Files
admin/.wiki/log.md
vitya 063910e265 incident(books-vds-es): true RCA — ransom-бот через открытый :9200, не оператор
Рецидив вчерашнего ES-инцидента вскрыл истинную причину. Вчерашняя
гипотеза «оператор в cutover попал на canonical» опровергнута.

Root cause: ES (stack 33) публиковал 0.0.0.0:9200 мимо traefik. Free-ES
7.10 без auth → порт открыт всему интернету. Ransom-бот сносил индексы
by-name (мимо Control #1 destructive_requires_name), оставлял read_me с
BTC-выкупом. accessLog (Control #2) пуст — бот шёл прямо в порт, не через
traefik. firewalld бесполезен (docker-publish обходит INPUT-зоны).

Fix (Control #3): убрана публикация host-порта из stack 33 (Portainer
PUT), дыра закрыта; re-restore epz/products/artmone из daily-2026-05-25.

Отдельный баг: epz-поиск падал у ОБОИХ тенантов — getTenantIdSeller(
config.get("tenant")) через node-config, а tenant не задан ни в
default.json, ни в env-маппинге (TENANT env = мёртвый груз). Добавлен
tenant в overlay default.json (slovo/bookva). products работал —
отдельный код-путь.

accessLog откатан (сторожил не ту дверь).

- wiki concept: correction-блок + секция «Рецидив 2026-05-29» + exposure-audit
- tasks: restore-es reopened+reclosed; new  harden-books-vds-exposed-ports
- NEXT_SESSION handoff

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-05-29 09:31:25 +03:00

34 lines
5.9 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# Wiki op-log
Append-only log of wiki operations (ingests, promotions, lints, migrations).
## [2026-05-28] ingest | concepts/es-destructive-delete-incident-2026-05-26 — RCA для удаления 3 user-индексов на canonical ES `elasticsearch.kzntsv.site` (books VDS stack 33) во время cutover-prep. Caller identity unrecoverable (audit log = X-Pack платный, traefik accessLog был выключен). 2 preventive controls applied + verified: ES env `action.destructive_requires_name=true` + traefik JSON accessLog. Источник: `.tasks/restore-elasticsearch-indices-books-vds.md` § Closure note.
## 2026-05-21
- bootstrap: empty wiki skeleton (CLAUDE.md, index.md, log.md, overview.md, raw/README.md) committed
## [2026-05-21] ingest | concepts/admin-infra-project
## [2026-05-21] seed | CLAUDE.md Domain conventions — design-context pointers (task: admin-infra-project-pointers)
## [2026-05-21] migrate | subtree-import 6 entities + 17 concepts + 3 sources from MoreThenCms (history preserved via subtree-split + temp-prefix merge)
## [2026-05-21] regen | index.md catalog refresh after subtree import
## [2026-05-21] ingest | concepts/ocis-on-vds-deploy-recipe — oCIS deploy recipe + gotchas (UID mismatch, PROXY_TLS, PROXY_ENABLE_BASIC_AUTH, LibreGraph user-create); updated entities/vds-kzntsv (added owncloud row to stack/hostnames/file-layout); source: .tasks/owncloud-vds-deploy.md
## [2026-05-22] update | concepts/ocis-on-vds-deploy-recipe — Gotcha 5: 60s HTTP timeout caps slow-uplink uploads at ~200MB on 3 MiB/s link; traefik buffering NOT the fix (500 from oxy buffer); workaround = VDS-side curl PUT loopback via throwaway sftp key. Source: .tasks/owncloud-vds-deploy.md close-note.
## [2026-05-23] ingest | concepts/windows-server-2025-core-bootstrap — default-blockers (SMB closed, IIS/URL-Rewrite/.NET unverified) + transfer-методов матрица (RDP-redirect / SMB / WinRM / SFTP); recommendation = SMB inbound с source-IP whitelist на время migration. Source: failed-robocopy инцидент 2026-05-23 18:08 (.tasks/iis-migration-to-ruvds.md Decisions log).
## [2026-05-24] ingest | concepts/books-ssh-access — SSH access audit на shared VDS pre-cutover Фазы 3 tenant-split. Retained keys table (1 key, vitya core dev), removed cosmetic dead root key, sshd hardening verified via `sshd -T` (not raw config), fail2ban active 2670 failed/37 banned, only vitya@94.19.247.14 в access log last 7 days. Source: .tasks/books-ssh-audit-shared-vds.md.
## [2026-05-24] ingest | RUVDS IIS migration + backup pipeline — entities/ruvds-iis-host (NEW, 80.64.31.36 Win Server 2025 Core, 25 SNI bindings, 2/24 hostnames DNS-flipped) + sources/iis-migration-to-ruvds-2026-05-23 (NEW, SSH/scp pivot после SMB-block by home-ISP) + sources/ruvds-backup-daily-kreknin-2026-05-24 (NEW, rclone+SFTP SYSTEM task daily 04:30) + concepts/traefik-acme-json-to-iis-cert-import (NEW, PFX+SNI recipe) + UPDATE concepts/windows-server-2025-core-bootstrap (SMB deprecate, HTTP middlebox warning, HTTP/2 note, backup-strategy + cert-import закрыты) + UPDATE entities/windows-recovery-host (IIS partial-cutover state, imgproxy SPOF carve-out) + UPDATE overview (RUVDS line). Sources: .tasks/iis-migration-to-ruvds.md, .tasks/ruvds-backup-daily-kreknin.md.
## [2026-05-28] ingest | vds-kzntsv DHCP outage postmortem — concepts/vds-kzntsv-dhcp-outage-2026-05-28 (NEW, симптомокартина + диагностический алгоритм + recovery runbook + RCA) + sources/vds-kzntsv-incident-2026-05-28 (NEW, timeline 05:25 backup OK → 08:15 detect → 08:43 statics fix → 08:50 disk GC; ticket text) + UPDATE entities/vds-kzntsv (Доступ §+subnet/gw/hypervisor/DNS resolvers; pass-store вместо .common/secrets; Known issues §NEW with 2026-05-28 incident; Open issues bump GC priority + iputils-ping note) + UPDATE concepts/rusonyx-vps-onboarding-quirks (quirk #9 NEW — gw в /18 надсети + recovery commands). Live session, no raw source ingested.
## [2026-05-28] update | vds-kzntsv DHCP outage post-resolution — после reset хостером в 13:52 MSK выяснилось что root cause — конфликт двух сетевых стеков (netplan+networkd поверх ожидаемого provider's ifupdown). Их `start/ipadd` ожидает чистый ifupdown, не мог auto-recover при разрыве DHCP binding. Resolution: mask netplan+systemd-networkd, reboot, provider положил `/etc/network/interfaces.d/ifcfg-eth0` с /18 netmask. UPDATE concepts/vds-kzntsv-dhcp-outage-2026-05-28 (revised RCA + permanent-fix § + anti-pattern + revised lessons-learned) + UPDATE sources/vds-kzntsv-incident-2026-05-28 (timeline до 14:00 + final config) + UPDATE entities/vds-kzntsv (mask /18, ifupdown stack, kernel cmdline net.ifnames=0) + UPDATE concepts/rusonyx-vps-onboarding-quirks (quirk #9 переписан про /18, quirk #10 NEW про ifupdown vs netplan stack). Memory `vds-kzntsv-rusonyx-network-recovery` переписан с новыми фактами.
## [2026-05-29] update | ES destructive-delete RECURRENCE — true root cause найден, диагноз сменился. UPDATE concepts/es-destructive-delete-incident-2026-05-26 (correction-блок + Hypothesis помечена ОПРОВЕРГНУТА + новая секция «Рецидив 2026-05-29 — true root cause»: ransom-бот через открытый `0.0.0.0:9200` мимо traefik+basicAuth, ES 7.10 free без auth; `read_me`=BTC-выкуп; снос by-name мимо Control #1; accessLog пуст т.к. бот шёл прямо в :9200; firewalld bypass docker-publish; Control #3 = убрать публикацию host-порта via Portainer PUT stack 33; restore из daily-2026-05-25; exposure-audit таблица — mongo/books-db/bookva-db/minio/bookva-minio тоже exposed но credentialed). bookva не пострадал (bookva-es порт не публикует). Live incident response session.