The hardening refactor + uid-collapse make the "what does left4me own
vs. ckn-bw own" question more pointed. The 2026-05-06 deployment
design already framed this: deploy/files/ in left4me mirrors target
paths, configmgmt integrates. Some artifacts have drifted into the
ckn-bw reactor since (systemd unit emissions, sysctl defaults); the
brainstorming session reconciles.
Sequenced after uid-collapse. Self-contained for a fresh Claude
session to read cold via superpowers:brainstorming.
Session-handoff updated to point at this as the next-next queued work.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The hardening refactor that just landed closes the same-uid attack
surface (FS view, ptrace, /proc visibility, signals) for the web +
gameserver units via systemd directives plus system-wide
kernel.yama.ptrace_scope=2. Keeping the script-sandbox on a separate
uid was the inconsistent half-step — defense-in-depth only, with
build-time-idmap complexity attached. One principle wins: harden
once, share the uid.
scripts/libexec/left4me-script-sandbox: drop the idmap block (uid
lookups, STAGING setup, cleanup_staging trap, mount --bind
--map-users), switch User=/Group= to left4me, point BindPaths at
\$OVERLAY_DIR directly. Header comment updated to reflect
hardening-not-uid as the same-uid defense. nsenter self-wrap kept —
it's about mount-namespace escape, not uid.
Tests + comments + companion docs updated. Build-time-idmap and
overlay-idmap plans marked SUPERSEDED; user-uid-split spec revised
to "1 user is correct"; one-line update notes on the hardening
specs and the build-overlay-unit-design.
Companion ckn-bw commit removes the l4d2-sandbox user + group and
tightens /var/lib/left4me from 0711 → 0755 (the traverse-only mode
was specifically for the sandbox uid).
Approved-but-not-executed plan to collapse the two-user model
(left4me + l4d2-sandbox) into one. The build-time-idmap that
translates sandbox writes back to left4me uid becomes a no-op when
source uid == target uid, so it's removed along with ~30 lines of
helper plumbing. Hardening already covers the same-uid attack
surface the sandbox uid was defending against, so collapsing makes
the architecture consistent with the web/server hardening-only
decision.
Plan: docs/superpowers/plans/2026-05-15-uid-collapse.md
Handoff: docs/superpowers/specs/2026-05-15-session-handoff.md
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The 1/2/3-user question is answered: stay at 2 (left4me + l4d2-sandbox).
The defenses that motivated a 3-user split (cross-uid ptrace,
cross-server contamination, web-side reach into gameserver state,
DB/env exposure to srcds) are closed by the systemd hardening
composition: PrivateUsers + PrivatePIDs + TemporaryFileSystem +
SystemCallFilter=~@debug + empty CapabilityBoundingSet.
The residual filesystem-ACL surface (mode 0640 root:left4me on DB and
web.env) is noted as a separate concern — covered for the current
deployment shape, revisit if shape changes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Four corrections noted by the test plan's executor in commit 461b8d0:
- PID-lookup race: pgrep+head can pick the wrong instance. Replace
with systemctl show -p MainPID --value left4me-server@N.service.
- gdb-from-host ptrace check: nsenter into only the mount namespace
with root caps bypasses the SECCOMP filter, so the test is a false
positive. Replace with systemd-run-with-same-directives probe, or
syscall-filter inspection.
- D5 pgrep pattern: 'srcds_linux.*\@2' doesn't match because @N is
in the unit name, not argv. Use systemctl show -p MainPID.
- scmp_sys_resolver is in the seccomp package on Debian 13, not
libseccomp-dev.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Verified during plan execution that the ckn-bw systemd-bundle emitter
handles tuples and empty values as expected. SocketBindAllow port
range hard-coded since systemd directive variable substitution is not
universal.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
12 tasks across left4me + ckn-bw: emitter verification, three Python
constants in the systemd_units reactor, spread into both managed units,
sysctl drop-in, annotated reference units, four spec bug fixes, mark
uid-split spec superseded, cross-repo push, bw apply + verify on host,
apt-remove test tooling. Each task has bite-sized steps with exact
commands and expected output.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Going back to the inline-in-reactor shape: hardening directives land in
ckn-bw's systemd_units reactor as shared Python dicts (HARDENING_COMMON
+ HARDENING_SERVER + HARDENING_WEB), spread into each unit's Service
block. Educational reference units in deploy/files/.../*.service stay
and get per-directive comments. Operator discipline hand-syncs the
reference to the reactor; no CI drift test.
The broader responsibility reshape — hardening drop-ins living in
left4me with ckn-bw as a thin file-shipper — is worth pursuing as a
separate dedicated session, not bundled into this refactor.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Hardening composition is application knowledge (which paths to bind, that
srcds is i386, what breaks sudo). It belongs in the left4me repo as
drop-in .conf files under deploy/files/etc/systemd/system/<unit>.d/.
ckn-bw shrinks: keeps the base units in its reactor, removes the
hardening keys, ships the drop-ins to /etc/systemd/system/. Existing
educational reference units in deploy/files/.../*.service are deleted in
favor of the drop-ins, which carry per-directive comments. Broader
configmgmt-responsibility reshape (base units leaving the reactor)
deliberately deferred to a future session.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The prior handoff pointed this session at running the test plan; that's
done (commit 461b8d0). Update the handoff to point the next session at
writing docs/superpowers/plans/2026-MM-DD-hardening-refactor.md against
the proven composition, including the two amendments (x86 arch,
PrivatePIDs) and the MDW permanent exclusion.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Ran the 11-test plan against left4me-server@1 (canary) and left4me-web
on left4.me / Debian 13 / systemd 257. Cleaned up all unit drop-ins;
kept the Test 9 sysctl (kernel.yama.ptrace_scope=2) per spec.
Outcomes:
- server@1 systemd-analyze: 7.5 EXPOSED → 1.3 OK
- left4me-web systemd-analyze: 8.7 EXPOSED → 4.1 OK
- All 8 attack vectors in Test 8 (D1.a-c, D2.a-c, D3, D5) blocked
- Test 6 (MemoryDenyWriteExecute) fails as predicted — Source engine
i386 .so files have text relocations; exclude from final composition.
- Test 11 (24-48h soak) skipped per operator decision.
Two amendments to the spec's proposed composition required for the
refactor:
- SystemCallArchitectures=native x86 (not bare 'native') — srcds_linux
is i386, the kernel kills every native-only call.
- PrivatePIDs=true added — ProtectProc=invisible alone cannot hide
gunicorn from srcds because both run as uid 980; PrivatePIDs gives
each instance its own PID namespace and closes D2.b.
Spec bugs surfaced and documented in the "Output" section: PID lookup
via pgrep (race vs. instance), Test 4/10 gdb-from-host doesn't
actually exercise the unit's SECCOMP filter, Test 8 D5 pgrep pattern
won't match. Tooling note corrected: scmp_sys_resolver is in
'seccomp' package, not 'libseccomp-dev'.
Next session: write docs/superpowers/plans/2026-MM-DD-hardening-refactor.md
against the proven composition; supersede the uid-split spec.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Reframe the queued uid-split decision into a broader hardening analysis.
Audit found the same-uid attack surface (DB readable from srcds, ptrace
allowed, RCON stored plaintext) is closable by either uid split or
systemd directive composition; the three specs ground that choice in a
threat model, survey the defenses, and lay out a self-contained test
plan to run on left4.me next. uid-split spec deferred pending results.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Short companion to the existing topic-specific handoff docs. Captures
the situationally-fresh state at the end of the 2026-05-15
deploy-dir-rethink + janitorial sweep so a fresh session can pick
up cold: what just landed, what's next (uid-split), what's NOT next
(build-overlay-unit, until uid-split decides), and the
decision-relevant signals that emerged during this session — mostly
that the 2-uid model was freshly load-bearing in the build-time-idmap
work and that srcds hardening already covers most of what a
gameserver-uid split would add.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Both items were operational verifications (not code changes) against
the deployed test host ovh.left4me (141.95.32.8).
Item 8: orphan idmap binds in PID 1's mount namespace.
`sudo findmnt --task 1 -o TARGET | grep /var/lib/left4me/runtime/.*/idmap/`
returned zero matches with left4me-server@{1,2}.service both active.
Either swept earlier or never appeared on this host; nothing to umount.
Item 9: Optimized Settings (overlay 8) files-overlay sanity.
Dir is left4me:left4me end-to-end; `sudo find /var/lib/left4me/overlays/8
-type f -uid 981` returned empty. The invariant "files-overlays are
populated by the web app as left4me, never through the sandbox helper"
holds.
Remaining live janitorial items: 7 (conditional on the build-overlay-unit
refactor) and 10 (SourceMod 1.13 calendar reminder, ~late 2026).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Janitorial item 6 in 2026-05-15-janitorial-cleanup.md. The v1 sandbox
design (2026-05-08-l4d2-script-overlays-design.md) was approved
2026-05-08 and superseded the same day by the v2 systemd-only design
(2026-05-08-l4d2-script-sandbox-v2-systemd.md). The current
left4me-script-sandbox helper uses systemd-run in service-unit mode;
no bwrap binary is invoked. The v1 spec still described bubblewrap as
the engine.
- v1 spec gets a top-of-file banner pointing at v2 as the supersede.
Body preserved; the rest of the v1 design (overlay-type unification,
resource caps, helper auth) is still valid — only the sandbox engine
changed.
- l4d2web/services/overlay_builders.py: ScriptBuilder docstring
"bubblewrap + systemd-run" → "hardened systemd-run transient
service" (the as-built reality).
- scripts/tests/test_script_sandbox.py: stray "/bwrap" in a comment
cleaned up. Negative regression assertions (`assert "bwrap" not in
text`) intentionally retained as the guard against accidental
re-introduction.
- Plan docs left untouched (historical action snapshots).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds the implementation plan that landed in the preceding commit
(2026-05-15-deploy-dir-rethink.md) under docs/superpowers/plans/, and
marks the two related specs:
- 2026-05-15-deploy-dir-rethink-design.md (the source handoff) gets a
"Resolved by …" banner at the top with a one-paragraph summary of
the decisions taken. Body preserved for archaeology.
- 2026-05-15-janitorial-cleanup.md gets a status banner noting that
items 1, 3, 4, 5 are fully resolved by the deploy-dir-rethink plan
and item 2 is partially resolved with a third option the original
enumeration didn't list: only the truly-dead two static units
(cake.service, nft-mark.service) deleted, the reactor-emitted set
(server@, web, workshop-refresh.{service,timer}, slices) retained
as curated examples. Resolved items left in place but flagged.
Remaining live janitorial items: 6 (bubblewrap doc drift), 7
(conditional on build-overlay-unit refactor), 8 (operational idmap
bind cleanup), 9 (Optimized Settings overlay verification), 10 (SM
1.13 calendar reminder).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two follow-ups bundled into a single commit:
- docs/superpowers/specs/2026-05-15-janitorial-cleanup.md collects
the "do later" small TODOs that surfaced across the recent idmap
+ consolidation work: dead cake-related artifacts, obsolete
static systemd units in deploy/files/, the bubblewrap→systemd-run
doc drift, stale gameserver-side idmap binds on un-checked
instances, calendar reminder for SM 1.13 stable. Each item is
small and self-contained.
- docs/l4d2-server-cvar-reference.md captures the research from
the early-session L4D2 cvar deep-dive: tickrate sweet spots,
nb_update_frequency cheat-protection + sm_cvar workaround,
cvars that don't exist in L4D2 (net_maxcleartime,
z_resolve_zombie_collision_multiplier per RCON probe), recommended
plugins, MetaMod/SourceMod branch tracking, and the empirically-
verified idmap-propagation-through-rebind kernel-6.12 quirk.
Reference material, not a spec — lives at docs/ root.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Explicit clarification so the next agent doesn't go looking for
user-unit friction. left4me-server@.service and left4me-web.service
are system units that drop to User=left4me; the 3-user split is a
literal one-line edit per unit. No lingering, no pam_systemd, no
per-user systemd instance bootstrap. The privileged
ExecStartPre/ExecStopPost steps stay root via the + prefix.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The 2-user split (left4me + l4d2-sandbox) has been inherited as a
constraint across multiple recent plans (idmap-on-mount, build-time-
idmap, helper consolidation) without ever being designed
end-to-end. Three plausible configurations: collapse to 1 user
(rejected for security), keep at 2 users (status quo), or split web
from game into 3 users for blast-radius limiting on either side.
Doc captures the threat-model heuristics, cross-uid file-access
plumbing options (shared group vs. world-read), idmap implications,
a step-by-step migration sketch for the 3-user variant, and explicit
out-of-scope items (per-instance gameserver uids, etc.). Detailed
enough that a future session can pick a configuration and execute
without re-deriving the design space.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The script content lives in the overlays.script DB column and the
unit's %i is the row id, so the worker-writes-script-to-fs step in
the original sketch is duplication. Document three options (worker
writes / unit fetches via helper / pipe to stdin) and recommend the
unit-fetches variant with RuntimeDirectory= auto-cleanup. Promote
this to the top of the open-decisions list since it shapes the
worker, the unit, and whether a fetch-script helper is added.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The build-time idmap landing today required a nsenter self-wrap in
left4me-script-sandbox to escape the web app's PrivateTmp namespace
before pre-creating the idmapped staging bind. Working but band-aid:
the helper is reinventing what a systemd template unit would do
declaratively. Mirror the left4me-server@.service pattern with a
build-overlay@.service template — ExecStartPre does the idmap bind in
PID 1's namespace by default, the hardening flags live in the unit
file, ExecStopPost tears down. Worker switches to sudo systemctl start.
Doc captures full proposed unit, worker rewrite sketch, sudoers
update, migration order, verification steps, and the ~5h estimate
so a future session can pick this up cold and execute.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Architectural cleanup: the uid translation is a build-time concern
(the sandbox produces sandbox-uid files); having the gameserver path
unwind that producer-side decision on every mount means the mount
helper carries idmap lifecycle code it shouldn't need. Moving the
idmap into the script-sandbox bind makes files land left4me-owned on
disk, drops ~140 lines from left4me-overlay, and makes all overlay
content (workshop + script-built) consistent on-disk.
Verified on left4.me kernel 6.12.86 that the kernel idmap propagates
through plain re-bind, so systemd-run's BindPaths can wrap a
pre-created idmapped staging path.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The 2026-05-15 script-consolidation pass landed a working but
half-finished mental model: deploy/files/ was retroactively promoted
from "historical reference" to "canonical source," but only for the
script files. Several adjacent things (sudoers/sysctl duplication
across both repos, the systemd unit files that ckn-bw's reactor
ignores, deploy-test-server.sh's role, dead-code apply-cake) didn't
get resolved. Capture the open questions and pointers so a future
session can pick this up and commit to a coherent shape.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Persist the implementation plan for adding idmapped bind mounts to
left4me-overlay so that overlay copy-up from l4d2-sandbox-owned lower
layers (script-built overlays) produces left4me-owned upperdir entries
the gameserver can write. Mechanism verified end-to-end on ovh.left4me
in a temp dir on 2026-05-14.
Plan for adding a per-server RCON console: HTMX append-swap input form,
fixed-height scrolling transcript replayed from CommandHistory on load,
multi-packet response handling, owner-only access, 30s timeout.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- _parse_duration wraps int() in try/except so malformed connected
durations raise RconError (not ValueError leaking past the poller's
except RconError).
- fake_rcon_server captures handler exceptions and re-raises at context
exit, so a buggy test handler surfaces as a real failure instead of
silently degrading into a client-side timeout.
- Two new parser tests: HH:MM:SS duration parsing and malformed input
coverage.
- Fix Steam ID formula typo in the spec doc (Z*2 + Y, not Y*2 + Z; Y is
the low bit). Code was already correct.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The two indexes ix_sps_server_open and ix_sps_server_recent were
byte-identical because SQLAlchemy's Index(name, *cols) form drops the
DESC ordering the spec intended. Rather than reach for text("left_at
DESC"), drop the second index entirely — SQLite scans the ASC index
backwards at no measurable cost. Spec and plan updated to match.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
RCON-based polling with run-length-encoded snapshots, session intervals
with min/max ping, Steam profile cache, and a server-detail roster of
current + recent players hot-linked from Steam CDN avatars.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Closes the gap where added workshop items never reach disk until an
admin presses the global refresh button. Downloads piggyback on the
per-overlay build_overlay job; daily updates come from a systemd
timer + CLI subcommand that enqueues the existing refresh job.
The matching design doc for the implementation plan committed in
6eb9bd0. Captures the session-invalidation reasoning (Django-style
"keep current session, kill others") and the open questions resolved
during brainstorming.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds /profile reachable via header username, with change-password form
as its first section. Industry-standard session semantics: other sessions
invalidated on password change, current session kept, via new
users.password_changed_at column + session marker.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Eight TDD tasks: sysctl extension, nftables marking (file + unit), CAKE
shaper (env + helper + unit), deploy-script wiring, README. Each task
adds one artifact with its assertion in test_deploy_artifacts.py and
ends in its own commit.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds Overlay.type='files' whose source-of-truth IS the overlay directory
itself. Users can:
* upload arbitrary files / whole folders by dragging from the OS onto a
folder row in the file tree (one POST per file, queue with
concurrency 3, per-file progress in a floating Uploads panel)
* move via drag-and-drop inside the tree (same gesture, source
distinguishes; refuses cycles)
* create / edit / rename / replace through a single editor modal
(text flavor for editable files, binary flavor with replace-upload
for everything else; filename input is the rename surface)
* mkdir empty folders (slashes allowed for nested intermediates)
* stream a folder as a zip download
* delete files and empty folders
Backend is type-agnostic past the new files_routes endpoints, so the
existing mount / spec / overlayfs / expose_server_cfg pipeline is reused
unchanged. is_editable gates the row's edit affordance and the /save
content rules. Three new safe-resolve helpers (write/delete/move) cover
the new operations with the same anchor-and-resolve pattern as listing
and download. FilesBuilder is a no-op so the build subsystem can
dispatch uniformly.
Spec: docs/superpowers/specs/2026-05-09-files-overlay-design.md
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Captures the symptom (Reset blew up on `umount target busy`), the
false starts (eager retry, lazy fallback, TimeoutStopSec bump — all
shipped briefly and reverted), the actual root cause (the helper's
own Python interpreter inheriting and pinning the unit's mount
namespace), and the fix (nsenter at the systemd Exec line).
The lessons section is the part future-me reads first: a retry loop
is a hint that something we own is the blocker; probe `/proc/*/ns/mnt`
before assuming kernel async; `+` Exec prefix doesn't escape the
unit's mount namespace.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Investigated whether to hard-pin each srcds instance to a single core
within the existing AllowedCPUs=1-7 set. Modern kernels (5.13+) no
longer expose kernel.sched_migration_cost_ns or the other classic CFS
"laziness" tunables, so a global cheap-fix is unavailable. Decision
for now: trust CFS + Nice=-5 + AllowedCPUs=1-7. Per-instance
CPUAffinity= remains an opt-in escape hatch in deploy/README.md.
Documents the revisit triggers and the preferred implementation path
when the time comes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Make explicit that the project uses system units (root systemctl, unit
under /usr/local/lib/systemd/system/, WantedBy=multi-user.target), so
`systemctl enable --now` is the correct verb to make instances survive
a host reboot. User units have different lifecycle rules and would not
auto-start at boot without enable-linger.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two TDD tasks: helper+service_control verb rename, then poller code
+ wiring + tests. Operator-side smoke test in F.3.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Switch lifecycle verbs from systemctl start/stop to enable --now /
disable --now (servers survive host reboot via WantedBy= symlinks),
plus a periodic state poller for runtime drift (OOM kills, manual
systemctl ops, exhausted Restart=on-failure).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two TDD tasks: deploy-script cpuset block + tests, README
"CPU isolation" subsection. Operator-side smoke test in F.3.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
cgroup-v2 AllowedCPUs= drop-ins for system/user/build/game slices.
Defaults: core 0 for everything-not-game, cores 1..N-1 for game,
computed from nproc. LEFT4ME_SYSTEM_CPUS / LEFT4ME_GAME_CPUS
overrides; single-core hosts skip with a warning.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Six tasks (TDD, one commit each): unit directives, slice files,
sysctl conf, sandbox slice + OOMScoreAdjust, deploy-script wiring,
README escape-hatch section. Final verification step with full
deploy + host + web pytest sweep.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
