Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Playing with friends? Godlike.Host runs hosted Minecraft Java servers from 10 data-center locations, with free DDoS protection.
A Minecraft Java server that runs for days and then disappears usually has a diagnosable failure, not a mysterious time limit. Stop it, make a complete copy of the server directory, then classify the evidence in latest.log, crash-reports, your hosting panel, and (on Linux) the kernel log. The signature normally points to one of five causes: a plugin or dependency fault, a watchdog shutdown caused by a stalled tick, an operating-system or container out-of-memory kill, a normal shutdown, or an incompatible Java/server/world version.
1. Preserve the world before changing anything
Do not begin by deleting plugins, changing JVM flags, or forcing a world to load. PaperMC’s troubleshooting guidance starts with: “Before following this guide, stop your server first.” A live copy can be writing region files while you investigate, so an incomplete backup can preserve the problem and damage the only good world copy.
- Stop the process from the hosting panel or send
stopin the server console. Wait until the console confirms that the world has been saved and the Java process has exited. - Archive the entire server directory, not just the world folder. Include
server.properties,eula.txt, startup scripts, thepluginsdirectory, configuration files, logs, crash reports, and all world folders. - Record the server jar name, Java version, startup command, allocated memory, hosting-panel memory limit, installed plugins, and the approximate time of the last successful run.
- Work on a copy when testing a suspected plugin or changing world settings. Keep the original archive read-only until you know which change solved the failure.
A backup is also essential for version problems. Opening a world with a newer server and then trying to force it into an older server can permanently corrupt data; a backup is the only reliable way back.
2. Collect the evidence that survives a delayed failure
“It stopped” is not a diagnosis. Look at the last several minutes of output before the process disappeared, then compare that time with panel and operating-system records.
#1 Best Overall
- Play and share with friends on console, mobile and Windows 10
- discover community creations in the new in-game store
- access new mini games and game modes through servers
| Evidence | What to inspect | What it can establish |
|---|---|---|
latest.log |
Exceptions, plugin names, shutdown messages, repeated warnings, and the final timestamp | Most Java and plugin failures, including a graceful /stop |
| Older compressed logs | Files such as 2026-10-15-1.log.gz (the date and suffix vary) |
Whether the same warning or stall appeared on earlier days |
crash-reports |
The exception type, message, and first stack frames | Which thread and component caused a Minecraft crash |
| Hosting-panel log and event history | Exit code, memory-limit events, automatic restarts, and scheduled tasks | Whether the panel or container terminated Java |
| Linux kernel log | dmesg -T | grep -i killed |
An operating-system out-of-memory kill that left no Minecraft crash report |
Start with the exception class and message, not the bottom of a long stack trace. The first application-specific frames usually identify the plugin, library, world operation, or server subsystem that was active when the failure occurred. A stack trace is a path through code, not proof that the lowest-level method is at fault.
3. Match the symptom to the failure class
Plugin or dependency fault
If a plugin package appears in the exception or in repeated errors immediately before the crash, treat it as the leading suspect. Update that plugin and any library it depends on, then test with plugins disabled. A plugin can fail only after a particular event, scheduled task, world operation, or accumulated state, which explains why a server may run for weeks first.
Shared libraries deserve separate attention. Protocol libraries and permission providers can break several otherwise unrelated plugins when their versions no longer match the server. Check the plugin’s stated Minecraft and server-platform support, its required dependencies, and the first release that supports your Java runtime.
To isolate one offender without guessing, use a binary split:
Rank #2
- Learn your way around one block at atime in the new tutorial mode
- Craft all sorts of items from boats todiamond armor and create enormousstructures
- Team up with friends in eight-playeronline multiplayer and four-player splitscreen modes
- Stop the server and copy the complete directory.
- Move roughly half of the plugins out of the active
pluginsdirectory. Do not delete them; preserve their configuration and data. - Start the copy and reproduce the workload or wait through the event that normally precedes the failure.
- If the failure remains, the suspect is in the half still enabled. If it disappears, it is in the half you removed.
- Split the suspect half again and repeat until one plugin or dependency remains.
After identifying it, install a compatible update or remove it from the production copy. If no compatible version exists, restore from backup and choose a supported replacement rather than leaving the server exposed to repeated crashes.
Watchdog shutdown or a tick stall
A watchdog report means the main server thread stopped responding. PaperMC describes the watchdog as reporting where the thread was stuck when the server took more than 10 seconds to respond. The current server.properties reference sets max-tick-time=60000 by default: one tick lasting 60,000 milliseconds is treated as a crash and the process is forcibly shut down. A normal Minecraft tick should be about 0.05 seconds, so a 60-second tick is an extreme stall, not ordinary lag.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchRead the watchdog dump’s first stack frames. They often show a slow disk operation, a world save or generation task, entity processing, a plugin callback, or a synchronized operation waiting on another thread. The stack tells you what the main thread was doing; it does not by itself prove which plugin initiated that work. Collect a profile with spark while the problem is occurring, then use the profile’s hottest methods and plugin attribution to choose a fix.
Do not set max-tick-time=-1 as a cure. That disables the safety shutdown and allows a wedged process to consume resources or remain unreachable. It can hide the symptom while making recovery harder. Fix the operation that blocks the main thread, reduce the workload, or update the responsible plugin instead.
Rank #3
- ENGAGING AND COOPERATIVE: Experience the thrill of Minecraft in a unique and cooperative board game format. Perfect for teams of 2 to 4 players, the game encourages joint strategies and shared victories
- EVER-CHANGING GAMEPLAY: With a map that changes every time you play, this game offers high replay value, making every game a unique and unpredictable experience
- QUALITY COMPONENTS: The game includes 25 wooden blocks, 18 world tiles, 4 inventory boards, 6 player cards, 9 buildings, 12 mob tokens and more. All components are crafted with care to enhance your gaming experience
- AGE-APPROPRIATE CHALLENGE: Aimed at ages 7 and up, this game offers a great balance of fun and challenge, making it a great choice for family game nights, parties, or gatherings
- TRUSTED BRAND: With over 130 years of experience, Ravensburger ensures top quality games that stimulate the mind, engage the hands, and touch the heart
Operating-system or container out-of-memory kill
A host can kill Java without giving Minecraft time to write a crash report. This is common when a container or hosting panel has a hard memory limit close to the Java heap maximum. On Linux, run dmesg -T | grep -i killed after the event and inspect the panel’s memory graph and event log. A kernel message naming the Java process, or a panel event saying the container exceeded its limit, distinguishes this case from a Java exception.
Leave headroom between the panel limit and -Xmx. The operating system, Java’s native memory, thread stacks, class metadata, direct buffers, and file cache all need space outside the Java heap. If the panel limit cannot be raised, reduce -Xmx; if the world and player load require more memory, raise the panel limit instead. Change one value at a time and record the result.
Normal or graceful stop
If the final console lines look as though an administrator typed /stop, a scheduled task ran, or a panel issued a restart command, investigate automation before changing Minecraft settings. Review cron jobs, systemd timers, panel schedules, crash-restart scripts, and access logs. To obtain more detail on an unexplained orderly shutdown, enable debug in server.properties, reproduce the stop, and inspect the resulting stack trace. This can reveal which code path requested the shutdown.
Java, server, or world version mismatch
UnsupportedClassVersionError means the Java runtime is older than the bytecode used by the server or a plugin. Install the Java major version required by your chosen server release and verify the executable used by the startup script, because a panel may launch a different Java binary than your interactive shell.
Rank #4
- Minecraft Dungeons Hero Edition includes the Hero Pass DLC content: a Hero Cape, two player skins, and a chicken pet. It also includes the Jungle Awakens and Creeping Winters DLC packs.
- Dungeon Creeper Battle new-and-nasty mobs in this all-new action-adventure, inspired by classic dungeon crawlers.
- Multiplayer Up to four players can team up and fight together in co-op mode.
- Power Up Unlock dozens of unique items and weapon enchantments for devastating special attacks.
- Options Personalize your character, then fight up-close and personal with melee swings, hang back with ranged attacks, or tank your way through swarms of mobs, by heavy armor
A message that a chunk or world was saved with a newer version means the data was opened by a newer server release than the one now trying to load it. Downgrading is unsupported. Do not force the load or remove version metadata: the result can permanently corrupt the world. Restore the backup, run the newer supported server version, and update plugins for that version.
Free tools Windows power users keep installed
One-click scans. No signup required.
4. Interpret memory correctly before tuning the JVM
Java often keeps memory it has allocated instead of immediately returning it to the operating system. A high resident-memory number alone therefore does not prove a leak. PaperMC notes that this behavior “is not a memory leak and will not cause out of memory crashes,” which are commonly caused by insufficient room for the operating system or by incorrectly configured container limits.
Look for evidence of an actual memory failure: a Java OutOfMemoryError, a kernel OOM message, a panel limit event, or a steadily rising usage pattern that ends at a hard cap. Capture heap and process metrics over time rather than reacting to one reading after startup. Keep the Java heap maximum below the hosting limit, and avoid assigning every available megabyte to -Xmx.
5. Check workload settings that can turn a slow server into a watchdog crash
Two server properties control how much world work is scheduled around players:
Best Value
- Official Minecraft Puzzle Game – Hands-On Logic Game Based on the Best-Selling Video Game.
- 40 Strategic Challenges – Use the Actions to Navigate & Solve Increasingly Tough Puzzles.
- No Screens, Just Strategy – Promotes Problem-Solving with Off-Screen Minecraft Fun.
- Perfect for Ages 8+ – Great Gift for Minecraft Fans Who Love a Challenge!
- Trusted by Parents Worldwide – With Over 50 Million Sold, ThinkFun is the World's Leading Manufacturer of Brain Games and Mind-Challenging Puzzles!
view-distance=10is the current PaperMC reference default and controls how many chunks the server sends around each player.simulation-distance=10is also the reference default and controls how many chunks actively run entities, redstone, and other simulation work.
Lowering these values can reduce CPU, memory, and disk pressure when player count or exploration spikes. Change them gradually, test with the same number of players and worlds, and note the gameplay trade-off: players see less distant terrain and fewer distant systems remain active.
pause-when-empty-seconds=-1 is the current default, meaning pausing is disabled. PaperMC keeps it disabled by default because plugins may be incompatible with a paused server. Enabling pausing can reduce work on an empty server, but test every plugin that schedules tasks, maintains connections, or expects ticks to continue.
6. Profile a stall instead of guessing
- Install a version of
sparkcompatible with your Paper-based server and Java runtime, following its documented installation procedure. - When ticks slow but the server still accepts commands, start a timed profile from the server console using the command syntax provided by your installed spark version.
- Capture the profile during the actual lag window, not immediately after a restart.
- Inspect tick duration, CPU hot spots, scheduler tasks, entity counts, and plugin attribution. Save the profile link or file with the matching log timestamp.
- Use the result to change one cause—such as a plugin task, excessive entities, a world-generation operation, or an over-large simulation distance—then repeat the same test.
If the main thread is blocked on disk or a network operation, profile results and the watchdog stack should agree on the waiting method. In that case, check storage latency, remote mounts, backups running against the live directory, and plugins that perform synchronous I/O.
7. A controlled recovery runbook
Use this order so that a tempting configuration change does not erase the evidence needed to identify the cause:
- Preserve: stop the server and archive the complete directory.
- Classify: decide whether the event is a Java crash, watchdog shutdown, host OOM kill, graceful stop, or version error.
- Reproduce safely: use a copy, the same world, comparable player load, and the same startup command.
- Remove variables: temporarily disable half the plugins, then use binary splitting if the failure is plugin-related.
- Validate versions: confirm Java, the server jar, every plugin, dependencies, and world data are mutually supported.
- Measure: use watchdog stacks, spark profiles, panel metrics, and kernel logs rather than memory impressions or console guesses.
- Change one thing: adjust one plugin, workload setting, heap limit, or host resource allocation per test.
- Observe a full cycle: keep the server under its normal scheduled jobs, backups, player count, and exploration pattern long enough to cover the delayed event.
- Document: record the change, date, server build, Java version, evidence, and result so a future update can be reversed.
8. Prevent a repeat after the server is stable
- Keep an automatic backup schedule, but run backups from a safe snapshot or stopped server when possible; do not overwrite the only known-good archive.
- Retain rotated logs and crash reports long enough to compare failures separated by days or weeks.
- Monitor process memory, container memory, CPU, disk space, disk latency, tick time, and restart events. Alert before a hard memory limit is reached.
- Apply server and plugin updates in a staging copy first. Read each plugin’s dependency and Java-version requirements before deploying.
- Use a process supervisor or hosting-panel restart policy for availability, but do not treat automatic restart as a fix. Preserve the pre-restart logs so the original failure remains visible.
- Keep
view-distance,simulation-distance, entity counts, and scheduled jobs appropriate for the host’s actual CPU, memory, and storage capacity.
The durable fix is the one supported by evidence: isolate the offending plugin, remove the blocking workload, provide memory headroom, correct the version mismatch, or repair the shutdown automation. A larger heap or a disabled watchdog can change when the server fails, but neither explains why it stopped.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
