Commit Graph
17 Commits
Author SHA1 Message Date
ricoromaandClaude Opus 5.5 96d2ecca3d docs: re-capture track2-reconciler evidence on 84f8c35
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-22 21:31:49 +03:00
ricoromaandClaude Opus 5.5 84f8c353e9 fix: address track2-reconciler verify findings
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-22 21:04:01 +03:00
ricoromaandClaude Opus 5.5 a6071dafa0 docs: add track2-reconciler evidence and report
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-22 20:23:01 +03:00
ricoromaandClaude Opus 5.5 569a16492a docs: re-capture upload-memory failure checks on 9d2ee30
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-22 16:17:53 +03:00
ricoromaandClaude Opus 5.5 9d2ee30cd2 fix: address upload-memory verify findings
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-22 16:11:03 +03:00
ricoromaandClaude Opus 5.5 79cc06b1e0 docs: add upload-memory evidence and report
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-22 15:39:58 +03:00
ricoromaandClaude Opus 5.5 511e28590b docs: record upload-memory benchmark hypotheses before runs
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-22 00:23:37 +03:00
ricoromaandClaude Opus 5.5 9dde16e638 docs: re-capture upload-consistency evidence on 5889ca2
Re-run failure checks (32 of 32) and the cleaner pass under a paused S3
on the two-step cleanup, fix the storage-consistency scenario count in
the student test evidence and point the report at the new commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-21 22:17:53 +03:00
ricoromaandClaude Opus 5.5 5889ca2ad7 fix: delete the object and the row in one locked cleanup transaction
Keep the committed mark, then delete the object and the row under the
same row lock, so an older replica without the mark check that waits on
the row finds it gone instead of promoting it to ready after the object
delete. A mixed-version stand is no worse than before the mark; document
the upgrade order (replicas first, cleaner last) and fix stale wording.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-21 22:08:29 +03:00
ricoromaandClaude Opus 5.5 6551839807 docs: add upload-consistency verify evidence and report
Capture the cleaner pass under a paused S3 (row stays marked pending,
next pass removes it), re-run failure checks (32 of 32 with red runs of
the new checks), student tests and the isolated run on 87c48b9, and
describe the marked three-step cleanup and its rolling-update window in
the report.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-21 21:42:31 +03:00
ricoromaandClaude Opus 5.5 f2dc0b368a docs: add upload-consistency evidence and report
- evidence: consistency (driver upload-consistency), cleaner process,
  cleaner S3 key, pending index plan on the working database
- re-run failure_checks (29/29), student_test via LB and test_isolated
  with the cleaner and storage-consistency coverage
- REPORT: database and S3 consistency, compensation table, grace period
  rationale, "PUT succeeded, COMMIT did not" answer, level 4 rows,
  limitations; PLAN stage 3 and AGENTS evidence map

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-21 06:32:48 +03:00
ricoromaandClaude Opus 5.5 6970dad871 docs: re-measure attachments rollout on a clean tree and update report
- bench_rollout label attachments from e1f8531 (dirty: false):
  replacement 8.67 s, 0 errors, 12001 messages = keys
- student_test through LB (197 passed) and failure_checks (24/24)
  after verify fixes
- REPORT: upload memory ~1.2 bodies with fixed mmap threshold,
  strict UTF-8 filename, env does not change attachment limits

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-21 02:26:35 +03:00
ricoromaandClaude Opus 5.5 8dc925e8ca docs: add attachments evidence and report
- re-capture coverage-exclusion evidence on the lab 4 stand: cache,
  auth storage, replay after API recreate, test-persistence,
  attachment-failover (stop, kill), attachment-storage
- green evidence/tests: student_test via LB, test_isolated,
  failure_checks; smoke, slow transfer, test-failover per replica
- bench_rollout label attachments: e1161f7 -> schema with attachments
  under load, no errors, messages = keys
- REPORT.md: attachments architecture, write order, defense answers,
  level 3/4 evidence rows, limitations; PLAN.md stage 2

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-21 01:50:10 +03:00
ricoromaandClaude Opus 5.5 1471f07a0e docs: re-capture s3-storage evidence after verify fixes
- evidence/s3 and evidence/tests re-captured from f57635a: first write
  after healthy, deleted bucket, netstat-based port check, readiness
  with s3 paused, prefix cleanup check
- PLAN: HEAD on a deleted bucket is an accepted limitation
- KNOWN_ISSUES: bootstrap after a manual bucket delete

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-20 21:47:01 +03:00
ricoromaandClaude Opus 5.5 f57635a5fb fix: address s3-storage verify findings
- s3 healthcheck and fault hook probe wait for the volume server to
  register with master (first write after restart no longer waits ~5 s)
- s3-bootstrap mounts /data as tmpfs: no anonymous volume per container
- S3 client: NoSuchBucket and AccessDenied are 503 with their own error
  logs, not a missing object or rejected credentials; delete ignores only
  a missing key
- test_isolated verifies prefix removal (weed shell exits 0 on errors)
- s3_check derives ports from listening sockets in the s3 container
- readiness without S3 is also checked with s3 paused

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-20 21:23:38 +03:00
ricoromaandClaude Opus 5.5 7d340c0673 docs: add s3-storage evidence and report
- evidence/s3: bootstrap, privacy (make s3_check), gRPC without token,
  restart and SIGKILL, client failures, memory of 20 concurrent uploads
- evidence/tests: failure_checks with own S3, test_isolated prefix
  cleanup (red tests: no attachment_ids until attachments)
- REPORT.md: S3 role, services table, bootstrap, readiness
  dependencies, skeleton S3_* names, evidence rows, limitations
- refine expected-red counts in tasks, design and PLAN

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-20 20:31:35 +03:00
Mikhail Verkovykh ec540a1a72 Подготовить skeleton лабораторной 4 по мессенджеру 2026-09-10 18:01:35 +03:00