fix(control-plane): stop Keycloak event storms #230
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,169 @@ | ||
| package auth | ||
|
|
||
| import ( | ||
| "bytes" | ||
| "encoding/json" | ||
| "fmt" | ||
| "log" | ||
| "net/http" | ||
| "net/http/httptest" | ||
| "strings" | ||
| "sync/atomic" | ||
| "testing" | ||
| "time" | ||
| ) | ||
|
|
||
| func TestTokenGrantLogContainsSafeRefreshDetails(t *testing.T) { | ||
| var output bytes.Buffer | ||
| priorWriter := log.Writer() | ||
| priorFlags := log.Flags() | ||
| priorPrefix := log.Prefix() | ||
| log.SetOutput(&output) | ||
| log.SetFlags(0) | ||
| log.SetPrefix("") | ||
| t.Cleanup(func() { | ||
| log.SetOutput(priorWriter) | ||
| log.SetFlags(priorFlags) | ||
| log.SetPrefix(priorPrefix) | ||
| }) | ||
|
|
||
| var grants atomic.Int32 | ||
| server := newTokenServer(t, &grants) | ||
| provider := NewTokenProvider("https://issuer.invalid", "client-id\nforged-entry", "client-secret") | ||
| provider.SetTokenEndpoint(server.URL) | ||
|
|
||
| if _, err := provider.Token(); err != nil { | ||
| t.Fatalf("Token() failed: %v", err) | ||
| } | ||
|
|
||
| message := output.String() | ||
| if !strings.Contains(message, `client "client-id\nforged-entry"; refresh in 4m0s`) { | ||
| t.Fatalf("token grant log = %q, want quoted client ID and refresh interval", message) | ||
| } | ||
| for _, forbidden := range []string{"token-1", "client-secret", "\nforged-entry"} { | ||
| if strings.Contains(message, forbidden) { | ||
| t.Fatalf("token grant log contains unsafe value %q: %q", forbidden, message) | ||
| } | ||
| } | ||
| } | ||
|
|
||
| func TestTokenReusesCachedToken(t *testing.T) { | ||
| t.Parallel() | ||
|
|
||
| var grants atomic.Int32 | ||
| server := newTokenServer(t, &grants) | ||
|
|
||
| provider := NewTokenProvider("https://issuer.invalid", "client-id", "client-secret") | ||
| provider.SetTokenEndpoint(server.URL) | ||
|
|
||
| first, err := provider.Token() | ||
| if err != nil { | ||
| t.Fatalf("first Token() call failed: %v", err) | ||
| } | ||
| second, err := provider.Token() | ||
| if err != nil { | ||
| t.Fatalf("second Token() call failed: %v", err) | ||
| } | ||
|
|
||
| if first != second { | ||
| t.Fatalf("Token() returned %q and %q; both calls must return the same token", first, second) | ||
| } | ||
| if got := grants.Load(); got != 1 { | ||
| t.Fatalf("token endpoint received %d grants; it must receive 1", got) | ||
| } | ||
| } | ||
|
|
||
| func TestConcurrentTokenCallsShareCachedToken(t *testing.T) { | ||
| t.Parallel() | ||
|
|
||
| var grants atomic.Int32 | ||
| server := newTokenServer(t, &grants) | ||
|
|
||
| provider := NewTokenProvider("https://issuer.invalid", "client-id", "client-secret") | ||
| provider.SetTokenEndpoint(server.URL) | ||
|
|
||
| const callCount = 16 | ||
| type result struct { | ||
| token string | ||
| err error | ||
| } | ||
|
|
||
| start := make(chan struct{}) | ||
| results := make(chan result, callCount) | ||
| for range callCount { | ||
| go func() { | ||
| <-start | ||
| token, err := provider.Token() | ||
| results <- result{token: token, err: err} | ||
| }() | ||
| } | ||
| close(start) | ||
|
|
||
| for range callCount { | ||
| result := <-results | ||
| if result.err != nil { | ||
| t.Fatalf("Token() failed: %v", result.err) | ||
| } | ||
| if result.token != "token-1" { | ||
| t.Fatalf("Token() returned %q; it must return %q", result.token, "token-1") | ||
| } | ||
| } | ||
|
|
||
| if got := grants.Load(); got != 1 { | ||
| t.Fatalf("token endpoint received %d grants; it must receive 1", got) | ||
| } | ||
| } | ||
|
|
||
| func TestTokenRefreshesAfterThreshold(t *testing.T) { | ||
| t.Parallel() | ||
|
|
||
| var grants atomic.Int32 | ||
| server := newTokenServer(t, &grants) | ||
|
|
||
| provider := NewTokenProvider("https://issuer.invalid", "client-id", "client-secret") | ||
| provider.SetTokenEndpoint(server.URL) | ||
|
|
||
| requestStarted := time.Now() | ||
| first, err := provider.Token() | ||
| requestFinished := time.Now() | ||
| if err != nil { | ||
| t.Fatalf("first Token() call failed: %v", err) | ||
| } | ||
|
|
||
| provider.mu.Lock() | ||
| refreshAt := provider.expiry | ||
| provider.expiry = time.Now().Add(-time.Second) | ||
| provider.mu.Unlock() | ||
|
|
||
| const refreshDelay = 4 * time.Minute | ||
| if refreshAt.Before(requestStarted.Add(refreshDelay)) || refreshAt.After(requestFinished.Add(refreshDelay)) { | ||
| t.Fatalf("refresh time = %v; want 80 percent of a 300-second lifetime", refreshAt) | ||
| } | ||
|
|
||
| second, err := provider.Token() | ||
| if err != nil { | ||
| t.Fatalf("second Token() call failed: %v", err) | ||
| } | ||
| if first != "token-1" || second != "token-2" { | ||
| t.Fatalf("Token() returned %q and %q; want %q and %q", first, second, "token-1", "token-2") | ||
| } | ||
| if got := grants.Load(); got != 2 { | ||
| t.Fatalf("token endpoint received %d grants; it must receive 2", got) | ||
| } | ||
| } | ||
|
|
||
| func newTokenServer(t *testing.T, grants *atomic.Int32) *httptest.Server { | ||
| t.Helper() | ||
|
|
||
| server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) { | ||
| grant := grants.Add(1) | ||
| w.Header().Set("Content-Type", "application/json") | ||
| _ = json.NewEncoder(w).Encode(tokenResponse{ | ||
| AccessToken: fmt.Sprintf("token-%d", grant), | ||
| ExpiresIn: 300, | ||
| TokenType: "Bearer", | ||
| }) | ||
| })) | ||
| t.Cleanup(server.Close) | ||
| return server | ||
| } |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Root cause confirmed and correctly fixed. The prior
time.Duration(float64(expiresIn) * 0.8)treatedexpiresIn(a value in seconds) as nanoseconds, so a 300s token was cached for ~240ns and effectively everyToken()call re-ran the client_credentials grant — that is the KeycloakCLIENT_LOGINevent storm.time.Duration(expiresIn) * time.Secondrestores the intended unit and the 80% threshold. Confidence: High.The integer form
ttl * 8 / 10avoids float rounding and cannot overflow for any realisticexpires_in(max 900s), so this is a clean choice.