Release notes for v2.40.74

Overview

This release primarily adds a GFQL endpoint to the server, major improvements and optimizations to logging, as well as a few smaller features and the usual patching.

NOTE: This bundles and extends patches from hotfix release 2.40.65

Graphistry Versions

Server 2.40.74 (was 2.40.60)

JS React+Vanilla 5.1.1 (was 5.0.1)

Python PyGraphistry client 0.33.8 (was 0.33.0)

Third-Party Versions

Arrow 10.0.1

Caddy 2.7.6 (was 2.6.4)

CUDA (In-Docker) 11.5

Dask 2023.1.1

DGL 1.1.1

Django 3.2

Docker (CE) 23.0.1

Docker Compose 2.17.3

fsspec 2023.1.0

Gremlinpython 3.5.0

Jupyter 1.0.0

Neo4j-python-driver 4.4.5

Nginx               1.25.4 (was 1.23.3)

NodeJS 14.21.3

Pandas 1.5.3

Postgres 14.11 (was 14.7)

Python 3.8.16

neo4j node driver 4.0, 4.1, 4.4

networkX 2.6.3

notebook 6.5.2

Nvidia driver (cloud) 470

RAPIDS 23.02

Redis 6.2.7

Splunk node SDK 1.9.0

Tornado 6.2

torch 1.12.1

torch-geometric 2.3.1

torch-god 0.3.1

Features

• Add a new django command to display the completeness of user data particularly on the SSO data completeness checking.
• Enhanced metrics collection for FEP by adding operational counters, improving metric export rates and efficiency, and refining cloud backend cost optimization.
• Enhanced OpenTelemetry tracing for frontend calls to streamgl-gpu service, plus added user-controlled telemetry export flag.
• Introduced an endpoint with new query functionality (GFQL) that allows for multi-hop queries to identify paths and connections within the graph dynamically along with filtering capabilities for nodes and edges for graph data manipulation. 
• Allow personal key authentication via POST request in addition to GET.
• PyGraphistry:
  • featurize(), on error, coerces object dtype cols to .astype(str) and retries.
  • Various fixes.

GraphistryJS

• Add class ClientPKey which allows a user to login using personal ID and secret to @graphistry/client-api >= 5.1.0 . Requires corresponding upgrade of Graphistry server if authenticating using a personal ID and secret with an organization.

Infrastructure

• Enhance migration logic to support OpenTelemetry configuration and utilize the release script instead of directly invoking Docker Compose.
• Improved span exporting logic to reduce telemetry noise.
• Persist python packages installed by a user through the notebook environment.
• Improved the deployment process for the telemetry services.

Admin

Security

• More uniform authentication and authorization logging. Monitor for `[AUDIT]` at INFO (permit) and ERROR (deny) levels.

Fixes & tweaks

• Fix continuous encoding not encoding using the full color array provided.
• Fix encoding bug where incorrect color mappings can be used if there is a mismatch between number of bins and number of mapping categories.

Docs

• Added GFQL quick start guide
• Added GFQL usage guide
• Added GFQL endpoint API documentation along with supporting documentation on GFQL operators and predicates

Migration

• Existing systems should update to source telemetry configuration from `data/config/telemetry.env`. Additionally, consider using the `release` utility for deploying the telemetry services.
• Note next release (2.41.X) will be CUDA 12.X
• See previous notes