fleet/server/service/base_client.go

281 lines
7.5 KiB
Go
Raw Normal View History

package service
import (
"crypto/tls"
"crypto/x509"
"encoding/json"
"errors"
"fmt"
"io"
"mime"
"net/http"
"net/url"
add headers denoting capabilities between fleet server / desktop / orbit (#7833) This adds a new mechanism to allow us to handle compatibility issues between Orbit, Fleet Server and Fleet Desktop. The general idea is to _always_ send a custom header of the form: ``` fleet-capabilities-header = "X-Fleet-Capabilities:" capabilities capabilities = capability * (,) capability = string ``` Both from the server to the clients (Orbit, Fleet Desktop) and vice-versa. For an example, see: https://github.com/fleetdm/fleet/commit/8c0bbdd291f54e03e19766bcdfead0fb8067f60c Also, the following applies: - Backwards compat: if the header is not present, assume that orbit/fleet doesn't have the capability - The current capabilities endpoint will be removed ### Motivation This solution is trying to solve the following problems: - We have three independent processes communicating with each other (Fleet Desktop, Orbit and Fleet Server). Each process can be updated independently, and therefore we need a way for each process to know what features are supported by its peers. - We originally implemented a dedicated API endpoint in the server that returned a list of the capabilities (or "features") enabled, we found this, and any other server-only solution (like API versioning) to be insufficient because: - There are cases in which the server also needs to know which features are supported by its clients - Clients needed to poll for changes to detect if the capabilities supported by the server change, by sending the capabilities on each request we have a much cleaner way to handling different responses. - We are also introducing an unauthenticated endpoint to get the server features, this gives us flexibility if we need to implement different authentication mechanisms, and was one of the pitfalls of the first implementation. Related to https://github.com/fleetdm/fleet/issues/7929
2022-09-26 10:53:53 +00:00
"os"
"path/filepath"
"strings"
"github.com/fleetdm/fleet/v4/pkg/fleethttp"
add headers denoting capabilities between fleet server / desktop / orbit (#7833) This adds a new mechanism to allow us to handle compatibility issues between Orbit, Fleet Server and Fleet Desktop. The general idea is to _always_ send a custom header of the form: ``` fleet-capabilities-header = "X-Fleet-Capabilities:" capabilities capabilities = capability * (,) capability = string ``` Both from the server to the clients (Orbit, Fleet Desktop) and vice-versa. For an example, see: https://github.com/fleetdm/fleet/commit/8c0bbdd291f54e03e19766bcdfead0fb8067f60c Also, the following applies: - Backwards compat: if the header is not present, assume that orbit/fleet doesn't have the capability - The current capabilities endpoint will be removed ### Motivation This solution is trying to solve the following problems: - We have three independent processes communicating with each other (Fleet Desktop, Orbit and Fleet Server). Each process can be updated independently, and therefore we need a way for each process to know what features are supported by its peers. - We originally implemented a dedicated API endpoint in the server that returned a list of the capabilities (or "features") enabled, we found this, and any other server-only solution (like API versioning) to be insufficient because: - There are cases in which the server also needs to know which features are supported by its clients - Clients needed to poll for changes to detect if the capabilities supported by the server change, by sending the capabilities on each request we have a much cleaner way to handling different responses. - We are also introducing an unauthenticated endpoint to get the server features, this gives us flexibility if we need to implement different authentication mechanisms, and was one of the pitfalls of the first implementation. Related to https://github.com/fleetdm/fleet/issues/7929
2022-09-26 10:53:53 +00:00
"github.com/fleetdm/fleet/v4/server/fleet"
"github.com/google/uuid"
)
var errInvalidScheme = errors.New("address must start with https:// for remote connections")
// httpClient interface allows the HTTP methods to be mocked.
type httpClient interface {
Do(req *http.Request) (*http.Response, error)
}
type baseClient struct {
baseURL *url.URL
http httpClient
urlPrefix string
insecureSkipVerify bool
add headers denoting capabilities between fleet server / desktop / orbit (#7833) This adds a new mechanism to allow us to handle compatibility issues between Orbit, Fleet Server and Fleet Desktop. The general idea is to _always_ send a custom header of the form: ``` fleet-capabilities-header = "X-Fleet-Capabilities:" capabilities capabilities = capability * (,) capability = string ``` Both from the server to the clients (Orbit, Fleet Desktop) and vice-versa. For an example, see: https://github.com/fleetdm/fleet/commit/8c0bbdd291f54e03e19766bcdfead0fb8067f60c Also, the following applies: - Backwards compat: if the header is not present, assume that orbit/fleet doesn't have the capability - The current capabilities endpoint will be removed ### Motivation This solution is trying to solve the following problems: - We have three independent processes communicating with each other (Fleet Desktop, Orbit and Fleet Server). Each process can be updated independently, and therefore we need a way for each process to know what features are supported by its peers. - We originally implemented a dedicated API endpoint in the server that returned a list of the capabilities (or "features") enabled, we found this, and any other server-only solution (like API versioning) to be insufficient because: - There are cases in which the server also needs to know which features are supported by its clients - Clients needed to poll for changes to detect if the capabilities supported by the server change, by sending the capabilities on each request we have a much cleaner way to handling different responses. - We are also introducing an unauthenticated endpoint to get the server features, this gives us flexibility if we need to implement different authentication mechanisms, and was one of the pitfalls of the first implementation. Related to https://github.com/fleetdm/fleet/issues/7929
2022-09-26 10:53:53 +00:00
// serverCapabilities is a map of capabilities that the server supports.
// This map is updated on each response we receive from the server.
serverCapabilities fleet.CapabilityMap
// clientCapabilities is a map of capabilities that the client supports.
// This list is given when the client is instantiated and shouldn't be
// modified afterwards.
clientCapabilities fleet.CapabilityMap
}
// parseResponse processes the status code and parses the response body.
// It does not close the response body (should be closed by the caller).
func (bc *baseClient) parseResponse(verb, path string, response *http.Response, responseDest interface{}) error {
switch response.StatusCode {
case http.StatusNotFound:
return notFoundErr{
msg: extractServerErrorText(response.Body),
}
case http.StatusUnauthorized:
errText := extractServerErrorText(response.Body)
if strings.Contains(errText, "password reset required") {
return ErrPasswordResetRequired
}
End-user authentication for Window/Linux setup experience: agent (#34847) <!-- Add the related story/sub-task/bug number, like Resolves #123, or remove if NA --> **Related issue:** Resolves #34528 # Details This PR implements the agent changes for allowing Fleet admins to require that users authenticate with an IdP prior to having their devices set up. I'll comment on changes inline but the high-level is: 1. Orbit calls the enroll endpoint as usual. This is triggered lazily by any one of a number of subsystems like device token rotation or requesting Fleet config 2. If the enroll endpoint returns the new `ErrEndUserAuthRequired` response, then it opens a window to the `/mdm/sso` Fleet page and retries the enroll endpoint every 30 seconds indefinitely. 3. Any other non-200 response to the enroll request is treated as before (limited # of retries, with backoff) # Checklist for submitter If some of the following don't apply, delete the relevant line. - [ ] Changes file added for user-visible changes in `changes/`, `orbit/changes/` or `ee/fleetd-chrome/changes`. See [Changes files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing- changes.md#changes-files) for more information. Will add changelog when story is one. ## Testing - [X] Added/updated automated tests Added test for new retry logic - [X] QA'd all new/changed functionality manually This is kinda hard to test without the associated backend PR: https://github.com/fleetdm/fleet/pull/34835 ## fleetd/orbit/Fleet Desktop - [X] Verified compatibility with the latest released version of Fleet (see [Must rule](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/workflows/fleetd-development-and-release-strategy.md)) This is compatible with all Fleet versions, since older ones won't send the new error. - [X] If the change applies to only one platform, confirmed that `runtime.GOOS` is used as needed to isolate changes This is compatible with all platforms, although it currently should only ever run on Windows and Linux since macOS devices will have end-user auth taken care of before they even download Orbit. - [ ] Verified that fleetd runs on macOS, Linux and Windows Testing this now. - [ ] Verified auto-update works from the released version of component to the new version (see [tools/tuf/test](../tools/tuf/test/README.md)) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added SSO (Single Sign-On) enrollment support for end-user authentication * Enhanced error messaging for authentication-required scenarios * **Bug Fixes** * Improved error handling and retry logic for enrollment failures <!-- end of auto-generated comment: release notes by coderabbit.ai -->
2025-11-03 22:41:57 +00:00
if strings.Contains(errText, "END_USER_AUTH_REQUIRED") {
return ErrEndUserAuthRequired
}
return ErrUnauthenticated
case http.StatusPaymentRequired:
return ErrMissingLicense
default:
if response.StatusCode >= 200 && response.StatusCode < 300 {
break
}
e := &statusCodeErr{
code: response.StatusCode,
body: extractServerErrorText(response.Body),
}
return fmt.Errorf("%s %s received status %w", verb, path, e)
}
bc.setServerCapabilities(response)
if responseDest != nil {
if e, ok := responseDest.(bodyHandler); ok {
if err := e.Handle(response); err != nil {
return fmt.Errorf("%s %s error with custom body handler contents: %w", verb, path, err)
}
} else if response.StatusCode != http.StatusNoContent {
b, err := io.ReadAll(response.Body)
if err != nil {
return fmt.Errorf("reading response body: %w", err)
}
if err := json.Unmarshal(b, &responseDest); err != nil {
Improved orbit debug logs when response contains a large HTML page. (#33195) Resolves #33219 Note: this only fixes orbit. The issue remains on osquery: [#33019](https://github.com/fleetdm/fleet/issues/33019) # Checklist for submitter - [x] Changes file added for user-visible changes in `changes/`, `orbit/changes/` or `ee/fleetd-chrome/changes`. See [Changes files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files) for more information. ## Testing - [x] Added/updated automated tests - [x] QA'd all new/changed functionality manually ## fleetd/orbit/Fleet Desktop - [x] Verified compatibility with the latest released version of Fleet (see [Must rule](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/workflows/fleetd-development-and-release-strategy.md)) - [x] Verified that fleetd runs on macOS, Linux and Windows - [x] Verified auto-update works from the released version of component to the new version (see [tools/tuf/test](../tools/tuf/test/README.md)) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - Bug Fixes - Improved error messages when servers return HTML instead of JSON. - Truncates oversized responses in logs to prevent overwhelming output while preserving context. - More robust parsing of non-JSON error responses. - Documentation - Added changelog entry noting enhanced debug logging for large HTML responses. - Tests - Added tests covering HTML, plain text, empty, long, and invalid JSON error bodies to validate error message handling. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
2025-09-19 22:00:19 +00:00
const maxBodyLen = 200
truncatedBytes, isHTML := truncateAndDetectHTML(b, maxBodyLen)
if isHTML {
return fmt.Errorf("decode %s %s response: %w, (server returned HTML instead of JSON), body: %s", verb, path, err, truncatedBytes)
}
return fmt.Errorf("decode %s %s response: %w, body: %s", verb, path, err, truncatedBytes)
}
if e, ok := responseDest.(fleet.Errorer); ok {
if e.Error() != nil {
return fmt.Errorf("%s %s error: %w", verb, path, e.Error())
}
}
}
}
add headers denoting capabilities between fleet server / desktop / orbit (#7833) This adds a new mechanism to allow us to handle compatibility issues between Orbit, Fleet Server and Fleet Desktop. The general idea is to _always_ send a custom header of the form: ``` fleet-capabilities-header = "X-Fleet-Capabilities:" capabilities capabilities = capability * (,) capability = string ``` Both from the server to the clients (Orbit, Fleet Desktop) and vice-versa. For an example, see: https://github.com/fleetdm/fleet/commit/8c0bbdd291f54e03e19766bcdfead0fb8067f60c Also, the following applies: - Backwards compat: if the header is not present, assume that orbit/fleet doesn't have the capability - The current capabilities endpoint will be removed ### Motivation This solution is trying to solve the following problems: - We have three independent processes communicating with each other (Fleet Desktop, Orbit and Fleet Server). Each process can be updated independently, and therefore we need a way for each process to know what features are supported by its peers. - We originally implemented a dedicated API endpoint in the server that returned a list of the capabilities (or "features") enabled, we found this, and any other server-only solution (like API versioning) to be insufficient because: - There are cases in which the server also needs to know which features are supported by its clients - Clients needed to poll for changes to detect if the capabilities supported by the server change, by sending the capabilities on each request we have a much cleaner way to handling different responses. - We are also introducing an unauthenticated endpoint to get the server features, this gives us flexibility if we need to implement different authentication mechanisms, and was one of the pitfalls of the first implementation. Related to https://github.com/fleetdm/fleet/issues/7929
2022-09-26 10:53:53 +00:00
bc.setServerCapabilities(response)
return nil
}
func (bc *baseClient) url(path, rawQuery string) *url.URL {
u := *bc.baseURL
u.Path = bc.urlPrefix + path
u.RawQuery = rawQuery
return &u
}
add headers denoting capabilities between fleet server / desktop / orbit (#7833) This adds a new mechanism to allow us to handle compatibility issues between Orbit, Fleet Server and Fleet Desktop. The general idea is to _always_ send a custom header of the form: ``` fleet-capabilities-header = "X-Fleet-Capabilities:" capabilities capabilities = capability * (,) capability = string ``` Both from the server to the clients (Orbit, Fleet Desktop) and vice-versa. For an example, see: https://github.com/fleetdm/fleet/commit/8c0bbdd291f54e03e19766bcdfead0fb8067f60c Also, the following applies: - Backwards compat: if the header is not present, assume that orbit/fleet doesn't have the capability - The current capabilities endpoint will be removed ### Motivation This solution is trying to solve the following problems: - We have three independent processes communicating with each other (Fleet Desktop, Orbit and Fleet Server). Each process can be updated independently, and therefore we need a way for each process to know what features are supported by its peers. - We originally implemented a dedicated API endpoint in the server that returned a list of the capabilities (or "features") enabled, we found this, and any other server-only solution (like API versioning) to be insufficient because: - There are cases in which the server also needs to know which features are supported by its clients - Clients needed to poll for changes to detect if the capabilities supported by the server change, by sending the capabilities on each request we have a much cleaner way to handling different responses. - We are also introducing an unauthenticated endpoint to get the server features, this gives us flexibility if we need to implement different authentication mechanisms, and was one of the pitfalls of the first implementation. Related to https://github.com/fleetdm/fleet/issues/7929
2022-09-26 10:53:53 +00:00
// setServerCapabilities updates the server capabilities based on the response
// from the server.
func (bc *baseClient) setServerCapabilities(response *http.Response) {
capabilities := response.Header.Get(fleet.CapabilitiesHeader)
bc.serverCapabilities.PopulateFromString(capabilities)
}
func (bc *baseClient) GetServerCapabilities() fleet.CapabilityMap {
return bc.serverCapabilities
add headers denoting capabilities between fleet server / desktop / orbit (#7833) This adds a new mechanism to allow us to handle compatibility issues between Orbit, Fleet Server and Fleet Desktop. The general idea is to _always_ send a custom header of the form: ``` fleet-capabilities-header = "X-Fleet-Capabilities:" capabilities capabilities = capability * (,) capability = string ``` Both from the server to the clients (Orbit, Fleet Desktop) and vice-versa. For an example, see: https://github.com/fleetdm/fleet/commit/8c0bbdd291f54e03e19766bcdfead0fb8067f60c Also, the following applies: - Backwards compat: if the header is not present, assume that orbit/fleet doesn't have the capability - The current capabilities endpoint will be removed ### Motivation This solution is trying to solve the following problems: - We have three independent processes communicating with each other (Fleet Desktop, Orbit and Fleet Server). Each process can be updated independently, and therefore we need a way for each process to know what features are supported by its peers. - We originally implemented a dedicated API endpoint in the server that returned a list of the capabilities (or "features") enabled, we found this, and any other server-only solution (like API versioning) to be insufficient because: - There are cases in which the server also needs to know which features are supported by its clients - Clients needed to poll for changes to detect if the capabilities supported by the server change, by sending the capabilities on each request we have a much cleaner way to handling different responses. - We are also introducing an unauthenticated endpoint to get the server features, this gives us flexibility if we need to implement different authentication mechanisms, and was one of the pitfalls of the first implementation. Related to https://github.com/fleetdm/fleet/issues/7929
2022-09-26 10:53:53 +00:00
}
// setClientCapabilities header is used to set a header with the client
// capabilities in the given request.
//
// This method is defined in baseClient because other clients generally have
// custom implementations of a method to perform the requests to the server.
func (bc *baseClient) setClientCapabilitiesHeader(req *http.Request) {
if len(bc.clientCapabilities) == 0 {
return
}
if req.Header == nil {
req.Header = http.Header{}
}
req.Header.Set(fleet.CapabilitiesHeader, bc.clientCapabilities.String())
}
func newBaseClient(
addr string,
insecureSkipVerify bool,
rootCA, urlPrefix string,
fleetClientCert *tls.Certificate,
capabilities fleet.CapabilityMap,
signerWrapper func(*http.Client) *http.Client,
) (*baseClient, error) {
baseURL, err := url.Parse(addr)
if err != nil {
return nil, fmt.Errorf("parsing URL: %w", err)
}
allowHTTP := insecureSkipVerify || strings.Contains(baseURL.Host, "localhost") || strings.Contains(baseURL.Host, "127.0.0.1")
if baseURL.Scheme != "https" && !allowHTTP {
return nil, errInvalidScheme
}
rootCAPool := x509.NewCertPool()
tlsConfig := &tls.Config{
// Osquery itself requires >= TLS 1.2.
// https://github.com/osquery/osquery/blob/9713ad9e28f1cfe6c16a823fb88bd531e39e192d/osquery/remote/transports/tls.cpp#L97-L98
MinVersion: tls.VersionTLS12,
}
if fleetClientCert != nil {
tlsConfig.Certificates = []tls.Certificate{*fleetClientCert}
}
switch {
case rootCA != "":
// read in the root cert file specified in the context
add headers denoting capabilities between fleet server / desktop / orbit (#7833) This adds a new mechanism to allow us to handle compatibility issues between Orbit, Fleet Server and Fleet Desktop. The general idea is to _always_ send a custom header of the form: ``` fleet-capabilities-header = "X-Fleet-Capabilities:" capabilities capabilities = capability * (,) capability = string ``` Both from the server to the clients (Orbit, Fleet Desktop) and vice-versa. For an example, see: https://github.com/fleetdm/fleet/commit/8c0bbdd291f54e03e19766bcdfead0fb8067f60c Also, the following applies: - Backwards compat: if the header is not present, assume that orbit/fleet doesn't have the capability - The current capabilities endpoint will be removed ### Motivation This solution is trying to solve the following problems: - We have three independent processes communicating with each other (Fleet Desktop, Orbit and Fleet Server). Each process can be updated independently, and therefore we need a way for each process to know what features are supported by its peers. - We originally implemented a dedicated API endpoint in the server that returned a list of the capabilities (or "features") enabled, we found this, and any other server-only solution (like API versioning) to be insufficient because: - There are cases in which the server also needs to know which features are supported by its clients - Clients needed to poll for changes to detect if the capabilities supported by the server change, by sending the capabilities on each request we have a much cleaner way to handling different responses. - We are also introducing an unauthenticated endpoint to get the server features, this gives us flexibility if we need to implement different authentication mechanisms, and was one of the pitfalls of the first implementation. Related to https://github.com/fleetdm/fleet/issues/7929
2022-09-26 10:53:53 +00:00
certs, err := os.ReadFile(rootCA)
if err != nil {
return nil, fmt.Errorf("reading root CA: %w", err)
}
// add certs to pool
if ok := rootCAPool.AppendCertsFromPEM(certs); !ok {
return nil, errors.New("failed to add certificates to root CA pool")
}
tlsConfig.RootCAs = rootCAPool
case insecureSkipVerify:
// Ignoring "G402: TLS InsecureSkipVerify set true", needed for development/testing.
tlsConfig.InsecureSkipVerify = true //nolint:gosec
default:
rootCAPool, err = x509.SystemCertPool()
if err != nil {
return nil, fmt.Errorf("loading system cert pool: %w", err)
}
tlsConfig.RootCAs = rootCAPool
}
httpClient := fleethttp.NewClient(fleethttp.WithTLSClientConfig(tlsConfig))
if signerWrapper != nil {
httpClient = signerWrapper(httpClient)
}
client := &baseClient{
baseURL: baseURL,
http: httpClient,
insecureSkipVerify: insecureSkipVerify,
urlPrefix: urlPrefix,
add headers denoting capabilities between fleet server / desktop / orbit (#7833) This adds a new mechanism to allow us to handle compatibility issues between Orbit, Fleet Server and Fleet Desktop. The general idea is to _always_ send a custom header of the form: ``` fleet-capabilities-header = "X-Fleet-Capabilities:" capabilities capabilities = capability * (,) capability = string ``` Both from the server to the clients (Orbit, Fleet Desktop) and vice-versa. For an example, see: https://github.com/fleetdm/fleet/commit/8c0bbdd291f54e03e19766bcdfead0fb8067f60c Also, the following applies: - Backwards compat: if the header is not present, assume that orbit/fleet doesn't have the capability - The current capabilities endpoint will be removed ### Motivation This solution is trying to solve the following problems: - We have three independent processes communicating with each other (Fleet Desktop, Orbit and Fleet Server). Each process can be updated independently, and therefore we need a way for each process to know what features are supported by its peers. - We originally implemented a dedicated API endpoint in the server that returned a list of the capabilities (or "features") enabled, we found this, and any other server-only solution (like API versioning) to be insufficient because: - There are cases in which the server also needs to know which features are supported by its clients - Clients needed to poll for changes to detect if the capabilities supported by the server change, by sending the capabilities on each request we have a much cleaner way to handling different responses. - We are also introducing an unauthenticated endpoint to get the server features, this gives us flexibility if we need to implement different authentication mechanisms, and was one of the pitfalls of the first implementation. Related to https://github.com/fleetdm/fleet/issues/7929
2022-09-26 10:53:53 +00:00
clientCapabilities: capabilities,
serverCapabilities: fleet.CapabilityMap{},
}
return client, nil
}
type bodyHandler interface {
Handle(*http.Response) error
}
type FileResponse struct {
DestPath string
DestFile string
destFilePath string
SkipMediaType bool
ProgressFunc func(n int)
}
func (f *FileResponse) Handle(resp *http.Response) error {
var filename string
if !f.SkipMediaType {
_, params, err := mime.ParseMediaType(resp.Header.Get("Content-Disposition"))
if err != nil {
return fmt.Errorf("parsing media type from response header: %w", err)
}
filename = params["filename"]
}
if filename == "" {
filename = f.DestFile
}
if filename == "" {
filename = uuid.NewString()
}
f.destFilePath = filepath.Join(f.DestPath, filename)
destFile, err := os.Create(f.destFilePath)
if err != nil {
return fmt.Errorf("creating file: %w", err)
}
defer destFile.Close()
var respBodyReader io.Reader = resp.Body
if f.ProgressFunc != nil {
respBodyReader = &progressReader{
Reader: respBodyReader,
progressFunc: f.ProgressFunc,
}
}
_, err = io.Copy(destFile, respBodyReader)
if err != nil {
return fmt.Errorf("copying from http stream to file: %w", err)
}
if err := destFile.Close(); err != nil {
return fmt.Errorf("closing file after copy: %w", err)
}
return nil
}
func (f *FileResponse) GetFilePath() string {
return f.destFilePath
}
type progressReader struct {
io.Reader
progressFunc func(n int)
}
func (pr *progressReader) Read(p []byte) (int, error) {
n, err := pr.Reader.Read(p)
pr.progressFunc(n)
return n, err
}