Skip to content

fix(directory): bound first provider lookup failures and align active configuration docs - #94

Merged
servicememeai merged 8 commits into
devfrom
fix/provider-stability-20261025
Oct 25, 2026
Merged

servicememeai merged 8 commits into
devfrom
fix/provider-stability-20261025

Conversation

@servicememeai

Copy link
Copy Markdown

Goal

Keep the accepted CF_DIRECTORY_PROVIDER > CF_ACL_PROVIDER > authenticated Directory declaration > external resolution unchanged. Add only a process-local cold-lookup guard: a single concurrent Directory request, success/legacy missing declaration pinned per worker, unavailable responses throttled for 30 seconds and retried later (503 while unavailable). No per-request recurring HTTP, new setting, table, Redis entry, timer or cross-system identity engine.

Update the existing generic docs/ENTERPRISE-DIRECTORY-CONTRACT.md and docs/CONFIGURATION.md rather than creating a customer-specific document. Correct documented FILE_LOCK/CHECKOUT default (both false); Directory-before-Hub rollout remains recommended, then Hub restart picks up the declaration.

Cheap pure unit coverage uses existing CI-registered test_subject_adapter_pure.py: concurrent one-shot, failure throttle/retry, old Directory fallback. No deployment/DEV access. EAP provider remains etech with existing data, no migration.

@servicememeai
servicememeai merged commit c452664 into dev Oct 25, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant