LCORE-951: example configuration files #770
Conversation
WalkthroughThis PR introduces six new YAML configuration example files for the Lightspeed Core Service, demonstrating various feature configurations including RAG-backed retrieval, conversation caching with PostgreSQL and SQLite backends, MCP server integration, and quota limiting with both database backends. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~15 minutes
Possibly related PRs
Suggested reviewers
Poem
Pre-merge checks and finishing touches✅ Passed checks (3 passed)
✨ Finishing touches🧪 Generate unit tests (beta)
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. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 5
🧹 Nitpick comments (3)
examples/lightspeed-stack-conversation-cache-pg.yaml (1)
19-27: Add documentation and address hardcoded credentials in examples.Example configurations expose hardcoded credentials (password:
123qwe, api_key:xyzzy). While acceptable in theexamples/directory for non-production use, adding inline comments would help users understand:
- When/how to replace these test values
- What each PostgreSQL connection parameter means
- Why
ssl_modeandgss_encmodeare set todisableConsider adding a comment block at the top or inline explaining the example nature and credential replacement.
conversation_cache: type: "postgres" postgres: + # Test/example credentials—replace with secure values in production host: 127.0.0.1 db: test user: tester password: 123qwe + # SSL/GSS disabled for local development; enable in production ssl_mode: disable gss_encmode: disableexamples/lightspeed-stack-mcp-servers.yaml (1)
9-12: Add documentation for api_key.The
api_keyat line 12 contains a placeholder test value (xyzzy). Consider adding an inline comment clarifying this is for development/examples only.llama_stack: use_as_library_client: false url: http://localhost:8321 + # Test api_key—replace with actual credentials in production api_key: xyzzyexamples/lightspeed-stack-byok-rag.yaml (1)
37-47: Add comments clarifying RAG configuration and file paths.The RAG configuration uses hardcoded paths in
/tmp/which may not persist or could conflict. Consider adding inline documentation explaining:
- The purpose of each RAG configuration
- How to customize db_path for production use
- The role of embedding_dimension and vector_db_id fields
byok_rag: + # Example RAG configurations using local FAISS vector stores - - rag_id: ocp_docs + - rag_id: ocp_docs # Example: OpenShift documentation RAG rag_type: inline::faiss embedding_dimension: 1024 vector_db_id: vector_byok_1 + # Customize db_path for production; /tmp is used only for examples db_path: /tmp/ocp.faiss - - rag_id: knowledge_base + - rag_id: knowledge_base # Example: Generic knowledge base RAG rag_type: inline::faiss embedding_dimension: 384 vector_db_id: vector_byok_2 db_path: /tmp/kb.faiss
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (6)
examples/lightspeed-stack-byok-rag.yaml(1 hunks)examples/lightspeed-stack-conversation-cache-pg.yaml(1 hunks)examples/lightspeed-stack-conversation-cache-sqlite.yaml(1 hunks)examples/lightspeed-stack-mcp-servers.yaml(1 hunks)examples/lightspeed-stack-quota-limiter-pg.yaml(1 hunks)examples/lightspeed-stack-quota-limiter-sqlite.yaml(1 hunks)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (4)
- GitHub Check: build-pr
- GitHub Check: Konflux kflux-prd-rh02 / lightspeed-stack-on-pull-request
- GitHub Check: e2e_tests (azure)
- GitHub Check: e2e_tests (ci)
| quota_handlers: | ||
| sqlite: | ||
| db_path: quota.sqlite | ||
| limiters: | ||
| - name: user_monthly_limits | ||
| type: user_limiter | ||
| initial_quota: 50 | ||
| quota_increase: 50 | ||
| period: "30 seconds" | ||
| - name: cluster_monthly_limits | ||
| type: cluster_limiter | ||
| initial_quota: 100 | ||
| quota_increase: 100 | ||
| period: "30 seconds" | ||
| scheduler: | ||
| # scheduler ticks in seconds | ||
| period: 10 |
There was a problem hiding this comment.
Fix period value format inconsistency (same as quota-limiter-pg).
The scheduler period (line 36) uses bare 10 while limiter periods use "30 seconds" strings. Align the format for consistency across all example configs.
scheduler:
# scheduler ticks in seconds
- period: 10
+ period: "10 seconds"📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| quota_handlers: | |
| sqlite: | |
| db_path: quota.sqlite | |
| limiters: | |
| - name: user_monthly_limits | |
| type: user_limiter | |
| initial_quota: 50 | |
| quota_increase: 50 | |
| period: "30 seconds" | |
| - name: cluster_monthly_limits | |
| type: cluster_limiter | |
| initial_quota: 100 | |
| quota_increase: 100 | |
| period: "30 seconds" | |
| scheduler: | |
| # scheduler ticks in seconds | |
| period: 10 | |
| quota_handlers: | |
| sqlite: | |
| db_path: quota.sqlite | |
| limiters: | |
| - name: user_monthly_limits | |
| type: user_limiter | |
| initial_quota: 50 | |
| quota_increase: 50 | |
| period: "30 seconds" | |
| - name: cluster_monthly_limits | |
| type: cluster_limiter | |
| initial_quota: 100 | |
| quota_increase: 100 | |
| period: "30 seconds" | |
| scheduler: | |
| # scheduler ticks in seconds | |
| period: "10 seconds" |
🤖 Prompt for AI Agents
In examples/lightspeed-stack-byok-rag.yaml around lines 20 to 36, the scheduler
period value uses a bare numeric 10 while limiter periods use string durations
like "30 seconds"; update the scheduler period to match the same string duration
format (e.g., "10 seconds") for consistency across example configs.
| conversation_cache: | ||
| type: "sqlite" | ||
| sqlite: | ||
| db_path: quota.sqlite |
There was a problem hiding this comment.
Fix YAML indentation inconsistency.
Line 22 uses 6 spaces of indentation instead of 4. For consistency with standard YAML conventions and the PostgreSQL variants (e.g., lightspeed-stack-conversation-cache-pg.yaml), nested keys should use 2-space increments.
conversation_cache:
type: "sqlite"
sqlite:
- db_path: quota.sqlite
+ db_path: quota.sqlite📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| conversation_cache: | |
| type: "sqlite" | |
| sqlite: | |
| db_path: quota.sqlite | |
| conversation_cache: | |
| type: "sqlite" | |
| sqlite: | |
| db_path: quota.sqlite |
🤖 Prompt for AI Agents
In examples/lightspeed-stack-conversation-cache-sqlite.yaml around lines 19 to
22, the nested key indentation is inconsistent (line 22 is indented 6 spaces);
update the YAML to use 2-space increments for nested keys so that "sqlite:"
remains indented 2 spaces under "conversation_cache:" and "db_path:" is indented
4 spaces under "sqlite:" (i.e., change line 22 from 6 spaces to 4 spaces).
| period: "30 seconds" | ||
| - name: cluster_monthly_limits | ||
| type: cluster_limiter | ||
| initial_quota: 100 | ||
| quota_increase: 100 | ||
| period: "30 seconds" | ||
| scheduler: | ||
| # scheduler ticks in seconds | ||
| period: 10 |
There was a problem hiding this comment.
Fix inconsistent period value format.
The scheduler period (line 42) uses a bare number 10 while the limiter periods (lines 34, 38) use string format "30 seconds". For consistency and clarity, align the format—either use numeric values with a clarifying comment, or use string format with units.
scheduler:
# scheduler ticks in seconds
- period: 10
+ period: "10 seconds"Alternatively, if the schema expects numeric values only in this context, update the limiter periods to match:
- name: user_monthly_limits
type: user_limiter
initial_quota: 50
quota_increase: 50
- period: "30 seconds"
+ period: 30📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| period: "30 seconds" | |
| - name: cluster_monthly_limits | |
| type: cluster_limiter | |
| initial_quota: 100 | |
| quota_increase: 100 | |
| period: "30 seconds" | |
| scheduler: | |
| # scheduler ticks in seconds | |
| period: 10 | |
| period: "30 seconds" | |
| - name: cluster_monthly_limits | |
| type: cluster_limiter | |
| initial_quota: 100 | |
| quota_increase: 100 | |
| period: "30 seconds" | |
| scheduler: | |
| # scheduler ticks in seconds | |
| period: "10 seconds" |
🤖 Prompt for AI Agents
In examples/lightspeed-stack-quota-limiter-pg.yaml around lines 34 to 42, the
scheduler period uses a bare number (10) while the limiter periods use string
values ("30 seconds"); make formats consistent by changing the scheduler period
to the same string format — replace 10 with "10 seconds" (or alternatively
convert the limiter periods to numeric seconds if the schema requires numbers)
so all period fields use the same type and unit.
| quota_handlers: | ||
| sqlite: | ||
| db_path: quota.sqlite |
There was a problem hiding this comment.
Fix YAML indentation consistency (SQLite block).
Line 22 uses 6 spaces indentation, inconsistent with standard 4-space YAML nesting. This matches the issue in lightspeed-stack-conversation-cache-sqlite.yaml. Ensure all nested keys follow 2-space increments.
quota_handlers:
sqlite:
- db_path: quota.sqlite
+ db_path: quota.sqliteCommittable suggestion skipped: line range outside the PR's diff.
🤖 Prompt for AI Agents
In examples/lightspeed-stack-quota-limiter-sqlite.yaml around lines 20–22, the
db_path line is indented with 6 spaces which is inconsistent; change the YAML
indentation so nested keys use 2-space increments (i.e., "sqlite:" indented 2
spaces under "quota_handlers:" and "db_path:" indented 2 spaces further under
"sqlite:") to make indentation consistent with the other YAML examples.
| scheduler: | ||
| # scheduler ticks in seconds | ||
| period: 10 |
There was a problem hiding this comment.
Fix period value format inconsistency.
The scheduler period (line 36) uses bare 10 while limiter periods (lines 28, 32) use "30 seconds" strings. This inconsistency appears across multiple example files. Standardize the format.
scheduler:
# scheduler ticks in seconds
- period: 10
+ period: "30 seconds" # or align to numeric format across all examples📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| scheduler: | |
| # scheduler ticks in seconds | |
| period: 10 | |
| scheduler: | |
| # scheduler ticks in seconds | |
| period: "10 seconds" |
🤖 Prompt for AI Agents
In examples/lightspeed-stack-quota-limiter-sqlite.yaml around lines 34 to 36,
the scheduler period is written as a bare number (10) while other periods in the
file use string durations like "30 seconds"; update the scheduler period to the
same string duration format (e.g., "10 seconds") so all period values are
consistent, quoted, and parsable as durations.
1 participant
Description
LCORE-951: example configuration files
Type of change
Related Tickets & Documents
Summary by CodeRabbit