testserver/Spigot-Server-Patches/0437-Improved-Watchdog-Supp...

597 lines
33 KiB
Diff
Raw Normal View History

From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
From: Aikar <aikar@aikar.co>
Date: Sun, 12 Apr 2020 15:50:48 -0400
Subject: [PATCH] Improved Watchdog Support
Forced Watchdog Crash support and Improve Async Shutdown
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
If the request to shut down the server is received while we are in
a watchdog hang, immediately treat it as a crash and begin the shutdown
process. Shutdown process is now improved to also shutdown cleanly when
not using restart scripts either.
If a server is deadlocked, a server owner can send SIGUP (or any other signal
the JVM understands to shut down as it currently does) and the watchdog
will no longer need to wait until the full timeout, allowing you to trigger
a close process and try to shut the server down gracefully, saving player and
world data.
Previously there was no way to trigger this outside of waiting for a full watchdog
timeout, which may be set to a really long time...
Additionally, fix everything to do with shutting the server down asynchronously.
Previously, nearly everything about the process was fragile and unsafe. Main might
not have actually been frozen, and might still be manipulating state.
Or, some reuest might ask main to do something in the shutdown but main is dead.
Or worse, other things might start closing down items such as the Console or Thread Pool
before we are fully shutdown.
This change tries to resolve all of these issues by moving everything into the stop
method and guaranteeing only one thread is stopping the server.
We then issue Thread Death to the main thread of another thread initiates the stop process.
We have to ensure Thread Death propagates correctly though to stop main completely.
This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
This also moves all plugins who register "delayed init" tasks to occur just before "Done" so they
are properly accounted for and wont trip watchdog on init.
diff --git a/src/main/java/com/destroystokyo/paper/Metrics.java b/src/main/java/com/destroystokyo/paper/Metrics.java
index 0b9e689d57705965721b5c55bc45d36657f360e4..dee00aac05f1acf050f05d4db557a08dd0f301c8 100644
--- a/src/main/java/com/destroystokyo/paper/Metrics.java
+++ b/src/main/java/com/destroystokyo/paper/Metrics.java
@@ -92,7 +92,12 @@ public class Metrics {
* Starts the Scheduler which submits our data every 30 minutes.
*/
private void startSubmitting() {
- final Runnable submitTask = this::submitData;
+ final Runnable submitTask = () -> {
+ if (MinecraftServer.getServer().hasStopped()) {
+ return;
+ }
+ submitData();
+ };
// Many servers tend to restart at a fixed time at xx:00 which causes an uneven distribution of requests on the
// bStats backend. To circumvent this problem, we introduce some randomness into the initial and second delay.
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
diff --git a/src/main/java/net/minecraft/server/CrashReport.java b/src/main/java/net/minecraft/server/CrashReport.java
index 95e6a6de7ccfc4445d0ac19c5f874c0d533b1712..cc6e6f245ee5e73bd570cf42381bf55ee0b364d3 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/net/minecraft/server/CrashReport.java
+++ b/src/main/java/net/minecraft/server/CrashReport.java
@@ -257,6 +257,7 @@ public class CrashReport {
}
public static CrashReport a(Throwable throwable, String s) {
+ if (throwable instanceof ThreadDeath) com.destroystokyo.paper.util.SneakyThrow.sneaky(throwable); // Paper
while (throwable instanceof CompletionException && throwable.getCause() != null) {
throwable = throwable.getCause();
}
diff --git a/src/main/java/net/minecraft/server/DedicatedServer.java b/src/main/java/net/minecraft/server/DedicatedServer.java
index a4db2527ff2defcfd25f08dd24722a37794a4a7e..f5792b999ce42acb13ae9a62ceb2ddec39abe209 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/net/minecraft/server/DedicatedServer.java
+++ b/src/main/java/net/minecraft/server/DedicatedServer.java
@@ -240,7 +240,7 @@ public class DedicatedServer extends MinecraftServer implements IMinecraftServer
long j = SystemUtils.getMonotonicNanos() - i;
String s = String.format(Locale.ROOT, "%.3fs", (double) j / 1.0E9D);
- DedicatedServer.LOGGER.info("Done ({})! For help, type \"help\"", s);
+ //DedicatedServer.LOGGER.info("Done ({})! For help, type \"help\"", s); // Paper moved to after init
if (dedicatedserverproperties.announcePlayerAchievements != null) {
((GameRules.GameRuleBoolean) this.getGameRules().get(GameRules.ANNOUNCE_ADVANCEMENTS)).a(dedicatedserverproperties.announcePlayerAchievements, (MinecraftServer) this);
}
@@ -368,6 +368,7 @@ public class DedicatedServer extends MinecraftServer implements IMinecraftServer
//this.remoteStatusListener.b(); // Paper - don't wait for remote connections
}
+ hasFullyShutdown = true; // Paper
System.exit(0); // CraftBukkit
}
@@ -701,7 +702,7 @@ public class DedicatedServer extends MinecraftServer implements IMinecraftServer
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
@Override
public void stop() {
super.stop();
- SystemUtils.h();
+ //SystemUtils.h(); // Paper - moved into super
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
}
@Override
diff --git a/src/main/java/net/minecraft/server/IAsyncTaskHandler.java b/src/main/java/net/minecraft/server/IAsyncTaskHandler.java
index bc15da4640a4a6107c9c186a01ce76df87511b41..27db247aa40e0516302c74b9bf00c631a8607af5 100644
--- a/src/main/java/net/minecraft/server/IAsyncTaskHandler.java
+++ b/src/main/java/net/minecraft/server/IAsyncTaskHandler.java
@@ -135,6 +135,7 @@ public abstract class IAsyncTaskHandler<R extends Runnable> implements Mailbox<R
try {
r0.run();
} catch (Exception exception) {
+ if (exception.getCause() instanceof ThreadDeath) throw exception; // Paper
IAsyncTaskHandler.LOGGER.fatal("Error executing task on {}", this.bj(), exception);
}
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
diff --git a/src/main/java/net/minecraft/server/MinecraftServer.java b/src/main/java/net/minecraft/server/MinecraftServer.java
index a8fe6c5f99c2be9dc7fedb8dfa8b12e4d0852534..2f5de20ea4e235d6f4664d7b3d0ac9ae31058b2f 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/net/minecraft/server/MinecraftServer.java
+++ b/src/main/java/net/minecraft/server/MinecraftServer.java
2020-06-26 16:20:03 +00:00
@@ -159,7 +159,7 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
public int autosavePeriod;
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
public boolean serverAutoSave = false; // Paper
public CommandDispatcher vanillaCommandDispatcher;
- private boolean forceTicks;
+ public boolean forceTicks; // Paper
// CraftBukkit end
// Spigot start
public static final int TPS = 20;
2020-06-26 16:20:03 +00:00
@@ -169,6 +169,8 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
public final SlackActivityAccountant slackActivityAccountant = new SlackActivityAccountant();
// Spigot end
+ public volatile Thread shutdownThread; // Paper
+
public static <S extends MinecraftServer> S a(Function<Thread, S> function) {
AtomicReference<S> atomicreference = new AtomicReference();
Thread thread = new Thread(() -> {
@@ -731,6 +733,7 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
// CraftBukkit start
private boolean hasStopped = false;
+ public volatile boolean hasFullyShutdown = false; // Paper
private final Object stopLock = new Object();
public final boolean hasStopped() {
synchronized (stopLock) {
@@ -745,6 +748,23 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
if (hasStopped) return;
hasStopped = true;
}
+ // Paper start - kill main thread, and kill it hard
+ shutdownThread = Thread.currentThread();
+ org.spigotmc.WatchdogThread.doStop(); // Paper
+ if (!isMainThread()) {
+ MinecraftServer.LOGGER.info("Stopping main thread (Ignore any thread death message you see! - DO NOT REPORT THREAD DEATH TO PAPER)");
+ while (this.getThread().isAlive()) {
+ this.getThread().stop();
+ try {
+ Thread.sleep(1);
+ } catch (InterruptedException e) {}
+ }
+ // We've just obliterated the main thread, this will prevent stop from dying when removing players
+ MinecraftServer.getServer().getWorlds().forEach(world -> {
+ world.tickingEntities = false;
+ });
+ }
+ // Paper end
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
// CraftBukkit end
MinecraftServer.LOGGER.info("Stopping server");
MinecraftTimings.stopServer(); // Paper
@@ -810,7 +830,18 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
this.getUserCache().b(false); // Paper
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
}
// Spigot end
+ // Paper start - move final shutdown items here
+ LOGGER.info("Flushing Chunk IO");
com.destroystokyo.paper.io.PaperFileIOThread.Holder.INSTANCE.close(true, true); // Paper
+ LOGGER.info("Closing Thread Pool");
+ SystemUtils.shutdownServerThreadPool(); // Paper
+ LOGGER.info("Closing Server");
+ try {
+ net.minecrell.terminalconsole.TerminalConsoleAppender.close(); // Paper - Use TerminalConsoleAppender
+ } catch (Exception e) {
+ }
+ this.exit();
+ // Paper end
}
public String getServerIp() {
@@ -903,6 +934,7 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
protected void w() {
try {
+ long serverStartTime = SystemUtils.getMonotonicNanos(); // Paper
if (this.init()) {
this.nextTick = SystemUtils.getMonotonicMillis();
this.serverPing.setMOTD(new ChatComponentText(this.motd));
@@ -910,6 +942,18 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
this.a(this.serverPing);
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
// Spigot start
+ // Paper start - move done tracking
+ LOGGER.info("Running delayed init tasks");
+ this.server.getScheduler().mainThreadHeartbeat(this.ticks); // run all 1 tick delay tasks during init,
+ // this is going to be the first thing the tick process does anyways, so move done and run it after
+ // everything is init before watchdog tick.
+ // anything at 3+ won't be caught here but also will trip watchdog....
+ // tasks are default scheduled at -1 + delay, and first tick will tick at 1
+ String doneTime = String.format(java.util.Locale.ROOT, "%.3fs", (double) (SystemUtils.getMonotonicNanos() - serverStartTime) / 1.0E9D);
+ LOGGER.info("Done ({})! For help, type \"help\"", doneTime);
+ // Paper end
+
+ org.spigotmc.WatchdogThread.tick(); // Paper
org.spigotmc.WatchdogThread.hasStarted = true; // Paper
Arrays.fill( recentTps, 20 );
long start = System.nanoTime(), curTime, tickSection = start; // Paper - Further improve server tick loop
@@ -965,6 +1009,12 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
this.a((CrashReport) null);
}
} catch (Throwable throwable) {
+ // Paper start
+ if (throwable instanceof ThreadDeath) {
+ MinecraftServer.LOGGER.error("Main thread terminated by WatchDog due to hard crash", throwable);
+ return;
+ }
+ // Paper end
MinecraftServer.LOGGER.error("Encountered an unexpected exception", throwable);
// Spigot Start
if ( throwable.getCause() != null )
@@ -996,14 +1046,14 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
} catch (Throwable throwable1) {
MinecraftServer.LOGGER.error("Exception stopping the server", throwable1);
} finally {
- org.spigotmc.WatchdogThread.doStop(); // Spigot
+ //org.spigotmc.WatchdogThread.doStop(); // Spigot // Paper - move into stop
// CraftBukkit start - Restore terminal to original settings
try {
- net.minecrell.terminalconsole.TerminalConsoleAppender.close(); // Paper - Use TerminalConsoleAppender
+ //net.minecrell.terminalconsole.TerminalConsoleAppender.close(); // Paper - Move into stop
} catch (Exception ignored) {
}
// CraftBukkit end
- this.exit();
+ //this.exit(); // Paper - moved into stop
}
}
@@ -1059,6 +1109,12 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
@Override
protected TickTask postToMainThread(Runnable runnable) {
+ // Paper start - anything that does try to post to main during watchdog crash, run on watchdog
+ if (this.hasStopped && Thread.currentThread().equals(shutdownThread)) {
+ runnable.run();
+ runnable = () -> {};
+ }
+ // Paper end
return new TickTask(this.ticks, runnable);
}
@@ -1301,6 +1357,7 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
try {
crashreport = CrashReport.a(throwable, "Exception ticking world");
} catch (Throwable t) {
+ if (throwable instanceof ThreadDeath) { throw (ThreadDeath)throwable; } // Paper
throw new RuntimeException("Error generating crash report", t);
}
// Spigot End
@@ -1758,7 +1815,8 @@ public abstract class MinecraftServer extends IAsyncTaskHandlerReentrant<TickTas
this.resourcePackRepository.a(collection);
this.saveData.a(a(this.resourcePackRepository));
datapackresources.i();
- this.getPlayerList().savePlayers();
+ if (Thread.currentThread() != this.serverThread) return; // Paper
+ //this.getPlayerList().savePlayers(); // Paper - we don't need to do this
this.getPlayerList().reload();
this.customFunctionData.a(this.dataPackResources.a());
this.ak.a(this.dataPackResources.h());
diff --git a/src/main/java/net/minecraft/server/PlayerChunkMap.java b/src/main/java/net/minecraft/server/PlayerChunkMap.java
index 87de764f74e8174da4c1a28238375caa6fdb4d5e..1b5c97c74b7224a8ac4d6e834925c0509a57d308 100644
--- a/src/main/java/net/minecraft/server/PlayerChunkMap.java
+++ b/src/main/java/net/minecraft/server/PlayerChunkMap.java
@@ -487,6 +487,7 @@ public class PlayerChunkMap extends IChunkLoader implements PlayerChunk.d {
MutableBoolean mutableboolean = new MutableBoolean();
do {
+ boolean isShuttingDown = world.getMinecraftServer().hasStopped(); // Paper
mutableboolean.setFalse();
list.stream().map((playerchunk) -> {
CompletableFuture completablefuture;
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
diff --git a/src/main/java/net/minecraft/server/PlayerList.java b/src/main/java/net/minecraft/server/PlayerList.java
index a1987b17adb5fa8fdeb601da948839fb1fd35aaf..fbd0ecb7bee67bf5affb6be6a57a8a0a8e2686b3 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/net/minecraft/server/PlayerList.java
+++ b/src/main/java/net/minecraft/server/PlayerList.java
@@ -432,7 +432,7 @@ public abstract class PlayerList {
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
cserver.getPluginManager().callEvent(playerQuitEvent);
entityplayer.getBukkitEntity().disconnect(playerQuitEvent.getQuitMessage());
- entityplayer.playerTick(); // SPIGOT-924
+ if (server.isMainThread()) entityplayer.playerTick(); // SPIGOT-924 // Paper - don't tick during emergency shutdowns (Watchdog)
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
// CraftBukkit end
// Paper start - Remove from collideRule team if needed
diff --git a/src/main/java/net/minecraft/server/SystemUtils.java b/src/main/java/net/minecraft/server/SystemUtils.java
index 1fe1df445ba56b2f176ee25502a774aa0a7bd00b..916a6a1fc1ccc73e4fb974ad77310d16fd8bd7e6 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/net/minecraft/server/SystemUtils.java
+++ b/src/main/java/net/minecraft/server/SystemUtils.java
@@ -123,6 +123,7 @@ public class SystemUtils {
return SystemUtils.f;
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
}
+ public static void shutdownServerThreadPool() { h(); } // Paper - OBFHELPER
public static void h() {
a(SystemUtils.e);
a(SystemUtils.f);
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
diff --git a/src/main/java/net/minecraft/server/World.java b/src/main/java/net/minecraft/server/World.java
index 4da88ce31cfb8baa6fa68c0115edadac0c5b2fba..d26fd68b2abd4907138ce77a6e6bb45c9af4af02 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/net/minecraft/server/World.java
+++ b/src/main/java/net/minecraft/server/World.java
@@ -798,6 +798,7 @@ public abstract class World implements GeneratorAccess, AutoCloseable {
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
gameprofilerfiller.exit();
} catch (Throwable throwable) {
+ if (throwable instanceof ThreadDeath) throw throwable; // Paper
// Paper start - Prevent tile entity and entity crashes
String msg = "TileEntity threw exception at " + tileentity.world.getWorld().getName() + ":" + tileentity.position.getX() + "," + tileentity.position.getY() + "," + tileentity.position.getZ();
System.err.println(msg);
@@ -872,6 +873,7 @@ public abstract class World implements GeneratorAccess, AutoCloseable {
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
try {
consumer.accept(entity);
} catch (Throwable throwable) {
+ if (throwable instanceof ThreadDeath) throw throwable; // Paper
// Paper start - Prevent tile entity and entity crashes
String msg = "Entity threw exception at " + entity.world.getWorld().getName() + ":" + entity.locX() + "," + entity.locY() + "," + entity.locZ();
System.err.println(msg);
diff --git a/src/main/java/net/minecraft/server/WorldServer.java b/src/main/java/net/minecraft/server/WorldServer.java
index 1c1dceb98413ccb24e3111435119b3b8c1b760c4..5d299a65417ff17eae3dfe52841c1a27a6d75677 100644
--- a/src/main/java/net/minecraft/server/WorldServer.java
+++ b/src/main/java/net/minecraft/server/WorldServer.java
@@ -65,7 +65,7 @@ public class WorldServer extends World implements GeneratorAccessSeed {
private final Queue<Entity> entitiesToAdd = Queues.newArrayDeque();
public final List<EntityPlayer> players = Lists.newArrayList(); // Paper - private -> public
public final ChunkProviderServer chunkProvider; // Paper - public
- boolean tickingEntities;
+ public boolean tickingEntities; // Paper - expose for watchdog
// Paper start
List<java.lang.Runnable> afterEntityTickingTasks = Lists.newArrayList();
public void doIfNotEntityTicking(java.lang.Runnable run) {
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
diff --git a/src/main/java/org/bukkit/craftbukkit/CraftServer.java b/src/main/java/org/bukkit/craftbukkit/CraftServer.java
index e5568d86b64465f54fe9bdec89b1a8f1d7639be7..4f5fe1f10195b1e3b09bf8301750dc75687b318a 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/org/bukkit/craftbukkit/CraftServer.java
+++ b/src/main/java/org/bukkit/craftbukkit/CraftServer.java
@@ -1792,7 +1792,7 @@ public final class CraftServer implements Server {
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
@Override
public boolean isPrimaryThread() {
- return Thread.currentThread().equals(console.serverThread); // Paper - Fix issues with detecting main thread properly
+ return Thread.currentThread().equals(console.serverThread) || Thread.currentThread().equals(net.minecraft.server.MinecraftServer.getServer().shutdownThread); // Paper - Fix issues with detecting main thread properly, the only time Watchdog will be used is during a crash shutdown which is a "try our best" scenario
}
@Override
diff --git a/src/main/java/org/bukkit/craftbukkit/Main.java b/src/main/java/org/bukkit/craftbukkit/Main.java
index 67331fa5463dd31e4aea3aebf6204ec5cb43d99e..8265879466333f4941c63c64b95d94f6ac0b5558 100644
--- a/src/main/java/org/bukkit/craftbukkit/Main.java
+++ b/src/main/java/org/bukkit/craftbukkit/Main.java
@@ -150,6 +150,37 @@ public class Main {
OptionSet options = null;
+ // Paper start - preload logger classes to avoid plugins mixing versions
+ tryPreloadClass("com.destroystokyo.paper.log.LogFullPolicy");
+ tryPreloadClass("org.apache.logging.log4j.core.Core");
+ tryPreloadClass("org.apache.logging.log4j.core.Appender");
+ tryPreloadClass("org.apache.logging.log4j.core.ContextDataInjector");
+ tryPreloadClass("org.apache.logging.log4j.core.Filter");
+ tryPreloadClass("org.apache.logging.log4j.core.ErrorHandler");
+ tryPreloadClass("org.apache.logging.log4j.core.LogEvent");
+ tryPreloadClass("org.apache.logging.log4j.core.Logger");
+ tryPreloadClass("org.apache.logging.log4j.core.LoggerContext");
+ tryPreloadClass("org.apache.logging.log4j.core.LogEventListener");
+ tryPreloadClass("org.apache.logging.log4j.core.AbstractLogEvent");
+ tryPreloadClass("org.apache.logging.log4j.message.AsynchronouslyFormattable");
+ tryPreloadClass("org.apache.logging.log4j.message.FormattedMessage");
+ tryPreloadClass("org.apache.logging.log4j.message.ParameterizedMessage");
+ tryPreloadClass("org.apache.logging.log4j.message.Message");
+ tryPreloadClass("org.apache.logging.log4j.message.MessageFactory");
+ tryPreloadClass("org.apache.logging.log4j.message.TimestampMessage");
+ tryPreloadClass("org.apache.logging.log4j.message.SimpleMessage");
+ tryPreloadClass("org.apache.logging.log4j.core.async.AsyncLogger");
+ tryPreloadClass("org.apache.logging.log4j.core.async.AsyncLoggerContext");
+ tryPreloadClass("org.apache.logging.log4j.core.async.AsyncQueueFullPolicy");
+ tryPreloadClass("org.apache.logging.log4j.core.async.AsyncLoggerDisruptor");
+ tryPreloadClass("org.apache.logging.log4j.core.async.RingBufferLogEvent");
+ tryPreloadClass("org.apache.logging.log4j.core.async.DisruptorUtil");
+ tryPreloadClass("org.apache.logging.log4j.core.async.RingBufferLogEventHandler");
+ tryPreloadClass("org.apache.logging.log4j.core.impl.ThrowableProxy");
+ tryPreloadClass("org.apache.logging.log4j.core.impl.ThrowableProxy$CacheEntry");
+ tryPreloadClass("org.apache.logging.log4j.core.impl.ExtendedClassInfo");
+ tryPreloadClass("org.apache.logging.log4j.core.impl.ExtendedStackTraceElement");
+ // Paper end
try {
options = parser.parse(args);
} catch (joptsimple.OptionException ex) {
@@ -245,8 +276,64 @@ public class Main {
} catch (Throwable t) {
t.printStackTrace();
}
+ // Paper start
+ // load some required classes to avoid errors during shutdown if jar is replaced
+ // also to guarantee our version loads over plugins
+ tryPreloadClass("com.destroystokyo.paper.util.SneakyThrow");
+ tryPreloadClass("com.google.common.collect.Iterators$PeekingImpl");
+ tryPreloadClass("com.google.common.collect.MapMakerInternalMap$Values");
+ tryPreloadClass("com.google.common.collect.MapMakerInternalMap$ValueIterator");
+ tryPreloadClass("com.google.common.collect.MapMakerInternalMap$WriteThroughEntry");
+ tryPreloadClass("com.google.common.collect.Iterables");
+ for (int i = 1; i <= 15; i++) {
+ tryPreloadClass("com.google.common.collect.Iterables$" + i, false);
+ }
+ tryPreloadClass("org.apache.commons.lang3.mutable.MutableBoolean");
+ tryPreloadClass("org.apache.commons.lang3.mutable.MutableInt");
+ tryPreloadClass("org.jline.terminal.impl.MouseSupport");
+ tryPreloadClass("org.jline.terminal.impl.MouseSupport$1");
+ tryPreloadClass("org.jline.terminal.Terminal$MouseTracking");
+ tryPreloadClass("co.aikar.timings.TimingHistory");
+ tryPreloadClass("co.aikar.timings.TimingHistory$MinuteReport");
+ tryPreloadClass("io.netty.channel.AbstractChannelHandlerContext");
+ tryPreloadClass("io.netty.channel.AbstractChannelHandlerContext$11");
+ tryPreloadClass("io.netty.channel.AbstractChannelHandlerContext$12");
+ tryPreloadClass("io.netty.channel.AbstractChannel$AbstractUnsafe$8");
+ tryPreloadClass("io.netty.util.concurrent.DefaultPromise");
+ tryPreloadClass("io.netty.util.concurrent.DefaultPromise$1");
+ tryPreloadClass("io.netty.util.internal.PromiseNotificationUtil");
+ tryPreloadClass("io.netty.util.internal.SystemPropertyUtil");
+ tryPreloadClass("org.bukkit.craftbukkit.scheduler.CraftScheduler");
+ tryPreloadClass("org.bukkit.craftbukkit.scheduler.CraftScheduler$1");
+ tryPreloadClass("org.bukkit.craftbukkit.scheduler.CraftScheduler$2");
+ tryPreloadClass("org.bukkit.craftbukkit.scheduler.CraftScheduler$3");
+ tryPreloadClass("org.bukkit.craftbukkit.scheduler.CraftScheduler$4");
+ tryPreloadClass("org.slf4j.helpers.MessageFormatter");
+ tryPreloadClass("org.slf4j.helpers.FormattingTuple");
+ tryPreloadClass("org.slf4j.helpers.BasicMarker");
+ tryPreloadClass("org.slf4j.helpers.Util");
+ tryPreloadClass("com.destroystokyo.paper.event.player.PlayerConnectionCloseEvent");
+ tryPreloadClass("com.destroystokyo.paper.event.entity.EntityRemoveFromWorldEvent");
+ // Minecraft, seen during saving
+ tryPreloadClass("net.minecraft.server.LightEngineLayerEventListener$Void");
+ tryPreloadClass("net.minecraft.server.LightEngineLayerEventListener");
+ tryPreloadClass("net.minecraft.server.ExceptionSuppressor");
+ // Paper end
+ }
+ }
+
+ // Paper start
+ private static void tryPreloadClass(String className) {
+ tryPreloadClass(className, true);
+ }
+ private static void tryPreloadClass(String className, boolean printError) {
+ try {
+ Class.forName(className);
+ } catch (ClassNotFoundException e) {
+ if (printError) System.err.println("An expected class " + className + " was not found for preloading: " + e.getMessage());
}
}
+ // Paper end
private static List<String> asList(String... params) {
return Arrays.asList(params);
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
diff --git a/src/main/java/org/bukkit/craftbukkit/util/ServerShutdownThread.java b/src/main/java/org/bukkit/craftbukkit/util/ServerShutdownThread.java
index 449e99d1b673870ed6892f6ab2c715a2db35c35d..c7ed6e0f8a989cec97700df2b15198c9c481c549 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/org/bukkit/craftbukkit/util/ServerShutdownThread.java
+++ b/src/main/java/org/bukkit/craftbukkit/util/ServerShutdownThread.java
@@ -12,12 +12,27 @@ public class ServerShutdownThread extends Thread {
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
@Override
public void run() {
try {
+ // Paper start - try to shutdown on main
+ server.safeShutdown(false, false);
+ for (int i = 1000; i > 0 && !server.hasStopped(); i -= 100) {
+ Thread.sleep(100);
+ }
+ if (server.hasStopped()) {
+ while (!server.hasFullyShutdown) Thread.sleep(1000);
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
+ return;
+ }
+ // Looks stalled, close async
org.spigotmc.AsyncCatcher.enabled = false; // Spigot
org.spigotmc.AsyncCatcher.shuttingDown = true; // Paper
+ server.forceTicks = true;
server.close();
+ while (!server.hasFullyShutdown) Thread.sleep(1000);
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
+ } catch (InterruptedException e) {
+ e.printStackTrace();
+ // Paper end
} finally {
try {
- net.minecrell.terminalconsole.TerminalConsoleAppender.close(); // Paper - Use TerminalConsoleAppender
+ //net.minecrell.terminalconsole.TerminalConsoleAppender.close(); // Paper - Move into stop
} catch (Exception e) {
}
}
diff --git a/src/main/java/org/spigotmc/RestartCommand.java b/src/main/java/org/spigotmc/RestartCommand.java
index aefea3a9a8b9b75c62bd20018be7cd166a213001..123de5ac9026508e21cdc225f0962f5c3c46fed5 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/org/spigotmc/RestartCommand.java
+++ b/src/main/java/org/spigotmc/RestartCommand.java
@@ -139,7 +139,7 @@ public class RestartCommand extends Command
// Paper end
// Paper start - copied from above and modified to return if the hook registered
- private static boolean addShutdownHook(String restartScript)
+ public static boolean addShutdownHook(String restartScript)
{
String[] split = restartScript.split( " " );
if ( split.length > 0 && new File( split[0] ).isFile() )
diff --git a/src/main/java/org/spigotmc/WatchdogThread.java b/src/main/java/org/spigotmc/WatchdogThread.java
index 9c1fc19851817a842779ab89bc9300e1adf3d2a2..ae8903ee1decd22e2ad6138f29fbc757b807e0a7 100644
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
--- a/src/main/java/org/spigotmc/WatchdogThread.java
+++ b/src/main/java/org/spigotmc/WatchdogThread.java
@@ -13,6 +13,7 @@ import org.bukkit.Bukkit;
public class WatchdogThread extends Thread
{
+ public static final boolean DISABLE_WATCHDOG = Boolean.getBoolean("disable.watchdog"); // Paper
private static WatchdogThread instance;
private long timeoutTime;
private boolean restart;
@@ -41,6 +42,7 @@ public class WatchdogThread extends Thread
{
if ( instance == null )
{
+ if (timeoutTime <= 0) timeoutTime = 300; // Paper
instance = new WatchdogThread( timeoutTime * 1000L, restart );
instance.start();
} else
@@ -71,12 +73,13 @@ public class WatchdogThread extends Thread
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
// Paper start
Logger log = Bukkit.getServer().getLogger();
long currentTime = monotonicMillis();
- if ( lastTick != 0 && timeoutTime > 0 && currentTime > lastTick + earlyWarningEvery && !Boolean.getBoolean("disable.watchdog") )
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
+ MinecraftServer server = MinecraftServer.getServer();
+ if (lastTick != 0 && timeoutTime > 0 && hasStarted && (!server.isRunning() || (currentTime > lastTick + earlyWarningEvery && !DISABLE_WATCHDOG) ))
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
{
- boolean isLongTimeout = currentTime > lastTick + timeoutTime;
+ boolean isLongTimeout = currentTime > lastTick + timeoutTime || (!server.isRunning() && !server.hasStopped() && currentTime > lastTick + 1000);
// Don't spam early warning dumps
if ( !isLongTimeout && (earlyWarningEvery <= 0 || !hasStarted || currentTime < lastEarlyWarning + earlyWarningEvery || currentTime < lastTick + earlyWarningDelay)) continue;
- if ( !isLongTimeout && MinecraftServer.getServer().hasStopped()) continue; // Don't spam early watchdog warnings during shutdown, we'll come back to this...
+ if ( !isLongTimeout && server.hasStopped()) continue; // Don't spam early watchdog warnings during shutdown, we'll come back to this...
lastEarlyWarning = currentTime;
if (isLongTimeout) {
// Paper end
@@ -118,7 +121,7 @@ public class WatchdogThread extends Thread
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
log.log( Level.SEVERE, "------------------------------" );
log.log( Level.SEVERE, "Server thread dump (Look for plugins here before reporting to Paper!):" ); // Paper
ChunkTaskManager.dumpAllChunkLoadInfo(); // Paper
- dumpThread( ManagementFactory.getThreadMXBean().getThreadInfo( MinecraftServer.getServer().serverThread.getId(), Integer.MAX_VALUE ), log );
+ dumpThread( ManagementFactory.getThreadMXBean().getThreadInfo( server.serverThread.getId(), Integer.MAX_VALUE ), log );
log.log( Level.SEVERE, "------------------------------" );
//
// Paper start - Only print full dump on long timeouts
@@ -139,9 +142,24 @@ public class WatchdogThread extends Thread
Forced Watchdog Crash support and Improve Async Shutdown If the request to shut down the server is received while we are in a watchdog hang, immediately treat it as a crash and begin the shutdown process. Shutdown process is now improved to also shutdown cleanly when not using restart scripts either. If a server is deadlocked, a server owner can send SIGHUP (or any other signal the JVM understands to shut down as it currently does) and the watchdog will no longer need to wait until the full timeout, allowing you to trigger a close process and try to shut the server down gracefully, saving player and world data. Previously there was no way to trigger this outside of waiting for a full watchdog timeout, which may be set to a really long time... Additionally, fix everything to do with shutting the server down asynchronously. Previously, nearly everything about the process was fragile and unsafe. Main might not have actually been frozen, and might still be manipulating state. Or, some reuest might ask main to do something in the shutdown but main is dead. Or worse, other things might start closing down items such as the Console or Thread Pool before we are fully shutdown. This change tries to resolve all of these issues by moving everything into the stop method and guaranteeing only one thread is stopping the server. We then issue Thread Death to the main thread of another thread initiates the stop process. We have to ensure Thread Death propagates correctly though to stop main completely. This is to ensure that if main isn't truely stuck, it's not manipulating state we are trying to save.
2020-04-16 08:53:50 +00:00
if ( isLongTimeout )
{
- if ( restart && !MinecraftServer.getServer().hasStopped() )
+ if ( !server.hasStopped() )
{
- RestartCommand.restart();
+ AsyncCatcher.enabled = false; // Disable async catcher incase it interferes with us
+ AsyncCatcher.shuttingDown = true;
+ server.forceTicks = true;
+ if (restart) {
+ RestartCommand.addShutdownHook( SpigotConfig.restartScript );
+ }
+ // try one last chance to safe shutdown on main incase it 'comes back'
+ server.safeShutdown(false, restart);
+ try {
+ Thread.sleep(1000);
+ } catch (InterruptedException e) {
+ e.printStackTrace();
+ }
+ if (!server.hasStopped()) {
+ server.close();
+ }
}
break;
} // Paper end
diff --git a/src/main/resources/log4j2.xml b/src/main/resources/log4j2.xml
index 476f4a5cbe664ddd05474cb88553018bd334a5b8..8af159abd3d0cc94cf155fec5b384c42f69551bf 100644
--- a/src/main/resources/log4j2.xml
+++ b/src/main/resources/log4j2.xml
@@ -1,5 +1,5 @@
<?xml version="1.0" encoding="UTF-8"?>
-<Configuration status="WARN" packages="com.mojang.util">
+<Configuration status="WARN" packages="com.mojang.util" shutdownHook="disable">
<Appenders>
<Queue name="ServerGuiConsole">
<PatternLayout pattern="[%d{HH:mm:ss} %level]: %msg%n" />