A client without the races patch crashes on a unit of race 12+, so the server now keeps
those characters out while custom races are off. The Custom Races card shows how many
player characters and bots that are (active realm, read only while the database runs) and
offers botcmd purgecustomraces for the companion bots.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The module list blocked Companions on every realm but Conquest of Azeroth, so the
multi-realm bot work could not be switched on. It is allowed (as experimental) when the
world server or its bots plugin carries the standard-class bot profiles of the multi-realm
bots build; SQUID and Dungeon Clear stay CoA-only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The uncommitted RC hardening of the multirealm Manager branch: a per-realm pending job
state machine reconciled against the database (no duplicate bots after a crash or an RA
timeout), cancelled by Stop spawning / Delete all bots, legacy file migrated; spawn
requests and build ids validated; tests JOB-01..JOB-14.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Subclasses/Hero Races and the /dev menu are parts of Custom Races (catalog part_of,
hidden from the grid): their switch and settings sit inside the Custom Races settings.
- The companion bots addon (CoABotUI) moves from the client facts into the addon list,
with install/update from the server package; the /dev menu addon of the V2 patch is
listed as installed by the patch, without its own buttons.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
addons/catalog.json pins SquidBots Lite, DragonUI (+Options), IncahootsCoA and CoACombat
to a commit and lists every file with its git blob SHA-1 and size. Install downloads
GitHub's archive of that commit, verifies each file (archives are not byte-stable, their
contents are), stages next to AddOns and moves the folders in; a foreign folder of the
same name goes to AddOns/.coa-addons-backup. .coa-addons.json records what the Manager
installed; update and remove touch only that. CoA_DevMenu ships with Custom Races V2, so
the separate dev menu addon is not in the catalog. Checked against the real GitHub
archives of the three small addons (ignored test).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- clear_game_cache removes Cache/WDB/<locale>/*.wdb before every Play (own server and
friend's), on by default, switch in the client card; nothing else in Cache or Data.
- The connection profile keeps the listed realm it was joined through; Play fills that
realm's saved account into GlueConfig.json every time. A host typed by hand drops it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Settings/restart-schedule.json (Settings, Startup) holds a local time and a countdown.
The Rust world supervisor asks the world for 'server restart <seconds>' once that time has
passed (by at most 15 minutes, for a world that ran before it), and starts the world
again when it exits with RESTART_EXIT_CODE; the database and auth keep running. While it
is under way (.state/scheduled-restart) the status says so and the Manager does not take
the down world for a crash. Extra worlds follow their server's schedule. Repacks with the
Python supervisor do not run the schedule; the card says the server needs an update.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Settings, Service edits Settings/mysql-custom.ini. Every start adds its name = value lines
(outside a section or in [mysqld]) to the rendered my.ini, after the defaults so they
win; the port, addresses, data and log paths, sockets and !include stay managed. Nothing
changes for a server without the file.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The worldserver ignores DisabledRaceIds saved against another ChrRaces catalog and with
it CoACustomRaces.Enable, so every server that had a restriction from 1.9.2 lost custom
races after the V2 update. Before a start the Manager binds the restriction to the new
catalog when every disabled race keeps its RaceID and name, judged by the server's own
snapshot (written by every save) or the table of every catalog a build has shipped
(race_catalogs.json, generated from the core history). Renamed races (1.9.1 reused 32/33)
and unknown catalogs still wait for the owner's review.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
V2 has 7,220 files. Switching rewrote the whole transaction journal with fsync after
every file (quadratic: minutes on fast disks, far longer with antivirus) and renamed
every file. Now:
- the journal is append-only JSON lines; recovery replays it, ignores a torn last
line and still reads the old journal.json of 0.9.13;
- only what the stock client loads is hidden (Data/*.MPQ, root dll/exe/bat, addons,
any file that replaced the player's own); Data/Retail, VulkanShaders and the
Esteria data stay in place, and a client disabled by 0.9.13 is recognised;
- payload hashes are remembered for the run, keyed by name, size and times.
Real V2 archives, release build: disable 1.9 s, enable 2.2 s, 224 of 7,220 files moved.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Settings, Startup lists every realm of the server with its name. The choice is kept in
Settings/realm-names.json of the main folder; the launcher, the realm-list setup and the
client's Config.wtf use it, and a running database gets it at once. Names are 1-32
characters without quotation marks or control characters, written as escaped SQL.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A reserved or taken database port no longer fails the start after 30 seconds: the
launcher recognises Windows excluded ranges (netsh), moves MySQL to a free port and
records it in Settings/repack.json. Only the server's own services use that port.
The database connection of with_database is rebuilt after the start.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The client update now takes ilusixn/CustomRacesCoa coa-custom-2.0: nine client archives plus the server
archive's client addons (subclass interface, dev menu, teleports, spell activation overlay), 76 race
archives and 7220 verified files in all, among them CoACustomRace.exe and its Vulkan renderer.
- The game starts CoACustomRace.exe when it is there; Ascension.exe is left alone, and a disabled V2
falls back to it.
- Updating from 1.9.2 retires its patched Ascension.exe and puts the player's own back (1.9.2 replaced
it; V2 does not ship one). Earlier installs keep their own hash lists, 1.9.2's included.
- Four V2 archives are byte-identical to 1.9.2's: a verified 1.9.2 copy in the cache or in Downloads
is used instead of downloading it again.
Verified: unit tests (800 passed); against the real archives, every bundled hash, a full install,
disable, enable and uninstall in a disposable client, and an update of a client that Manager 0.9.12
had put 1.9.2 on (its own Ascension.exe comes back, and again after uninstalling).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A console shutdown request is sometimes lost, and the world then keeps
running until the stop times out. While the world still runs and its
console still answers after 15 s, the launcher asks again (twice at most);
asking twice is harmless. Verified on a disposable repack with the first
World::ShutdownServ call suppressed in a debugger: the second request
stopped the world.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Repacks ship no .bat files, and the Manager starts servers with
coa-core::launcher, so the realm-profile patch it wrote into manage.py
served no one. It is no longer applied (realm creation, realm switching and
start).
Recognising the edits older Managers made (realm patch, bridge import fix)
moves into legacy_launcher, so updates and file checks still treat such a
file as unmodified. Nothing there writes a file.
manage.py itself stays complete in the repack: Managers up to 0.9.10 run it
through their bridge and patch its text, so it cannot become a thin wrapper
until those Managers are no longer supported.
Verified: coa-core unit tests; examples/manager_launcher_e2e on a disposable
repack (manage.py unchanged after a start with a realm profile).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
launcher::relay does what BugReport/relay.py and manage.py relay-worker do,
with the same spool: the <key>.status files the world reads, the journal
relay.sqlite3 (deliveries, settings), relay.lock (byte 0) and the
relay-ready record. Service answers map to the same states and retry times;
a post interrupted mid-flight stays uncertain and is never resent.
When coa-launcher.exe supervises a world it runs the relay itself; the
Python supervisor still runs relay.py.
Verified:
- examples/relay_equivalence: the same reports and scripted service answers
through relay.py and the Rust relay give identical statuses, journal rows
(including retry times) and posts, and each continues the other's journal.
- examples/manager_launcher_e2e on a disposable repack: supervisor and relay
are coa-launcher.exe, start/stop interoperates with manage.py.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
send_without_answer closed the socket right after writing "server shutdown".
With the reply still unread, Windows resets the connection, and the world
can drop the command: a world supervised by coa-launcher.exe ignored the
shutdown and the stop timed out after 90 seconds. The reply is now read up
to the next prompt, the end of the connection or 10 seconds.
The E2E also checks which supervisor runs the world and that manage.py
stops a world the Manager started.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
driver::run now calls coa-core::launcher::ops::run in-process instead of
running Scripts/manage.py through launcher_bridge.py. Failures carry a typed
code (Error::Service) instead of being guessed from stderr text.
Moved into the launcher: realm profiles (realm id, name and databases from
Settings/realm-profile.json), extra worlds sharing the main database,
the shared world startup deadline and console readiness check, the owned-world
check on shutdown, force-stop-all, and an environment block without
AC_REALM_SERVER_PORT for update validation.
The world supervisor is Scripts/coa-launcher.exe when the repack ships it,
otherwise manage.py watch-world. Extra worlds get an unmodified manage.py.
The realm patch of manage.py stays for the repack's own .bat files.
Verified with examples/manager_launcher_e2e on a disposable repack.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
std::process::Command lets a Windows child inherit every inheritable handle of the parent: a
MySQL started by the launcher kept the caller's files and pipes open (found by the SQUID test,
whose log stayed locked). Services now start through CreateProcessW with
PROC_THREAD_ATTRIBUTE_HANDLE_LIST = [log, NUL], as Python's subprocess does. The MySQL client's
outputs are read while it runs (a full pipe stopped it) and its last error line is reported.
E2E on a disposable repack: launcher_e2e PASS again; squid_equivalence PASS: Rust and
squid_playerbots.py leave identical journals, acore_playerbots tables (row counts) and
playerbots.conf (4106 lines), and a second run of either changes nothing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- services: exclusive-bind port check, readiness by listener row or the server side of a
loopback connection, MySQL client, start MySQL/auth, stop-all (world through its console,
database last), status.
- world: start-world through a supervisor process (coa-launcher watch-world) that keeps the
bug-report relay (still BugReport/relay.py via the repack's Python) alive; detailed logging.
- squid: squid_playerbots.py in Rust with the same journals (coa_squid_migrations,
coa_squid_pending, coa_bots_installed).
- coa-launcher.exe: the manage.py commands and error texts (the Manager reads them).
- Services no longer inherit the launcher's stdin/stdout/stderr: a server holding the
output pipe kept whoever read the launcher's output waiting.
E2E (examples/launcher_e2e, disposable stable repack on free ports): coa-launcher start-all ->
manage.py sees and stops it; manage.py start-all -> coa-launcher sees and stops it; a second
Rust-only cycle leaves nothing running.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
First part of moving Scripts/manage.py into coa-core (plan §17, PY-1). Same formats on disk as
manage.py so both can run on one server: .state records, the byte-0 control lock msvcrt.locking
takes, configuration.json. prepare renders byte-for-byte what manage.py renders (checked against
the real script, COA_MANAGE_PY). State files are flushed to disk before they replace the old ones,
and a damaged record counts as no process instead of failing every later start.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Starting a server that is open to friends adds the firewall rules for the
world ports of all realms that start together (only missing ones; a refused
Windows prompt does not undo the start).
- Settings shows one public-listing block per realm that starts together. The
main realm keeps its plain local id, so an existing listing is untouched.
- A realm that runs as an extra world is reached through that world's own
console port, so its listing reports its own state and capabilities.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Settings gets two options, one per extra realm; turning on both runs all
three realms. Each extra world keeps its own folder, configs, logs and
ports (multiworld now handles a set of worlds instead of one second world),
and Warcraft Reborn gets its own status card, realm list entry and firewall
rule.
Starting a server also sets CoA.ClassModel of the realm chosen on Overview
to exactly that realm's model, so the shipped "coa wcr" default no longer
shows two class models on a Conquest of Azeroth realm.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>