mirror of
https://github.com/coder/coder.git
synced 2026-10-31 19:11:55 +08:00
Susana Ferreira
dba45cede7
## Problem When a key returned 401 or 403, the pool marked it permanently unavailable for the lifetime of that in-memory pool. This is bad UX: a transient auth failure or a briefly-misconfigured key could take a key out of rotation until the operator either restarted Coder or reconfigured the key (even re-saving the same working value). ## Changes - **403 removed from key failover**: it's a per-request authorization failure, not a key-level problem, so it's surfaced to the caller as-is without marking the key or failing over. - **401 now applies a temporary cooldown** (like 429) so the key recovers on its own instead of staying blocked. - When every key is in an auth-failure cooldown, the pool reports a `502` with no `Retry-After`, but the keys still recover automatically once the cooldown elapses. Closes https://linear.app/codercom/issue/AIGOV-421/ai-gateway-a-quarantined-centralized-key-never-recovers-without-a Closes https://linear.app/codercom/issue/AIGOV-533/403s-misclassifying-keys-as-permanently-down-in-ai-gateway > [!NOTE] > Initially generated by Claude Opus 4.7, modified and reviewed by @ssncferreira
70 lines
2.5 KiB
Go
70 lines
2.5 KiB
Go
package integrationtest
|
|
|
|
import (
|
|
"io"
|
|
"net/http"
|
|
"testing"
|
|
|
|
"github.com/stretchr/testify/require"
|
|
"github.com/tidwall/sjson"
|
|
|
|
"github.com/coder/coder/v2/aibridge/config"
|
|
"github.com/coder/coder/v2/aibridge/fixtures"
|
|
"github.com/coder/coder/v2/aibridge/internal/testutil"
|
|
"github.com/coder/coder/v2/aibridge/recorder"
|
|
)
|
|
|
|
// TestInterceptionUpstreamErrorRecorded verifies that a failed interception
|
|
// records a categorized upstream error on the ended record.
|
|
//
|
|
// The default test provider is centralized (backed by a single-key pool), so a
|
|
// 401 marks the key temporary and exhausts the pool. Both blocking and
|
|
// streaming interceptors preserve the *keypool.Error so the cause is
|
|
// categorized as "unauthorized", the auth-failure exhaustion outcome.
|
|
func TestInterceptionUpstreamErrorRecorded(t *testing.T) {
|
|
t.Parallel()
|
|
|
|
cases := []struct {
|
|
name string
|
|
provider string
|
|
fixture []byte
|
|
path string
|
|
streaming bool
|
|
wantType recorder.ErrorType
|
|
}{
|
|
{"anthropic_blocking", config.ProviderAnthropic, fixtures.AntSimple, pathAnthropicMessages, false, recorder.ErrorTypeUnauthorized},
|
|
{"anthropic_streaming", config.ProviderAnthropic, fixtures.AntSimple, pathAnthropicMessages, true, recorder.ErrorTypeUnauthorized},
|
|
{"openai_blocking", config.ProviderOpenAI, fixtures.OaiChatSimple, pathOpenAIChatCompletions, false, recorder.ErrorTypeUnauthorized},
|
|
{"openai_streaming", config.ProviderOpenAI, fixtures.OaiChatSimple, pathOpenAIChatCompletions, true, recorder.ErrorTypeUnauthorized},
|
|
}
|
|
|
|
for _, tc := range cases {
|
|
t.Run(tc.name, func(t *testing.T) {
|
|
t.Parallel()
|
|
|
|
fix := fixtures.Parse(t, tc.fixture)
|
|
upstream := testutil.NewMockUpstream(t.Context(), t,
|
|
testutil.NewErrorResponse(http.StatusUnauthorized, ""),
|
|
)
|
|
upstream.AllowOverflow = true
|
|
|
|
bridgeServer := newBridgeTestServer(t.Context(), t, upstream.URL, withProvider(tc.provider))
|
|
|
|
reqBody, err := sjson.SetBytes(fix.Request(), "stream", tc.streaming)
|
|
require.NoError(t, err)
|
|
|
|
resp, err := bridgeServer.makeRequest(t, http.MethodPost, tc.path, reqBody)
|
|
require.NoError(t, err)
|
|
_, _ = io.Copy(io.Discard, resp.Body)
|
|
require.NoError(t, resp.Body.Close())
|
|
|
|
intcs := bridgeServer.Recorder.RecordedInterceptions()
|
|
require.Len(t, intcs, 1)
|
|
ended := bridgeServer.Recorder.RecordedInterceptionEnd(intcs[0].ID)
|
|
require.NotNil(t, ended, "interception should be ended")
|
|
require.Equal(t, tc.wantType, ended.ErrorType)
|
|
require.NotEmpty(t, ended.ErrorMessage)
|
|
})
|
|
}
|
|
}
|