Repository navigation
fix(etw): recover after collector and guard loss - #969
Merged
Merged
Conversation
7 of 8 tasks
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Losing both the diagnostic ETW collector and its primary guard left no surviving elevated owner that could recover the original session. Start an independent fallback before StartTrace and reserve the original per-capture GUID over mutually authenticated restricted pipes. It retains both original peer objects, waits for their exits, rechecks session identity and stops only by the matching handle. The broker validates fallback receipts and exits before main can permit explicit retry. Capture remains FAILED with unknown stage totals and an observation gap; production ETW remains disabled by default.
Add four fixed own-process crash cases (both ready loss orders, before arm and before start) to an eleven-capture packaged profile. Single guard loss also gains verified fallback cleanup. Broker loss continues to block retry without an authenticated terminal. No automatic capture/UAC retry, name-based stop, elevated kill, renderer IPC, workflow, lockfile or machine-setting change.
Type of change
Testing
5a09b16ae2140ee91cb1772bc23aca7173f10603, including independent session presence/absence, original-peer exits, actual Sensors FAILED state, sync/pending-overlapped read witnesses and a subsequent normal capture. Portable evidence binds source/tree, receipts and package hashes. Final commit changes documentation/evidence only.b45faaa84d7c870f45964df66605a15e03db6f66: CI, Windows fixtures. These hosted checks are separate from the live packaged evidence above.ENOENT); no scanner success is claimed. Two-pass self-review completed; no independent human review is claimed.Loss of all three elevated peers, protected durable recovery/service authority, protected installation, alternate credentials, actual sleep/wake, full I/O coverage and production/path admission remain unqualified. Faults cover both coordinated loss orders; instruction-level simultaneous termination is not claimed. Raw profiles/screenshots/receipts remain private under the existing manual seven-day review policy. The dirty primary checkout's seven tracked hashes and complete status were unchanged.
Checklist