fix(failover): patch P1-P3 reliability gaps surfaced by team review

- Forwarder buffers non-streaming bodies and primes streaming first
  chunk before signaling success, so body timeouts and SSE first-chunk
  failures route through the circuit breaker instead of being recorded
  as success on response-header arrival
- Atomic enable-failover: switch to P1 before persisting the flag, and
  roll back auto-added queue entries when the switch is rejected
  (e.g. official providers)
- Hot-reload circuit breaker config on per-app proxy config change
  instead of waiting for a proxy restart
- FailoverToggle / FailoverQueueManager / AutoFailoverConfigPanel
  require proxy takeover for the active app; the backend command also
  rejects enabling when takeover is off
- ProviderHealthBadge consumes the backend is_healthy flag instead of
  hardcoding the 5-failure threshold

Cleanup:
- impl From<&AppProxyConfig> for CircuitBreakerConfig and use it from
  the command layer
- Collapse three identical TabsContent blocks into a single map
This commit is contained in:
Jason
2026-05-14 21:35:02 +08:00
parent 940161fb0e
commit b642ef0633
14 changed files with 389 additions and 98 deletions
+17 -8
View File
@@ -10,6 +10,7 @@ import {
useAutoFailoverEnabled,
useSetAutoFailoverEnabled,
} from "@/lib/query/failover";
import { useProxyStatus } from "@/hooks/useProxyStatus";
import { cn } from "@/lib/utils";
import { useTranslation } from "react-i18next";
import type { AppId } from "@/lib/api";
@@ -24,8 +25,11 @@ export function FailoverToggle({ className, activeApp }: FailoverToggleProps) {
const { data: isEnabled = false, isLoading } =
useAutoFailoverEnabled(activeApp);
const setEnabled = useSetAutoFailoverEnabled();
const { takeoverStatus } = useProxyStatus();
const takeoverEnabled = takeoverStatus?.[activeApp] ?? false;
const handleToggle = (checked: boolean) => {
if (checked && !takeoverEnabled) return;
setEnabled.mutate({ appType: activeApp, enabled: checked });
};
@@ -36,15 +40,20 @@ export function FailoverToggle({ className, activeApp }: FailoverToggleProps) {
? "Codex"
: "Gemini";
const tooltipText = isEnabled
? t("failover.tooltip.enabled", {
const tooltipText = !takeoverEnabled
? t("failover.tooltip.takeoverRequired", {
app: appLabel,
defaultValue: `${appLabel} 故障转移已启用\n按队列优先级(P1→P2→...)选择供应商`,
defaultValue: `请先接管 ${appLabel},再启用故障转移`,
})
: t("failover.tooltip.disabled", {
app: appLabel,
defaultValue: `启用 ${appLabel} 故障转移\n将立即切换到队列 P1,并在失败时自动切换到下一个`,
});
: isEnabled
? t("failover.tooltip.enabled", {
app: appLabel,
defaultValue: `${appLabel} 故障转移已启用\n按队列优先级(P1→P2→...)选择供应商`,
})
: t("failover.tooltip.disabled", {
app: appLabel,
defaultValue: `启用 ${appLabel} 故障转移\n将立即切换到队列 P1,并在失败时自动切换到下一个`,
});
return (
<div
@@ -69,7 +78,7 @@ export function FailoverToggle({ className, activeApp }: FailoverToggleProps) {
<Switch
checked={isEnabled}
onCheckedChange={handleToggle}
disabled={setEnabled.isPending || isLoading}
disabled={setEnabled.isPending || isLoading || !takeoverEnabled}
/>
</div>
);
+1
View File
@@ -725,6 +725,7 @@ function ProviderQueueItem({
{/* 健康徽章 */}
<ProviderHealthBadge
consecutiveFailures={health?.consecutive_failures ?? 0}
isHealthy={health?.is_healthy}
/>
</div>
);