Skip to content

Treat NoneType provider shape errors as retryable#33136

Closed
Brixyy wants to merge 1 commit into
NousResearch:mainfrom
Brixyy:fix/nonetype-provider-error-retry
Closed

Treat NoneType provider shape errors as retryable#33136
Brixyy wants to merge 1 commit into
NousResearch:mainfrom
Brixyy:fix/nonetype-provider-error-retry

Conversation

@Brixyy

@Brixyy Brixyy commented May 27, 2026

Copy link
Copy Markdown

Summary

  • do not classify TypeError: 'NoneType' object is not iterable as a non-retryable local validation error
  • use the classifier-extracted status code, including status codes found in exception cause chains
  • add focused tests for local validation error classification

Why

Telegram/gateway turns can currently abort with Non-retryable error (HTTP None): 'NoneType' object is not iterable after a malformed provider/SDK response. That leaves the bot apparently online but unable to answer simple messages. These shape failures should go through the retry/fallback path instead of being treated as a local request validation bug.

Verification

  • python -m py_compile run_agent.py agent/error_classifier.py
  • focused helper assertion script for _is_nonretryable_local_validation_error

Full pytest could not run in this local Windows environment because the global Python lacks the openai package and the repo addopts expect pytest-xdist.

@alt-glitch alt-glitch added type/bug Something isn't working P3 Low — cosmetic, nice to have comp/agent Core agent loop, run_agent.py, prompt builder codex labels May 27, 2026
@markojak

Copy link
Copy Markdown

I'm experiencing this as well

⚠️ API call failed (attempt 1/3): TypeError
🔌 Provider: openai-codex Model: gpt-5.5
🌐 Endpoint: https://chatgpt.com/backend-api/codex
📝 Error: 'NoneType' object is not iterable
⚠️ Non-retryable error (HTTP None) — trying fallback...
❌ Non-retryable error (HTTP None): 'NoneType' object is not iterable
❌ Non-retryable client error (HTTP None). Aborting.
🔌 Provider: openai-codex Model: gpt-5.5
🌐 Endpoint: https://chatgpt.com/backend-api/codex
💡 This type of error won't be fixed by retrying.
I can start by checking #188’s current merge readiness now if you want.

@teknium1

Copy link
Copy Markdown
Contributor

Salvaged onto current main via #33399 (merged commit on main). Your authorship is preserved on the fix commit. Thanks @Brixyy — the classifier carve-out was exactly the right shape. The salvage applies the equivalent guard inline at the canonical extracted location (agent/conversation_loop.py:2869) since the target function moved out of the monolithic run_agent.py in a recent refactor.

@teknium1 teknium1 closed this May 27, 2026
jerome-benoit pushed a commit to jerome-benoit/hermes-agent that referenced this pull request May 27, 2026
…ble provider shape error

Salvages the intent of NousResearch#33136 (@Brixyy) onto current main. The original PR
was written against the pre-refactor monolithic run_agent.py and added a
top-level _is_nonretryable_local_validation_error() helper. Both target
functions have since been extracted to agent/conversation_loop.py:2869,
so the salvage applies the equivalent guard inline at that canonical
location rather than reintroducing the helper.

## Why

After NousResearch#33042 made our own Codex consumer structurally immune to NoneType
crashes, third-party shims, mocked clients, and any future code path that
hasn't migrated could still surface TypeError: 'NoneType' object is not
iterable as a wire-shape mismatch. The agent loop's classifier currently
treats ALL TypeError as a local programming bug and aborts non-retryable
— users on stale Telegram/gateway turns saw bare "Non-retryable error
(HTTP None)" with no recovery.

This is a provider/SDK shape mismatch, not a local programming bug. The
retry/fallback path should run, not be short-circuited.

## What

agent/conversation_loop.py: extend is_local_validation_error to exclude
TypeErrors whose message matches the NoneType-not-iterable shape (case-
insensitive, both "NoneType" and "not iterable" must appear).

tests/run_agent/test_jsondecodeerror_retryable.py:
- update the mirror predicate to match the production check
- add TestNoneTypeNotIterableIsRetryable class with 3 tests (the basic
  shape, message variants, unrelated TypeErrors still abort)
- add TestAgentLoopSourceHasNoneTypeCarveOut to enforce the source-level
  invariant matches the test mirror

## Validation

tests/run_agent/test_jsondecodeerror_retryable.py +
tests/run_agent/test_31273_402_not_retried.py → 14/14 passing

Co-authored-by: Brixyy <subrtt@gmail.com>
jerome-benoit pushed a commit to jerome-benoit/hermes-agent that referenced this pull request May 27, 2026
mathias3 pushed a commit to mathias3/hermes-agent that referenced this pull request May 28, 2026
…ble provider shape error

Salvages the intent of NousResearch#33136 (@Brixyy) onto current main. The original PR
was written against the pre-refactor monolithic run_agent.py and added a
top-level _is_nonretryable_local_validation_error() helper. Both target
functions have since been extracted to agent/conversation_loop.py:2869,
so the salvage applies the equivalent guard inline at that canonical
location rather than reintroducing the helper.

## Why

After NousResearch#33042 made our own Codex consumer structurally immune to NoneType
crashes, third-party shims, mocked clients, and any future code path that
hasn't migrated could still surface TypeError: 'NoneType' object is not
iterable as a wire-shape mismatch. The agent loop's classifier currently
treats ALL TypeError as a local programming bug and aborts non-retryable
— users on stale Telegram/gateway turns saw bare "Non-retryable error
(HTTP None)" with no recovery.

This is a provider/SDK shape mismatch, not a local programming bug. The
retry/fallback path should run, not be short-circuited.

## What

agent/conversation_loop.py: extend is_local_validation_error to exclude
TypeErrors whose message matches the NoneType-not-iterable shape (case-
insensitive, both "NoneType" and "not iterable" must appear).

tests/run_agent/test_jsondecodeerror_retryable.py:
- update the mirror predicate to match the production check
- add TestNoneTypeNotIterableIsRetryable class with 3 tests (the basic
  shape, message variants, unrelated TypeErrors still abort)
- add TestAgentLoopSourceHasNoneTypeCarveOut to enforce the source-level
  invariant matches the test mirror

## Validation

tests/run_agent/test_jsondecodeerror_retryable.py +
tests/run_agent/test_31273_402_not_retried.py → 14/14 passing

Co-authored-by: Brixyy <subrtt@gmail.com>
mathias3 pushed a commit to mathias3/hermes-agent that referenced this pull request May 28, 2026
Bryce-huang pushed a commit to wbkunlun/hermes-agent that referenced this pull request May 29, 2026
…ble provider shape error

Salvages the intent of NousResearch#33136 (@Brixyy) onto current main. The original PR
was written against the pre-refactor monolithic run_agent.py and added a
top-level _is_nonretryable_local_validation_error() helper. Both target
functions have since been extracted to agent/conversation_loop.py:2869,
so the salvage applies the equivalent guard inline at that canonical
location rather than reintroducing the helper.

## Why

After NousResearch#33042 made our own Codex consumer structurally immune to NoneType
crashes, third-party shims, mocked clients, and any future code path that
hasn't migrated could still surface TypeError: 'NoneType' object is not
iterable as a wire-shape mismatch. The agent loop's classifier currently
treats ALL TypeError as a local programming bug and aborts non-retryable
— users on stale Telegram/gateway turns saw bare "Non-retryable error
(HTTP None)" with no recovery.

This is a provider/SDK shape mismatch, not a local programming bug. The
retry/fallback path should run, not be short-circuited.

## What

agent/conversation_loop.py: extend is_local_validation_error to exclude
TypeErrors whose message matches the NoneType-not-iterable shape (case-
insensitive, both "NoneType" and "not iterable" must appear).

tests/run_agent/test_jsondecodeerror_retryable.py:
- update the mirror predicate to match the production check
- add TestNoneTypeNotIterableIsRetryable class with 3 tests (the basic
  shape, message variants, unrelated TypeErrors still abort)
- add TestAgentLoopSourceHasNoneTypeCarveOut to enforce the source-level
  invariant matches the test mirror

## Validation

tests/run_agent/test_jsondecodeerror_retryable.py +
tests/run_agent/test_31273_402_not_retried.py → 14/14 passing

Co-authored-by: Brixyy <subrtt@gmail.com>

#AI commit#
Bryce-huang pushed a commit to wbkunlun/hermes-agent that referenced this pull request May 29, 2026
mosaiq-systems pushed a commit to mosaiq-systems/hermes-agent that referenced this pull request May 29, 2026
…ble provider shape error

Salvages the intent of NousResearch#33136 (@Brixyy) onto current main. The original PR
was written against the pre-refactor monolithic run_agent.py and added a
top-level _is_nonretryable_local_validation_error() helper. Both target
functions have since been extracted to agent/conversation_loop.py:2869,
so the salvage applies the equivalent guard inline at that canonical
location rather than reintroducing the helper.

## Why

After NousResearch#33042 made our own Codex consumer structurally immune to NoneType
crashes, third-party shims, mocked clients, and any future code path that
hasn't migrated could still surface TypeError: 'NoneType' object is not
iterable as a wire-shape mismatch. The agent loop's classifier currently
treats ALL TypeError as a local programming bug and aborts non-retryable
— users on stale Telegram/gateway turns saw bare "Non-retryable error
(HTTP None)" with no recovery.

This is a provider/SDK shape mismatch, not a local programming bug. The
retry/fallback path should run, not be short-circuited.

## What

agent/conversation_loop.py: extend is_local_validation_error to exclude
TypeErrors whose message matches the NoneType-not-iterable shape (case-
insensitive, both "NoneType" and "not iterable" must appear).

tests/run_agent/test_jsondecodeerror_retryable.py:
- update the mirror predicate to match the production check
- add TestNoneTypeNotIterableIsRetryable class with 3 tests (the basic
  shape, message variants, unrelated TypeErrors still abort)
- add TestAgentLoopSourceHasNoneTypeCarveOut to enforce the source-level
  invariant matches the test mirror

## Validation

tests/run_agent/test_jsondecodeerror_retryable.py +
tests/run_agent/test_31273_402_not_retried.py → 14/14 passing

Co-authored-by: Brixyy <subrtt@gmail.com>
mosaiq-systems pushed a commit to mosaiq-systems/hermes-agent that referenced this pull request May 29, 2026
KKT-OPT pushed a commit to KKT-OPT/hermes-agent that referenced this pull request May 31, 2026
…ble provider shape error

Salvages the intent of NousResearch#33136 (@Brixyy) onto current main. The original PR
was written against the pre-refactor monolithic run_agent.py and added a
top-level _is_nonretryable_local_validation_error() helper. Both target
functions have since been extracted to agent/conversation_loop.py:2869,
so the salvage applies the equivalent guard inline at that canonical
location rather than reintroducing the helper.

## Why

After NousResearch#33042 made our own Codex consumer structurally immune to NoneType
crashes, third-party shims, mocked clients, and any future code path that
hasn't migrated could still surface TypeError: 'NoneType' object is not
iterable as a wire-shape mismatch. The agent loop's classifier currently
treats ALL TypeError as a local programming bug and aborts non-retryable
— users on stale Telegram/gateway turns saw bare "Non-retryable error
(HTTP None)" with no recovery.

This is a provider/SDK shape mismatch, not a local programming bug. The
retry/fallback path should run, not be short-circuited.

## What

agent/conversation_loop.py: extend is_local_validation_error to exclude
TypeErrors whose message matches the NoneType-not-iterable shape (case-
insensitive, both "NoneType" and "not iterable" must appear).

tests/run_agent/test_jsondecodeerror_retryable.py:
- update the mirror predicate to match the production check
- add TestNoneTypeNotIterableIsRetryable class with 3 tests (the basic
  shape, message variants, unrelated TypeErrors still abort)
- add TestAgentLoopSourceHasNoneTypeCarveOut to enforce the source-level
  invariant matches the test mirror

## Validation

tests/run_agent/test_jsondecodeerror_retryable.py +
tests/run_agent/test_31273_402_not_retried.py → 14/14 passing

Co-authored-by: Brixyy <subrtt@gmail.com>
KKT-OPT pushed a commit to KKT-OPT/hermes-agent that referenced this pull request May 31, 2026
gweeteve pushed a commit to gweeteve/hermes-agent that referenced this pull request Jun 2, 2026
…ble provider shape error

Salvages the intent of NousResearch#33136 (@Brixyy) onto current main. The original PR
was written against the pre-refactor monolithic run_agent.py and added a
top-level _is_nonretryable_local_validation_error() helper. Both target
functions have since been extracted to agent/conversation_loop.py:2869,
so the salvage applies the equivalent guard inline at that canonical
location rather than reintroducing the helper.

## Why

After NousResearch#33042 made our own Codex consumer structurally immune to NoneType
crashes, third-party shims, mocked clients, and any future code path that
hasn't migrated could still surface TypeError: 'NoneType' object is not
iterable as a wire-shape mismatch. The agent loop's classifier currently
treats ALL TypeError as a local programming bug and aborts non-retryable
— users on stale Telegram/gateway turns saw bare "Non-retryable error
(HTTP None)" with no recovery.

This is a provider/SDK shape mismatch, not a local programming bug. The
retry/fallback path should run, not be short-circuited.

## What

agent/conversation_loop.py: extend is_local_validation_error to exclude
TypeErrors whose message matches the NoneType-not-iterable shape (case-
insensitive, both "NoneType" and "not iterable" must appear).

tests/run_agent/test_jsondecodeerror_retryable.py:
- update the mirror predicate to match the production check
- add TestNoneTypeNotIterableIsRetryable class with 3 tests (the basic
  shape, message variants, unrelated TypeErrors still abort)
- add TestAgentLoopSourceHasNoneTypeCarveOut to enforce the source-level
  invariant matches the test mirror

## Validation

tests/run_agent/test_jsondecodeerror_retryable.py +
tests/run_agent/test_31273_402_not_retried.py → 14/14 passing

Co-authored-by: Brixyy <subrtt@gmail.com>
gweeteve pushed a commit to gweeteve/hermes-agent that referenced this pull request Jun 2, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

codex comp/agent Core agent loop, run_agent.py, prompt builder P3 Low — cosmetic, nice to have type/bug Something isn't working

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants