Correct the 2-BM storage-tier account against the beamline's archival page - #696
Merged
Conversation
… page Read the beamline's data-management page, which turns out to be an in-flight archival status page rather than the standing how-it-works doc (item_018 is that one). Three things it settles. The tier has two names at the source: the beamline's current pages say DM, its older ones say Sojourner, and staff say Voyager in email. Only the /gdata mount is stable across all three, so the deployment page now leads with APS Data Management and records the alias rather than picking a brand and pretending the source is consistent. The upstream tiers are not merely transient. /data2 was lost outright on 2026-08-14 when the tomodata2 array failed, taking every dataset DM did not already hold, and the beamline now marks the survivors "DM is authoritative". Deleting a local copy DM already holds is gated on an emailed approval, not a schedule. That durability asymmetry is what decides where a durable Distribution has to point, so it belongs on the page rather than in a session's notes. 2026-08-Haridy-1015116 is already past the transition and lives only under /gdata. The same page supplies a real counterexample for an invariant mint_capture_path_locator so far stated only in the abstract: /data3 carries per-person workspaces directly at its root, so a root set to /data3/vnikitin would embed a username in the very segment the locator treats as safe to keep. Named it in the docstring, together with the confirmation that depth below the root genuinely varies by tier, which is why discarding everything between root and filename is the right rule. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Coverage report
Click to see where and how coverage changed
This report was generated by python-coverage-comment-action |
||||||||||||||||||||||||
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.
What
Corrects the 2-BM deployment page's storage-tier account against the beamline's own archival status page, and adds a concrete counterexample to an invariant
mint_capture_path_locatorhad only stated in the abstract.Three things the source settles
The tier has two names, and neither is the stable one. The beamline's current pages say DM, its older ones say Sojourner, and staff say Voyager in email. Only the
/gdatamount is common to all three, so the page now leads with APS Data Management and records the alias rather than picking a brand and implying the source is consistent.The upstream tiers are not merely transient.
/data2was lost outright on 2026-08-14 when the tomodata2 array failed, taking every dataset DM did not already hold. The beamline now marks the survivors "DM is authoritative", and deleting a local copy DM already holds is gated on an emailed approval rather than a schedule. That durability asymmetry decides where a durable Distribution has to point, so it belongs on the page.2026-08-Haridy-1015116is already past the transition and lives only under/gdata.A real counterexample for a documented invariant.
mint_capture_path_locatorembeds the matched root verbatim and requires it to be the facility-level tier, never a path containing personal data. The docstring said so abstractly./data3carries per-person workspaces directly at its root (/data3/vnikitin,/data3/sboyer), so a root pointed at one of those would embed a username in the very segment the locator treats as safe. Named in the docstring, along with the confirmation that depth below the root genuinely varies by tier, which is why discarding everything between root and filename is the right rule.Verification
mkdocs build --strictclean; architecture suite passes; ruff, pyright, tach clean. Docs-only plus one docstring, no behavior change.🤖 Generated with Claude Code