fix: fixed issue #1587

[richardz403]

Description

Fixes issue #1587, where resetting a flow causes the LLM selection to revert to a stale or default model.

In src/api/settings/langflow_sync.py, the fallback branch of _update_langflow_model_values (called during reapply_all_settings()) was only iterating and refreshing configured embedding providers. It completely bypassed updating the LLM providers, meaning the chat flow never received a force_llm_update=True signal.

Changes

• Modified the fallback block in src/api/settings/langflow_sync.py inside reapply_all_settings().
• If everything is blank, it automatically falls back to your system's default agent settings (config.agent) and forces an update (force_llm_update=True) to the chat flow using that default LLM provider and model.

Summary by CodeRabbit

• Bug Fixes
  • Improved Langflow model synchronization by adding a reliable fallback to the currently configured LLM provider/model when explicit embedding/LLM settings aren’t provided, improving reliability.
• Infrastructure
  • Updated Docker Compose to expose the backend service on host port 8000 for easier local access.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 860523e4-c758-4160-83fd-b8e6a8ada2d4

📥 Commits

Reviewing files that changed from the base of the PR and between 9077fb0 and a79c8d1.

📒 Files selected for processing (1)

• src/api/settings/langflow_sync.py

💤 Files with no reviewable changes (1)

• src/api/settings/langflow_sync.py

---

Walkthrough

Adds a fallback branch in _update_langflow_model_values that forces an LLM sync with the current config when no model-provider arguments are supplied. Also exposes the backend container port 8000 to the host in Docker Compose for local access.

Changes

LLM Sync Fallback and Local Deployment

Layer / File(s)	Summary
Fallback force-update branch src/api/settings/langflow_sync.py	Adds a conditional block in _update_langflow_model_values: when no model or provider arguments are passed, forces a Langflow model-value sync using the currently configured LLM provider/model (force_llm_update=True) and logs success before proceeding to update embedding providers.
Backend service port exposure docker-compose.yml	Adds a ports mapping to the openrag-backend service to expose container port 8000 on the host ("8000:8000").

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Possibly related issues

• bug: reapply_all_settings() skips LLM sync in fallback branch #1587: The fallback force_llm_update=True branch added in this PR directly addresses the bug reported in this issue, where LLM sync was skipped when no model-provider arguments were supplied.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 inconclusive)

Check name	Status	Explanation	Resolution
Title check	❓ Inconclusive	The title 'fix: fixed issue #1587' is vague and generic, referencing only an issue number without describing the actual change or fix implemented.	Replace with a more descriptive title that explains the fix, e.g., 'fix: restore LLM selection on flow reset by forcing model update in fallback path' or similar.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[NishadA05]

I checked out this PR locally and reviewed the diff. The change is narrow and limited to src/api/settings/langflow_sync.py. It updates the fallback path in _update_langflow_model_values() so that when no explicit model/provider overrides are passed, the configured agent LLM provider/model is forced back into the Langflow chat flow with force_llm_update=True.

That seems to address the main issue where the fallback path refreshed embedding providers but did not force an LLM update after a flow reset/reapply.

One note: the issue prompt mentioned iterating all configured LLM providers similar to the embedding provider loop. This PR instead updates only the active/default config.agent LLM provider/model. That seems sufficient if the goal is to restore the active chat model after reset, but it is narrower than the originally suggested approach.