sscoble/pi-mono
1
0
Fork 0
mirror of https://github.com/badlogic/pi-mono.git synced 2026-11-05 23:37:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 4

fix(ai): retry Azure peak-load capacity errors

Browse source 
Closes #9669
This commit is contained in:
David Brailovsky 2026-09-17 12:01:39 +02:00
parent 7140838fdd
commit e98f287ee4
3 changed files with 11 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

1
packages/ai/CHANGELOG.md
View file

@ -25,6 +25,7 @@
- Fixed OpenAI-compatible Responses errors to identify the actual provider instead of always labeling them as OpenAI errors ([#9298](https://github.com/earendil-works/pi/issues/9298)).
- Fixed Baseten requests to send session-affinity headers from `sessionId` for automatic prompt-cache routing ([#9629](https://github.com/earendil-works/pi/issues/9629)).
- Fixed retry classification for Cloudflare 520 responses ([#9627](https://github.com/earendil-works/pi/issues/9627)).
- Fixed retry classification for transient Azure peak-load capacity errors ([#9669](https://github.com/earendil-works/pi/issues/9669)).
## [0.85.1] - 2026-09-05

1
packages/ai/src/utils/retry.ts
View file

@ -26,6 +26,7 @@ const NON_RETRYABLE_PROVIDER_LIMIT_ERROR_PATTERN = buildProviderErrorPattern([
const RETRYABLE_PROVIDER_ERROR_PATTERN = buildProviderErrorPattern([
// Generic provider load, HTTP status, and server-side transient failures.
"overloaded",
"currently experiencing high demand",
"rate.?limit",
"too many requests",
"429",

9
packages/ai/test/retry.test.ts
View file

@ -12,6 +12,8 @@ const bunFetchSocketClosedMessage =
const openAIResponsesEarlyEofMessage = "OpenAI Responses stream ended before a terminal response event";
const wrappedDnsLookupError =
"The pending stream has been canceled (caused by: getaddrinfo ENOTFOUND bedrock-runtime.us-east-1.amazonaws.com)";
const azurePeakLoadError =
"The system is currently experiencing high demand and cannot process your request. Your request exceeds the maximum usage size allowed during peak load. For improved capacity reliability, consider switching to Provisioned Throughput.";
describe("provider retry classification", () => {
it("matches explicit provider retry guidance", () => {
@ -68,6 +70,13 @@ describe("provider retry classification", () => {
).toBe(true);
});
it("matches Azure peak-load capacity errors", () => {
// Regression for #9669.
expect(
isRetryableAssistantError(fauxAssistantMessage("", { stopReason: "error", errorMessage: azurePeakLoadError })),
).toBe(true);
});
it("keeps provider limit errors non-retryable", () => {
expect(
isRetryableAssistantError(