On SteamOS and Bazzite /home is a link to /var/home; Wine keeps the path it was
given, so the resolved one alone can miss the command line. The start order is
unchanged (database up and healthy before the realm list is updated): a comment
says so.
client::launch started Ascension.exe directly, which cannot work on Linux, and
client::is_running found nothing (it reads the Windows process list). wine.rs is
the Linux counterpart, and both hand over to it on a Linux host.
It runs the command of the project's own launcher script: umu-run with a Proton
build (GAMEID=0, PROTONPATH, WINEPREFIX, started from the client folder), else
plain Wine, with WINEDLLOVERRIDES=divxtac=d (DivxTac.dll deadlocks the world
loading screen under wine-mono). The newest GE-Proton found for Steam is used, or
umu's own choice; COA_PROTONPATH, COA_WINEPREFIX and client-launch.json override
the detection; the prefix defaults to ~/Games/umu/coa-client. The game runs in its
own process group, so closing the Manager does not close it, and prints to
client.log. Whether it runs is read from /proc, where the command line holds the
Windows form of the client path. A clear message says what to install when neither
umu nor Wine is found.
Tests: detection (newest build, no build, Wine only, nothing, overrides), the
command, the /proc match, and a real spawn with a stand-in umu-run that records its
environment. Not run: the real client.
The Realm card showed "io: No such file or directory (os error 2)" on a Docker
server: the Wildcard check read Core/worldserver.exe. The card lets the owner
switch realm, so it is made to work rather than hidden.
What the repack's patched launcher did at each start is done by the Docker
backend: finish an interrupted switch (realms::before_start), name the realms in
the realm list, and give the containers the databases of the selected realm
(acore_world_wildcard, acore_characters_wildcard). The switch itself is the shared
code. prepare_launcher has nothing to patch on a Docker server, and Wildcard
support is read from Core/worldserver as it is from Core/worldserver.exe (the .exe
wins when both exist, so a repack is read as before).
Tests: the support check on both binary names, no launcher to adapt, and the
databases passed to the containers for each realm. Creating the Wildcard realm
itself needs a build and a database with the Wildcard migrations and was not run.
Found by installing a server through the new screen: the update check of a Docker
server read the Windows "stable" channel and offered "0.261003.17 (you have
0.261003.1)". Applying it would have put the repack's executables, MySQL and
launcher into a Linux server folder.
Refused in the core (update::preview, apply and rollback), so no caller can do it.
The Settings tab says that updates are not available for Docker servers yet; the
background check stays quiet, as for any failed check.
A test applies the same update package to the same folder with and without
Settings/docker.json: refused and untouched with it, committed without it.
The install commands ask coa_core::platform::flavor() once and hand a Docker
install to src-tauri/src/docker_install.rs; the repack arm of each command is
untouched (an early return and an extra optional parameter are all that changed
in lib.rs). New command install_environment tells the screen the kind of server,
the suggested folder (~/CoaServer) and why Docker cannot be used, if it cannot.
On Linux the install screen shows the game data folder (used in place, read-only)
and the server package (a folder or an address; there is no default address until
a signed Linux package is published), and disables Install until both are given
and Docker works. It shows Docker's own message with a "Check again" button when
Docker is missing, stopped or not allowed for the user. The Windows screen is
unchanged (checked side by side in the browser mock).
A debug build (tauri dev) accepts another package key named in
COA_DEV_TRUSTED_KEY, to try the installer on a package signed with a throwaway
key; a release build always uses the embedded key.
New text is English only for now (other languages fall back to it).
- Add UnsavedChangesGuard modal (Save/Discard/Cancel) when navigating between tabs
- Wire guard into ModulesPage and SettingsPage with floating action bar
- Implement crash parser and explanation engine in coa-core
- Expose list_crashes Tauri command and CrashItem API
- Add auto-restart on unexpected server crash with crash loop protection
- Add crash counter badge and CrashHistoryModal with human-readable explanations and full crashlogs
- Support all 6 locales (en, ru, de, fr, es, zh)
Only the client's own connection errors (ERROR 2003 and 2002, "Can't connect") count as a stopped
database. The SQLSTATE 22003 of an out-of-range value contains "2003" and hid the real error of a
failing release migration. The message now also carries the client's last line.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Version bump only. Collects: the update check no longer fails after a crash and no longer
starts MySQL in the background, the Wildcard table repair checksum error, same-version files
edited by hand are not an update, a stopped server stays stopped after an update, the layout
fixes, one card per realm, remove-server and backup-folder settings, the Accounts tab
(bot accounts hidden), a world server started from the coa-bots folder, and the new
SQUID bot Dashboard tab.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Accounts named with AiPlayerbot.RandomBotAccountPrefix (default rndbot) belong to the bot
system like the companions' accounts: they are not listed and cannot be renamed or deleted.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The dashboard now runs through a small launcher that avoids the immediate socket shutdown that
cut answers above 64 KB on Windows, supplies the missing level-builds file, removes history
files zeroed by a crash, is addressed as 127.0.0.1 and opens in the browser through its own
command. The realm drop-down is hidden while both worlds start together.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The Dashboard tab (shown with SQUID Playerbots) downloads the dashboard release of the
bots' release series, starts it on a free local port and shows it in the window. After an
update, a server that was stopped stays stopped unless the owner asked for it to start.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The migration manager_repair__20261005_missing_wildcard_tables was published with two
different checksums; a server that applied one refused the update carrying the other.
Both versions have the same effect, so the Manager treats them as interchangeable.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
An update is decided by version: files the owner edited at the same version no longer
open a keep/replace decision. Repair shows which files differ or are missing.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Backups can be stored in a chosen folder (new backups go there, older ones stay listed),
and a backups folder that is a link to another drive no longer fails the update's safety
backup check. SQUID settings that declare a fixed list of choices are edited as a drop-down
and validated against it.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
- The main area and the console log are now positioning contexts, so screen-reader-only
text inside them can no longer extend the whole window (the second scrollbar in the update
dialog and the window scroll on the Console database tab).
- The client button group on the server card wraps instead of leaving the card.
- With both worlds starting together, each realm gets its own card.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
- Settings: a card removes a stopped server from the Manager (no files deleted).
- The Players tab is now Accounts in all languages.
- The five-minute background check no longer starts and stops MySQL on a stopped
server; only an explicit check may start it.
- A failed install keeps its marker until the folder is complete, cleans up after a
failure that happens after the move, and reports the launcher's output; install and
update errors have a Copy button.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A crash during a launcher write can leave .state/*.json zero-filled, which made
every launcher start (and so the update check) fail with "Something went wrong".
Zero-filled state files are now discarded before each launcher verb, the check
degrades to a warning when the database cannot be inspected, and database start
failures carry the launcher's own output.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The signed base package holds the MySQL data directory of a Windows repack. On a
Linux computer extract-schema-base now turns it into a Docker server folder whose
database container opens that directory (mysqlData in docker.json), so the release
SQL, the schema contract and export-baseline run with the Docker installation code.
Windows is unchanged: it still bootstraps the repack's own database program.
Version bump only. Collects: one-click diagnostics and report routing, LAN
address selector, Auction House Bot card without a GUID, experimental Linux
Docker installer, and the fixes already on master since 0.6.5.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Sharing on the local network used to take whatever address the Manager found
first, which is often a virtual adapter. The Friends page now lists the usable
IPv4 addresses of the computer (the automatic one marked) and accepts a custom
one. The choice is kept independently of the sharing mode, so it survives
restarts and switching to other modes. A chosen address that is not available
is reported instead of silently switching to another adapter.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
feat: install a new server as a Docker installation (Linux)
Also give the installer's test helper the compatible_sha256 field that master
added to manifest::Migration after the pull request was branched.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
- The diagnostic package now also holds crash reports and small crash dumps from
Core/Crashes, Windows' own crash records for the server programs (needed for
crashes that bypass the server's crash handler), the newest update journals,
the bot, supervisor and updater logs, and a view of which server programs run
and from where. The folder with the file opens after export.
- JSON files in the package keep valid JSON after secrets are removed.
- The problem report asks what the problem is about and opens the new-issue page
of the Manager, CoA Companions or SQUID Playerbots, preselecting the bot system
that is switched on. One button collects diagnostics and opens the report.
- Targets come from the module catalog; only their new-issue pages may be opened.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The module never deducts money from the auction character when it bids or buys, so the hint no longer says the character needs gold.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The GUID setting moves to the advanced list and the card text and the seller hint no longer ask for a character.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
docker::install is the Linux counterpart of install::install_base and shares none
of its code (the Windows installer and its folder checks are untouched). Same
order of events: signed manifest, folder / space / game data checks, verified
download, extraction into <dest>.installing, database, one rename. Docker must be
usable; a failure removes the containers, the volume and the staging folder.
The databases start from the package's own dumps, Database/baseline/<kind>.sql.zst,
in the format of the Manager's backups, taken in a state where every migration of
the package is applied. The installer imports them, records the migrations as
applied (baseline) in a few bulk statements, and creates the acore account, the
realm port and the console account. Rebuilding from the repository's SQL does not
work: pending migrations such as rev_20260903_01_live_class_baseline are guards
generated from the maintainers' live database and refuse to run on a database
built from the base scripts, whatever the order.
- coa-release export-baseline --server DIR --out DIR produces the dumps from a
prepared server (repack or Docker).
- The game data stays where the person has it (dataDir in docker.json, mounted
read-only), checked at install time.
- ensure_main_configs creates worldserver.conf / authserver.conf from their
.dist templates when missing, at install and at every start, never overwriting.
- Fix (affects the backend merged in 0.6.0): the database container's health check
is now a ping over TCP. On a new volume the image first runs a temporary server
that answers on its socket but takes no network connection, so the container was
healthy about four seconds (10 s against 14 s measured) before TCP worked.
- docs/ARCHITECTURE.md: D13 (installation, package contract) and D14 (how Windows
and Linux code are kept apart).
Tested: 12 new tests (folder rules, settings, package checks, bulk ledger SQL,
failure cleanup with an in-memory docker), and real runs of the whole chain
through examples/docker_install: baseline dumps exported from a prepared
server, package signed with a throwaway key, installed in 171 s, started by the
Manager (database healthy, world ready in 8 s, auth up), console reached with the
account the installer created, 1630 migrations recorded as applied.
Db runs mysql, mysqldump and mysqladmin inside the database container with
docker exec -i when the server folder is a Docker installation, connecting over
TCP on the container's loopback like the repack's commands do on the host. The
password still travels only in MYSQL_PWD. Every caller keeps using Db unchanged
(accounts, backups, population, migrations, the RA service account), and the
host needs no MySQL client. The repack path is untouched: one line before its
tools are looked up. A stopped container or a missing one is reported as a
database that is not running.
Tested: 4 new tests (commands built for the container, no password on a command
line, repack unchanged, error classification), and a real run against the
Docker folder: SQL queries, a quick backup (156 characters tables), restore with
the previous database kept, a damaged backup refused, the RA service account
created by SQL, the world server console reached through its published
loopback port, an account created through it and listed by the Manager.
A server folder with Settings/docker.json is a Docker installation: MySQL, the
auth server and the world server run in containers controlled through the
docker command line. Every other folder is a repack and goes through the
unchanged Windows code. driver::run, process::observe and layout::scan check
for the marker first and hand over to coa_core::docker.
- start: network, runtime image (libraries only, binaries are mounted from
Core/), database healthy, world listening, then auth. A container that stops
while starting reports the cause found in its log (DatabaseNotRunning,
PortInUse, ...). New error code DockerUnavailable (docker missing, stopped,
or not usable by this user).
- stop: world (180 s to save), auth, database; exit code 0 observed.
- observe: container state, health and port readiness mapped to the existing
ServiceState. Readiness is a connection that stays open, because Docker's
proxy accepts connections before the server inside listens.
- passwords go by environment, never on a command line; the remote console is
published on 127.0.0.1 only; the server's own database updater is turned off
(it needs a mysql client the image does not have and stops the server).
Tested: 18 new tests with an in-memory Docker, and a real run through
examples/drive: database, world (ready in 11 s) and auth started from a
package-shaped folder, stopped with exit code 0 and restarted.
Not included: installation, database setup, backups and SQL through the
container, updates, firewall/exposure, the client under Wine/Proton.
A server without a module's active file (an old repack updated by the Manager) made two modules log a
'missing property' line on every world tick.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A choice about files the owner changed appears in place below the buttons. First-time client setup keeps its dialog.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
- main: set WEBKIT_DISABLE_DMABUF_RENDERER=1 on Linux unless the user did.
WebKitGTK's DMA-BUF renderer ended the process at startup with a Wayland
"Protocol error" (seen with an NVIDIA card on Wayland).
- data_dir: use $XDG_DATA_HOME (absolute only) or ~/.local/share on Linux
instead of the temp folder. Windows keeps %LOCALAPPDATA%.
- open_link: xdg-open on Linux, explorer.exe on Windows as before.
- tailscale_exe: look in the usual Linux locations.
Every Windows branch is the previous code unchanged.
- config snapshots: two snapshots of the same scope in the same millisecond
shared one folder. A restore snapshots the current state just before reading
the snapshot it restores, so on a fast disk the snapshot being restored was
overwritten with the current files and the restore did nothing. Each snapshot
now gets its own folder (numeric suffix when the name is taken). This was the
intermittent failure of snapshot_restore_roundtrip_is_byte_exact_and_undoable.
- package extract: restore the execute bit on ELF binaries and #! scripts (the
archive carries no modes). No effect on Windows.
- fsx::starts_with_ci: case-insensitive on Windows only; /srv/CoA and /srv/coa
are different folders on Linux.
- registry test: use a root path that is a root on the running OS.
cargo test --workspace on Linux: 157 passed, 0 failed.
The collection switches and every other documented setting become tabs of the same
panel, are searched together with the curated settings and are saved by the same bar.
The Modules page drops the status legend; Playerbots is named SQUID's Playerbots.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A server whose module configs lack a key logs the same warning on every tick, which filled the 512 KB log tails of a
bug report and hid the actual failure. Read 8 MB back, fold the repeats, then keep the last part.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-10-01 20:29:30 +05:00
246 changed files with 21015 additions and 724 deletions
- PLAY launches only the client with the selected realmlist; starting the local server remains a separate action.
- Server multipliers accept decimals typed with a dot or comma. Talent and rested experience descriptions now explain their actual behavior.
- Added database and file checks, and Repair with backups and signed package verification. Unresolved schema damage is reported instead of guessed repairs.
- Fixed character database rebuilds retaining migration history that could skip required SQL. Applied migration hash changes are now rejected.
- Added optional simultaneous CoA and Wildcard worlds on Windows, with separate processes, ports and status indicators.
## 0.6.0 — 2026-10-03
- Added Simplified Chinese for the interface and all server and bot settings.
- The closed game client now follows the selected CoA or Wildcard realm. Play checks the saved realm before launching; existing settings are preserved and backed up.
- Fixed GitHub reports opening Windows Explorer instead of the default browser. Long reports use the clipboard fallback.
- Fixed configuration snapshots created in the same millisecond overwriting each other during restore.
- Added experimental Linux startup: persistent state uses the XDG data folder, Tailscale detection supports Linux, and a Wayland rendering workaround respects existing user settings.
- Added experimental start, stop and status controls for manually prepared Docker servers, with private database networking and a remote console bound to localhost.
- Linux extraction restores executable permissions for ELF binaries and scripts, and paths retain case-sensitive filesystem semantics.
### Availability
This release ships the Windows installer and signed updater assets. Linux support is a development preview, not a one-click installation flow. Docker installation/database setup, SQL operations and backups, updates, Wildcard realm profiles, and Wine/Proton client launching remain unfinished. Servers must be prepared manually and use Linux binaries built with the packaged configuration layout.
Server shutdown and starting-zone scaling issues require separate server releases and are not fixed by updating the Manager alone.
- [x]**Предупреждение о несохранённых изменениях при переходе между вкладками (Unsaved Changes Guard / Modal)**:
- Если у пользователя есть несохранённые изменения в настройках (`ModulesPage`, `SettingsPage` — сервер, боты и др.) и он переходит на другую вкладку (в боковом меню `Shell.tsx` или кнопками навигации), показывать pop-up диалог подтверждения.
- Опции в диалоге:
- **Сохранить (Save)** — сохранить внесённые изменения и перейти на выбранную страницу.
- **Сбросить (Discard)** — отменить несохранённые изменения и перейти на выбранную страницу.
- **Отмена (Cancel)** — остаться на текущей странице с сохранением введённых данных.
- *Контекст:* Предотвратит случайную потерю настроек, когда пользователь меняет параметры модуля/сервера и сразу уходит на вкладку Overview для запуска сервера (см. [Issue #45](https://github.com/Corfirean/coa-server-manager/issues/45)).
## Сервер и стабильность
- [x]**Автоподнятие сервера при падении (Auto-restart on crash) со счётчиком и расшифровкой крашей**:
- **Настройка:** Чекбокс «Автоматически перезапускать сервер при падении» в настройках сервера / менеджера.
- **Поведение автоподнятия:**
- Перезапускать процесс сервера только если падение произошло внезапно (процесс завершился с ненулевым exit code, крашем или без команды от менеджера).
- Если пользователь нажал кнопку «Стоп» (Stop Server), сервер останавливается полностью, автоподнятие не срабатывает.
- Защита от бесконечного цикла крашей (Crash Loop Protection): если за короткий интервал (например, 2-3 минуты) сервер падает более 3-5 раз, автоподнятие временно прекращается, а в UI выводится критическое оповещение.
- **Счётчик падений в интерфейсе:**
- Рядом со статусом сервера (или в шапке/панели состояния) показывать бейдж со счётчиком крашей с момента старта (например: `Крашей: 2`).
- При клике на счётчик открывать окно / модалку со списком последних падений с точным временем.
- **Расшифровка причин краша (Crash Explanation Engine):**
- *Как парсить и определять причину:*
1. Читать свежий текстовый файл из `Core/Crashes/<hash>_worldserver.exe_[date].txt`.
2. Смотреть `Exception code` (например, `C0000005` — нарушение доступа к памяти/access violation, `C0000420` — сработавший assert ядра, `C00000FD` — переполнение стека, `E06D7363` — исключение C++ / std::bad_alloc).
3. При `C0000420` парсить блок `ASSERTION FAILED` (`Location`, `Function`, `Condition`).
4. Анализировать последние строки `Core/Logs/Server.log` и `Errors.log` на критические маркеры (потеря соединения с MySQL, нехватка RAM/OOM, ошибка бинда портов).
5. Проверять по встроенному словарю известных сигнатур (паттерны частых крашей с понятным описанием, например: «Сбой удаления объекта из сетки карты», «Ошибка запроса к базе данных», «Нехватка оперативной памяти»).
- *Fallback:* Если сигнатура не опознана в словаре известных проблем, показывать краткую техническую выжимку (код исключения + сработавший ассерт или вершина Call Stack) и кнопку разворота полного текста крашлога / копирования.
title:The game client follows the realm selected in the Manager
---
Switching realms updates the closed client's saved realm automatically. Play checks it again before launch, preventing login attempts to the previous realm. If the client is running, the update waits until the next launch.
title:Experimental Linux startup and Docker server controls
---
The Manager can start on Linux and control manually prepared Docker servers. One-click installation, updates, backups and Wine or Proton client launch are not available yet.
title:Switch between CoA and Wildcard while keeping separate characters and module settings
---
Run one world at a time with shared accounts. Wildcard keeps its own progress; CoA companions are unavailable there, and other modules show experimental compatibility. Backups and server updates cover both realms.
title:Choose which item types the auction bot sells
---
Switch 15 item categories on or off and adjust listing frequency by quality. Labels and descriptions are translated into all six languages. Changes affect new listings; existing auctions remain.
title:Module settings now offer translated controls and an enemy damage slider
---
Configure scaling, auctions and enchantments with translated controls. Module switches no longer appear twice. Modules without settings hide their settings button; short lists need no search. New server builds support dungeon and raid damage from 25% to 200%.
title:Failed server updates restore databases and files together
---
Updates save a full backup and recover databases and files together. Interrupted SQL is not replayed. Checks cover Wildcard tables and every supported race/class pair, including existing characters. Unfinished updates block startup until recovered.
title:Check database structure and repair server files with a safety backup
---
Settings can check both realm databases and restore official files plus pending SQL. Installation and updates check character-save columns. Clean packages no longer keep old character migration records when rebuilding their database.
title:Configure partial dungeon parties and open-world life steal in module settings
---
Choose whether Dungeon Finder enters with the current party and enable open-world life steal with a healing percentage from 0 to 100. Available on server builds that support these options.
title:Start CoA and Wildcard together from Settings
---
Both worlds can run on separate ports with their own characters, settings and logs, sharing accounts. Stop saves both worlds before shutting down the shared database.
title:Configure SQUID Playerbots and prevent conflicting bot systems
---
The Bots tab shows translated settings for the active bot system. Bot module cards keep enable switches without duplicate settings. Enable one bot system at a time; the Manager checks manual config changes before starting the server. Recovery points also include the SQUID database when installed.
title:Download and update the game client without installing a server
---
Choose to host a server or connect to someone else's. Players joining a friend can download, update and launch the client, save the host's address and use the Manager without installing a local server.
title:Imported SquidBots servers warn before replacing their original binary
---
Imports explain that the original launcher and updater check the server binary hash. Updating that binary requires an explicit choice to keep or replace it.
title:Install a server from a package on Linux, running in Docker (experimental)
---
Experimental. The Manager creates the database, imports the packaged baseline and starts the server in Docker containers. It needs a Linux server package and your own game data folder.
title:Linux:install from the install screen, play with Wine or Proton (experimental)
---
Experimental. The install screen offers Docker installs on Linux, realm profiles and the Play button work on a Docker server, and updates are not offered for it.
title:Finish a server update that was blocked by a startup failure
---
The Manager repairs the launcher import path and lets you retry startup for an already applied update. Successful validation finishes the update without replaying SQL or restoring old databases. Failure details are preserved and rollback remains available.
title:SquidBots settings follow the options shipped by the module
---
New settings use the module's descriptions, groups, defaults and value limits while retaining existing translations. Unknown formats or invalid fields use built-in controls so the settings page remains available.
title:SquidBots shows its bundled release and includes it in problem reports
---
The Bots and Modules pages and problem reports include the release identifiers. Imported repacks use their recorded bot revision; empty tags are hidden. The settings file's source commit is not presented as the build revision.
title:Server updates accept verified published migration variants without rerunning them
---
Signed updates can recognise known published checksums without changing your migration history. Interrupted or failed SQL still requires recovery before retrying.
title:Server updates replace older launchers adapted by the Manager for Wildcard
---
Updates recognize the Manager’s own realm and import adjustments even when older metadata still records the original launcher. New server binaries receive their matching launcher; genuine custom edits still require your decision.
title:Server packages must include a complete database schema check
---
Release tools generate the expected database structure on an isolated copy of the signed base, verify all archive files and refuse packages without the schema check. Corrective SQL can restore missing Wildcard tables without replaying old updates.
title:The Auction House Bot card no longer asks for a character GUID
---
Turning the module and the seller on is enough: a server build with the automatic auction character creates its own account and character. The GUID field moved to the advanced settings.
The buyer option hint no longer says it spends the auction character's gold.
title:Choose and remember the local network address friends use to connect
---
Select an adapter or custom IPv4 address when sharing on LAN. Your choice survives restarts and mode changes; unavailable addresses show a warning instead of silently switching adapters.
title:Large SQUID database imports have more time to finish on first startup
---
The Manager allows up to 45 minutes when SQUID Playerbots is enabled, covering its initial database import and bot creation. Ordinary server startup and shutdown keep their existing deadlines.
title:Update history conflicts are detected before server files change
---
Damaged update records block changes and remain visible. Recovery checks saved copies before restoring, and unfinished updates protect backups from deletion and cleanup.
title:Pending database updates remain visible when the server version matches
---
SQL files are checked before replacing server files. Update and recovery attempts refresh their saved status, including errors. Backup deletion cannot race with an update.
title:Diagnostic files with database checks stay readable after secrets are removed
---
Removing a line that mentioned a secret could break the JSON files in a diagnostic package. Secret values are now replaced instead, so the files stay valid.
title:One button collects all diagnostics, including crashes and Windows crash records
---
Report a problem collects logs, crash reports and small dumps, Windows crash records, update journals and the running server programs in one file, then shows its folder. Crashes without a server crash report are no longer missing.
Choose what the problem is about: Manager and other modules, CoA Companions, or SQUID Playerbots. The report opens in that project's GitHub page with the versions filled in. The bot system that is switched on is preselected.
title:Choose where backups are stored, for example on another drive
---
Backups can now live in any folder outside the server folder. Existing backups stay listed. A backups folder that you moved and linked from the old place also works again for updates.
title:The bot dashboard loads completely and can be opened in the browser
---
The dashboard no longer fails on large pages, starts again after a crash left its history files empty, and its Open in the browser button works. With both realms started together the realm drop-down is hidden.
title:A failing database update is no longer reported as "database not running"
---
An SQL error whose message contained the number 2003, such as an out-of-range value, was shown as "The database is not running". The real error is now shown.
title:A Dashboard tab shows what the SQUID bots are doing
---
With SQUID Playerbots on, the Dashboard tab downloads the dashboard release that matches your bots, starts it on this computer and shows it inside the Manager.
title:A stopped server stays stopped after an update, unless you ask it to start
---
Updating no longer leaves a server running that was stopped before. Settings has a switch to start it afterwards, and the client card has one to open the game after a client update.
title:Checking for updates no longer fails with "Something went wrong" after a crash
---
A crash or power loss could leave a damaged status file that made every database start fail. The Manager now removes such files itself, still compares the files when the database is unavailable, and shows the real reason.
title:A world server started from the coa-bots folder is recognised
---
The Manager no longer reports a port conflict when the world server runs from the coa-bots folder inside the server folder, and a failed update now says that undoing it can take several minutes.
title:Failed server startup after an update automatically restores the previous version
---
The Manager restores server files and databases, then checks the recovered server. If recovery cannot be confirmed, it preserves the recovery point and blocks further startup until the problem is resolved.
title:Windows update recovery preserves database procedures triggers and views
---
A verified copy of stopped MySQL restores all databases and accounts together. Interrupted restoration can be retried, and the replaced database directory is retained.
The "Bots can be custom races" setting now configures AiPlayerbot.ExcludedBotRaces in playerbots.conf. Off excludes all added races (recommended), while on allows stable added races (experimental).
title:Blocked updates identify database objects that prevent automatic recovery
---
Diagnostics include the complete list of unsupported database objects across the recovery point, helping resolve update failures without removing user data.
title:Background world servers no longer stop because the local console is closed
---
The Manager automatically disables the local interactive console before starting the world server. Its remote console remains available, and no manual configuration changes are needed.
title:Updates can follow signed channels that select a complete server release
---
The Manager checks the channel signature and downloads every package part from the same release. Existing installations continue using the previous download source until the new channel is published.
title:Updates repair missing database defaults from the verified release schema
---
Only missing literal defaults are restored when column properties match. Existing rows, custom defaults, types and nullability are preserved; unresolved incompatibilities still block the update.
title:Server updates no longer replay module installation scripts on existing databases
---
Update packages contain incremental SQL only. Module table initialization and seed scripts remain available when preparing a clean installation, preventing existing content from being reset during updates.
title:Windows server updates are tested on a private copy before changing your installation
---
The Manager tests migrations and server startup using your databases and configurations. A failed rehearsal leaves the installed server unchanged and preserves a diagnostic report.
title:Server updates can require a tested combination of platform and installed version
---
Signed release metadata identifies core and module revisions, supported starting versions and database contracts. Incompatible packages are rejected before shutdown; changed schema contracts and incompatible databases block installation before migration.
title:Unsaved settings are protected and crashed servers can restart automatically
---
The Manager asks before leaving unsaved settings. Optional automatic restart pauses after repeated crashes, and crash history explains common failures.
title:Large server updates no longer repeatedly rewrite the recovery journal
---
Updates containing rebuilt navigation data now prepare all recovery copies before replacing files. Interrupted updates remain recoverable without rewriting the full journal for every file.
title:Normalize rehearsal server roots to prevent Windows path prefix errors
---
Server update rehearsals now canonicalize the installation root before collecting files, resolving a "prefix not found" error during rehearsal preparation on Windows.
title:Custom race client installation follows the current upstream release
---
The installer verifies the current race archives and includes the two additional MPQs. Legacy patch hashes remain available so upgrading never mistakes the old custom executable for an original launcher file.
title:Module cards show installed server revisions and the client race patch version
---
Module revisions come from the currently installed release metadata. The race card shows the client patch version separately; missing version metadata is reported honestly instead of showing a download version as installed.
title:Diagnostics record blocked startup requests and maintenance owner information
---
A start refused before the launcher runs is now logged with its reason. Exported diagnostics include the last recorded maintenance owner, while clearly distinguishing old metadata from a currently held operating-system lock.
title:Client race patches can be repaired and removed while preserving original files
---
Every overwritten client file gets an original backup. Switching preserves newer launcher files; checksums detect missing or replaced assets. Repair and removal are always available, with rollback if a file operation fails.
title:Checking an already installed server update no longer reports incompatibility
---
The installed release can be checked again even when its compatibility list only names earlier releases. Platform checks and protection against downgrades still apply.
title:Failed server startup shows service errors even when the launcher reported ready
---
Update rehearsal and startup failures include recent world, authentication and database error lines. Missing tables and late startup failures are visible directly in the error details, with secrets redacted.
title:Disabling custom races for SQUID bots now excludes all nonstandard races
---
The standard-races preset keeps only the ten original playable races. High elves, worgen and other custom race IDs are excluded, and the displayed setting reads complete race IDs rather than matching number substrings.
title:Starting the server waits for update checks and identifies ongoing maintenance
---
Downloading an update manifest no longer blocks startup. A start waits up to 30 seconds for a check to finish; other active maintenance still blocks it safely and shows its operation, process and start time.
Loaded 100 of 246 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.