feat: add LightRAG + DocumentDB playground#333
Draft
xgerman wants to merge 4 commits intodocumentdb:mainfrom
Draft
feat: add LightRAG + DocumentDB playground#333xgerman wants to merge 4 commits intodocumentdb:mainfrom
xgerman wants to merge 4 commits intodocumentdb:mainfrom
Conversation
Add a self-contained playground demonstrating LightRAG (graph-based RAG engine) using DocumentDB as its MongoDB-compatible storage backend. Includes: - Helm chart with init-container that patches LightRAG for DocumentDB compatibility (skips unsupported createIndex with collation) - Ollama deployment manifest for in-cluster LLM inference - Automated deploy.sh and cleanup.sh scripts - Comprehensive README with architecture, setup, configuration, DocumentDB compatibility matrix, and troubleshooting guide Storage mapping: - KV, Graph, DocStatus → MongoKVStorage/MongoGraphStorage (DocumentDB) - Vectors → NanoVectorDBStorage (local, since DocumentDB lacks $vectorSearch) Tested end-to-end on Kind with DocumentDB Kubernetes Operator. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Replace manual service/secret lookups with the connection string from the DocumentDB resource status field, matching the pattern documented in the official networking docs. The status field contains embedded kubectl commands that are resolved via eval. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Document the three patched methods, why each is needed for DocumentDB compatibility, how the init container applies the patches, and the impact on functionality. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Findings from AKS E2E testing of the KEDA playground apply here too: - Fix eval quoting: use eval "echo \"...\"" to prevent & in connection string query params from being interpreted as shell background operator - Replace ClusterIP with DNS name for cross-namespace service resolution (status.connectionString uses ClusterIP) - Update README eval example to use the corrected two-step pattern Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds a self-contained playground demonstrating LightRAG — a graph-based RAG engine — using DocumentDB as its MongoDB-compatible storage backend.
What's included
README.mdhelm/lightrag/helm/ollama.yamlhelm/lightrag-values.yamlscripts/deploy.shscripts/cleanup.shDocumentDB compatibility
LightRAG's MongoDB storage layer requires patching for DocumentDB:
createIndexwith collation$vectorSearch$listSearchIndexesStorage mapping
$vectorSearch)Tested
End-to-end on Kind cluster with DocumentDB Kubernetes Operator: