A new `api/client/proxy/Client` has been added to interact with
the SSH and gRPC servers that the Proxy serves on its SSH port.
The client will first try connecting to the gRPC server and if
that fails it will fall back to the SSH server.
Much of the SSH functionality mimics the existing behavior of the
`ProxyClient` in `lib/client`. This is the first part of phasing
out that client in favor of the new client. There will be a follow
up PR that migrates `lib/client` to make use of the new client instead.
Part of #19812
Sets `http.Server.IdleTimeout` and `http.Client.IdleConnTimeout`
on clients and servers which didn't have them set. A default of
360s was chosen to be on par with the default of an NLB without
being identical.
This was added as another safety measure to prevent leaking any
idle connections indefinitely as seen in #22757.
`apidefaults.DefaultDialTimeout` was also renamed to
`apidefaults.DefaultIOTimeout` to better reflect its usage.
The `http.Client.Transport` created in `newWebClient` wraps a
chain of `http.RoundTripper` over an underlying `http.Transport`.
Since we cannot guarantee that each `http.RoundTipper` has a
`CloseIdleConnections` method the usage of
`defer clt.CloseIdleConnections()` does not guarantee that the http
connections created during Find/Ping/etc are closed.
To prevent leaking connections implementations of `http.RoundTripper`
have added a `CloseIdleConnections` method added that forwards the
request on to the wrapped interface. The `otelhttp.Transport` does
not implement this method either so care has been taken to wrap it
in a `http.RoundTripper` which will call the root transport in
`enforceCloseIdleConnections`. An upstream issue has been filed
with otelhttp: https://github.com/open-telemetry/opentelemetry-go-contrib/issues/3543
to get the method added to their `Transport` implementation.
This wasn't noticed prior to upgrading to go1.20 becuase prior to
[this](https://github.com/golang/go/commit/4e7e7ae1406c70d9cc0809ec11105a55a60a0b70)
commit the `ReadHeaderTimeout` set on the Proxy web api http.Server
would cause the connection to appear idle and get terminated by the
server.
`TestWebClientClosesIdleConnections` was added to capture the leak
in connections as reported in #22757 and prevent any regressions.
Before this commit, the `tsh` HTTP requests that had the extra headers
were those that did not use `roundtrip`.
This commit leverages `http.RoundTripper.RoundTrip` to ensure that all
requests have the the extra headers.
AttemptDeviceLogin, which is the main entry point for device authentication, now
checks the Ping response and skips the attempt entirely if device trust is
disabled.
The main objective is to avoid a needless roundtrip if the feature is disabled,
as one should only pay for what is in use.
There's actual little consequence in attempting the roundtrip, apart from the
added latency on logins, so I've gone with a negative flag ("Disabled" instead
of "Enabled"). The negative is less harmful if, for some reason, it's wrongly
absent (say, because of some future Ping code branch).
gravitational/teleport.e#514
By only providing the tunnel address from the `reversetunnel.Resolver`
callers would still need to lookup the proxy listener mode to determine
how to dial the address. This results in sending a request to
`/webapi/find` once by the resolver to get the tunnel address and then
a second request to `/webapi/find` by users of the `Resolver` to determine
the proxy listener mode. Propagating the listener mode along with the
tunnel address by the `Resolver` ensures only one `/webapi/find` call
is needed.
This is especially impactful because the `reversetunnel.TunnelAuthDialer`
which is used by the auth http client would do this everytime the
`http.Client` connection pool was empty. When the `http.Client` needed
to dial the auth server it was incurring the additional roundtrip to the
proxy.
Update metalinter, fix a few lint warnings and replace deprecated linters.
`deadcode`, `structcheck` and `varcheck` are abandoned and now replaced by [`unused`][1].
Since 1.19, `go fmt` reformats godocs according to https://go.dev/doc/comment. I've done a bulk-reformatting of the codebase to keep the linter happy. Backporting is mostly harmless (the exception being `lib/services/role_test.go`, that for some reason breaks the _old_ linter using the new format).
[1]: https://golangci-lint.run/usage/linters/
* Bump golangci-lint version
* Replace abandoned linters
* Fix bodyclose on lib/auth/github.com
* Fix bodyclose on lib/kube/proxy/streamproto/proto_test.go
* Fix bodyclose on lib/srv/alpnproxy/proxy_test.go
* Fix bodyclose on lib/web/conn_upgrade_test.go
* Silence staticcheck on lib/kube/proxy/forwarder_test.go
* Silence staticcheck on lib/utils/certs_test.go
* Address BuildNameToCertificate deprecation warnings
* Run `go fmt ./...`
* Run `go fmt ./...` on api/
* Ignore formatting in role_test.go
* Remove redundant initializers in lib/srv/uacc/
* Update e/
* Add spans to TeleportClient/ProxyClient/NodeSession
* Improve dialer tracing
* Alter default http client span name to include path
* propagate tracing options in NewClientConn
* Respect timeout in `(directDial).DialTimeout`
* Add a Timeout to the api/client/webclient calls
* Respect timeout in (proxyDial).DialTimeout
* Applied suggestion
Co-authored-by: Alan Parra <alan.parra@goteleport.com>
Co-authored-by: Alan Parra <alan.parra@goteleport.com>
Leverage otelhttp to generate spans for http clients and http handlers.
Additionally add spans for grpc clients and servers by using the
otelgrpc interceptors. This is step one in implementation of RFD 65.
Until further work is done, the spans generated will be a no-op
until the tracing provider is initialized.
Implement passwordless cluster settings, as described by RFD 52: Passwordless
[1]. See https://github.com/gravitational/teleport/pull/11716 for recent RFD
changes.
Add the "passwordless" connector and allows cluster admins to configure it as
the cluster default.
teleport.yaml:
```yaml
auth_service:
authentication:
type: local
second_factor: on
webauthn:
rp_id: example.com
passwordless: true # enables/disables passwordless
connector_name: passwordless # sets passwordless as the default
```
`tsh` obeys the connector name if present. User overrides are possible via `tsh
--auth=local` or `tsh --auth=passwordless`. The `tsh --pwdless` flag is now
removed, as discussed in recent RFD changes.
Web API `/ping` and `/ping/{connector}` endpoints are changed to support
"passwordless" and to provide other necessary information.
#9160
[1] https://github.com/gravitational/teleport/blob/master/rfd/0052-passwordless.md#cluster-settings
* Add passwordless toggle to AuthPreferenceSpecV2 proto
* Update generate protos
* Add AuthPreference fields and validation
* Reserve system connector names
* Distinguish between type=local and local connector
* Add passwordless option to teleport.yaml
* Do /ping and /ping/{connector} changes
* Make tsh respect the passwordless connector, remove --pwdless
* Include the possible connector values if supplied value is invalid.
* Update messaging for auth flag.
* Return error message from failed ping response.
* Collect the connector names in the existing loops.
This change updates NO_PROXY handling to allow blocking specific host:port combinations, rather than just the host. It also adds a special case for downgrading requests to plain HTTP when --insecure is true and the request goes through a plain HTTP proxy at localhost (i.e. HTTP_PROXY=http://localhost).
Give Teleport clients a consistent hint of which MFA method they should
favor when facing multiple options.
To be used by tsh and Web UI to decide whether they should refer to
"U2F" or "WebAuthn" during the transitional period.
* Return preferred MFA method on ping endpoints
Add Webauthn support to a few outstanding endpoints and RPCs.
* Refactor and add Webauthn support to ChangePassword
* Add explicit JSON tag to AuthenticateUserRequest.Webauthn
* Add Webauthn support to /webapi/users/password/token
* Add Webauthn to /webapi/ping
* Add default Webauthn device name to ChangeUserAuthentication
* Tidy Ping tests
Using the standard logger results in log messages being routed to the
console even in non-debug mode, so reverting to `logrus` which will
honour the debug setting
This patch will allow the web api to (optionally) fall back from HTTPS
to HTTP under very particular circumstances:
1. The appropriate `insecure` flag was set, and
2. The target host is the loopback
If both conditions are met, this patch will allow the WebAPI client to
fall back to plain HTTP when attempting to login to the auth server.
Adds a configurable MotD to the auth service. If a message is set, any
attempt to login into a cluster via `tsh` will display that message and
not begin to log the user in until that message is acknowledged.
rosstimothy