feat(gitea): add retry logic for 5xx errors #69

Merged
aweiker merged 7 commits from issue-68 into main 2026-05-11 12:59:50 +00:00

7 Commits

Author SHA1 Message Date
Rodin 1b38e6ad00 fix(gitea): ensure consistent lastErr return for network errors
PR Ready Gate / clear-labels (pull_request) Successful in 2s
CI / test (pull_request) Successful in 17s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 35s
CI / review (gpt-5, security, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 58s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 1m37s
Move lastErr assignment outside the retry condition so that both
network errors and HTTP 5xx paths return lastErr consistently.

Previously, on the final retry attempt, a network error would return
the raw err variable instead of lastErr. While they held the same
value in practice, the inconsistency was confusing when reading the
code.

Now both paths:
- Network errors: assign lastErr before checking retry, return lastErr
- HTTP 5xx: assign lastErr before checking retry, return lastErr

Addresses review finding #3 (MINOR) from sonnet review on PR #69.
2026-05-11 05:31:39 -07:00
Rodin 5498dccd60 fix(gitea): address MINOR review findings on retry logic
PR Ready Gate / clear-labels (pull_request) Successful in 2s
CI / test (pull_request) Successful in 17s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 39s
CI / review (gpt-5, security, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 50s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 1m28s
1. Fix non-deterministic test TestDoGet_RetriesOnTemporaryNetError:
   - Replace timing-dependent listener approach with mockTransport
   - mockTransport allows controlled injection of net.OpError failures
   - Test now makes deterministic assertions: exactly 3 attempts (2 fail + 1 success)
   - Added SetHTTPClient() method for test transport injection

2. Sanitize error content in retry warning logs:
   - Added sanitizeErrorForLog() helper that omits response body content
   - For APIError: logs only 'HTTP <status>' instead of full body
   - For other errors: preserves error type information
   - Addresses security concern about logging server error content at WARN level
   - Full error with body still returned to caller for proper error handling

Both changes have corresponding test coverage.
2026-05-11 05:15:07 -07:00
Aaron Weiker ecfbfddc7c fix(gitea): redact query params from retry warning logs
PR Ready Gate / clear-labels (pull_request) Successful in 1s
CI / test (pull_request) Successful in 17s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 34s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 1m14s
CI / review (gpt-5, security, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 1m17s
Addresses security review finding: retry warnings were logging the full
request URL which could inadvertently leak sensitive query parameters
if future callers pass them.

Added redactURL() helper that:
- Strips query parameters from URLs before logging (replaces with [redacted])
- Returns [invalid URL] for unparseable URLs to avoid leaking any data
- Preserves the base path for debugging context

The error itself (lastErr) is kept as-is since APIError.Error() already
truncates response bodies to 200 chars, and network errors don't contain
user-controlled data.
2026-05-11 04:52:41 -07:00
Rodin ac53ecfa5d fix(gitea): improve retry logic precision for net.OpError
PR Ready Gate / clear-labels (pull_request) Successful in 2s
CI / test (pull_request) Successful in 17s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 46s
CI / review (gpt-5, security, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 1m11s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 1m23s
Address review feedback on isTemporaryNetError being too broad:

1. RetryBackoff field: Added doc comment clarifying it must be
   configured before the first request (addresses concurrency concern).

2. isTemporaryNetError: Now inspects the underlying syscall error
   instead of treating all net.OpError as retriable. Only retries on:
   - ECONNREFUSED (connection refused)
   - ECONNRESET (connection reset)
   - ENETUNREACH (network unreachable)
   - EHOSTUNREACH (host unreachable)
   - ETIMEDOUT (connection timed out)

   Permanent errors like EACCES, EPERM are no longer retried.

3. DNS errors: Changed from Temporary() to IsTimeout, since
   "no such host" is permanent and shouldn't be retried.

4. Empty backoff slice: Added comment explaining that retry without
   delay is intentional when caller explicitly configures it.

Addresses MINOR findings from sonnet-review-bot and gpt-review-bot.
2026-05-11 04:32:15 -07:00
Rodin 090ae3848c fix(gitea): make retry backoff configurable and retry temp net errors
PR Ready Gate / clear-labels (pull_request) Successful in 1s
CI / test (pull_request) Successful in 17s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 34s
CI / review (gpt-5, security, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 45s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 1m34s
Address review feedback:

1. Make backoff delays injectable via Client.RetryBackoff field
   - Defaults to {1s, 2s} when nil for production
   - Tests can set shorter values for fast execution
   - Fixes slow unit tests that previously waited 3+ seconds

2. Add retry on temporary network errors (net.OpError, net.DNSError)
   - Connection refused, network unreachable, DNS failures now retry
   - Non-temporary network errors still fail immediately
   - Context cancellation still respected during backoff

Added isTemporaryNetError helper and TestIsTemporaryNetError test.
Updated existing retry tests to use configurable short backoffs.
2026-05-11 04:23:27 -07:00
Rodin 23da7eedf5 fix(gitea): address review feedback on retry logic
PR Ready Gate / clear-labels (pull_request) Successful in 2s
CI / test (pull_request) Successful in 23s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 30s
CI / review (gpt-5, security, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 45s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 1m30s
- Remove dead backoff[0] element; array now only contains retry delays
- Fix time.After timer leak by using time.NewTimer with timer.Stop()
- Add io.LimitReader (64KB) for error body reads to bound memory allocation

Addresses feedback from sonnet-review-bot, security-review-bot, and gpt-review-bot.
2026-05-11 01:08:01 -07:00
Rodin 7279cdd216 feat(gitea): add retry logic for 5xx errors in doGet
CI / test (pull_request) Successful in 28s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 30s
CI / review (gpt-5, security, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 1m27s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 1m33s
Addresses transient HTTP 500 errors from Gitea API during pattern fetches.

Changes:
- Add retry with exponential backoff (1s, 2s) to doGet(), max 3 attempts
- Add IsServerError() helper to detect 5xx responses
- No retry on 4xx errors (client errors should propagate immediately)
- Respects context cancellation during backoff waits
- Logs retries at WARN level for observability

All existing tests pass. New tests:
- TestIsServerError: validates 5xx detection across edge cases
- TestDoGet_RetriesOn500: verifies recovery after transient errors
- TestDoGet_FailsAfterMaxRetries: verifies proper failure after exhaustion
- TestDoGet_NoRetryOn4xx: ensures client errors don't retry
- TestDoGet_RespectsContextCancellation: validates cancellation during backoff

Closes #68
2026-05-11 00:59:47 -07:00