors returned to LDAP clients go in diagnostic messages; keep them
non-technical, they are shown verbatim by LDAP browsers.