Satisfactory: Reading Server Logs

When your server stops responding or refuses to start, the `FactoryGame.log` is your only source of truth. Here is how to translate the technical gibberish into actionable fixes.

Common Error Signatures

Error: "Out of Video Memory trying to allocate a rendering resource"

• Cause: The server is trying to render graphics, which it shouldn't do.
• Fix: Ensure you are launching with the `-nullrhi` argument. This forces "Null Rendering Hardware Interface", telling the engine "I am a headless server, do not draw graphics."
• Fix 2: Sometimes this is actually a System RAM error disguised as VRAM. Upgrade your plan if you have a massive save file.

Error: "UNetConnection::Tick: Connection TIMED OUT"

• Cause: The server is overloaded and cannot send "heartbeats" to clients fast enough.
• Fix: Reduce `Max Autosave Frequency`. Autosaving freezes the main thread. If it takes 20 seconds to save, everyone times out.
• Fix 2: Check `NetworkQuality` settings in `Game.ini`.

Error: "Session ID: Invalid" or "Server is offline"

• Cause: The server failed to authenticate with Epic Online Services (EOS).
• Fix: Check your `FactoryGame.log` for "EOS: Login Failed". It usually means your server is not bound to 0.0.0.0 or the Query Port (7777) is blocked by a firewall.

Error: "SIGABRT (Abort)" or "Assertion Failed"

• Cause: A bug in the game engine or mod conflict. The server encountered a condition it can't handle.
• Fix: Disable all mods and try to start the server. If it works, re-enable mods one by one to identify the conflict.
• Fix 2: Update the server to the latest version. The bug might have been fixed in a patch.
• Fix 3: Check if you're running an outdated mod version. Update all mods to their latest compatible versions.

Error: "SIGSEGV (Segmentation Fault) - Address 0x0"

• Cause: Null pointer exception. The server tried to access memory that doesn't exist. Common with corrupted save files.
• Fix: Load an earlier save file. If the server won't start with any save, your installation might be corrupted.
• Fix 2: Verify the server files using SteamCMD: `app_update 1690800 validate`.
• Fix 3: If this happens consistently with a specific save, the save is corrupted. Restore from backup.

Error: "Memory allocation failed" or "Out of memory"

• Cause: Your save file has grown too large for your available RAM. Satisfactory servers with massive factories (1,000+ buildings) can consume 32GB+ of RAM.
• Fix: Upgrade to a higher RAM plan. 16GB is minimal, 32GB recommended for large factories.
• Fix 2: Reduce `Max Objects` in your server config. This limits the number of entities the server tracks.
• Fix 3: Delete unused buildings and decorations using SCIM. Smaller save files require less RAM.
• Fix 4: Use the `MaxAutosaveFrequency` setting to prevent autosaves from consuming too much memory.

Error: "Package could not be loaded" or "Missing package"

• Cause: The server is missing game files, or a mod requires assets that aren't installed.
• Fix: Run `app_update 1690800 validate` to repair the game installation.
• Fix 2: If you're using mods, ensure all required mod files are uploaded to the server's `mods/` directory.
• Fix 3: Check if you're trying to load a save with mods installed, but the server doesn't have those mods installed.

Error: "Failed to bind socket" or "Address already in use"

• Cause: Another process is using the required ports (7777, 15000, 15777), or a previous server instance didn't shut down properly.
• Fix: Check for stuck processes: `ps aux | grep FactoryGame`. Kill any stuck processes with `kill -9 [PID]`.
• Fix 2: Verify no other Satisfactory server is running on the same machine.
• Fix 3: Change the ports in your server's `Engine.ini` configuration file if you need to run multiple servers.

Error: "Unable to login to EOS"

• Cause: Epic Online Services authentication failure. Can be due to network issues or expired tokens.
• Fix: Restart the server. EOS tokens sometimes expire and need renewal.
• Fix 2: Check your internet connection. EOS requires outbound internet access.
• Fix 3: Verify your server's time is synchronized. Incorrect time can cause token validation failures.
• Fix 4: If using a VPN or proxy, ensure it's not blocking EOS endpoints.

Error: "Save file is incompatible"

• Cause: You're trying to load a save from a different game version (e.g., Experimental save on Stable server).
• Fix: Update the server to match the save's version. Use SteamCMD to switch to the correct branch (stable or experimental).
• Fix 2: If you downgraded from Experimental to Stable, you cannot load the Experimental save. You'll need to restore a pre-Experimental backup.
• Fix 3: Use SCIM to downgrade the save version if absolutely necessary (risky).

Error: "World save file is corrupted"

• Cause: The save file's internal data structures are damaged. Can happen due to unexpected shutdowns or disk errors.
• Fix: Load the most recent backup. Supercraft creates automatic backups.
• Fix 2: Use SCIM's repair function. Sometimes minor corruption can be fixed.
• Fix 3: As a last resort, use SCIM to export buildings and recreate the world. You'll lose some progress but can recover the factory.
• Prevention: Ensure your server has adequate disk space and the filesystem is healthy. Run disk checks periodically.

Error: "Physics engine failed to initialize"

• Cause: The physics simulation encountered a critical error during world loading.
• Fix: Disable advanced physics in `Game.ini`. Set `bPhysicsOptimization=1` and `MaxPhysicsThreads=1`.
• Fix 2: Remove complex physics objects (e.g., massive chains of vehicles) using SCIM.
• Fix 3: Reduce the number of physics objects in your save. Vehicles, trains, and factories with many moving parts consume physics resources.

Log Analysis Workflow

Step 1: Find the Critical Error

1. Open `FactoryGame.log` in a text editor.
2. Search for: `Error:` (with colon) to find all error messages.
3. Search for: `Fatal:` to find fatal errors that caused the crash.
4. Look for the last error before the crash - that's usually the root cause.

Step 2: Check Timestamps

Each log entry has a timestamp in the format: `Log: [2024.01.09-10.30.45:123]`

• Compare the crash timestamp with your actions. Did the crash happen during autosave?
• Look for patterns. Does the crash happen at the same time daily?
• Check for player join/leave events. Sometimes crashes correlate with specific player actions.

Step 3: Identify the Module

Log entries indicate which game module generated the error:

• LogCore: Core engine issues (memory, files, initialization).
• LogNet: Network/Connectivity problems.
• LogRenderer: Graphics/rendering errors (shouldn't appear on servers with `-nullrhi`).
• LogScript: Script or mod errors.
• LogPhysics: Physics simulation errors.

Step 4: Check for Memory Patterns

Look for these memory-related patterns in the minutes before the crash:

• Increasing memory allocation warnings (e.g., " allocating 4GB" repeatedly).
• Garbage collection warnings (e.g., "Garbage collection taking too long").
• Fragmentation warnings (e.g., "Memory is heavily fragmented").

Step 5: Autosave Correlation

Many crashes happen during autosaves. Look for:

• LogSaveGame: Shows when autosaves start and complete.
• Large time gaps between "Saving..." and "Save complete" - indicates autosave is taking too long.
• If crashes consistently occur during autosaves, increase your autosave interval or save size limits.

Performance Metrics in Logs

Frame Time Monitoring

The logs track server frame time (time to process one game tick):

• Normal: 16.67ms or lower (60 FPS).
• Lag Warning: 33ms+ (30 FPS or below).
• Severe Lag: 100ms+ (10 FPS or below - unplayable).

Server Load Patterns

Look for these patterns to understand server load:

• Tick Rate: The game should tick at 30 Hz. If you see "Tick took 100ms", the server is overloaded.
• Player Count: More players = more calculations. Track how performance degrades as player count increases.
• Entity Count: High entity counts (factories, vehicles, items) increase CPU load. Look for warnings like "Too many entities".

Common Warning Patterns

Warning: "Out of Memory - Garbage Collection"

This is not fatal yet, but indicates the server is running low on memory.

• Immediate Action: Upgrade RAM plan or reduce factory complexity.
• Long-term: This will eventually lead to a crash.

Warning: "Network packet dropped"

Indicates network issues between server and clients.

• If client-side: Check player's internet connection.
• If server-side: Check server's network bandwidth and firewall rules.

Warning: "Vehicle physics error"

A vehicle encountered a physics problem. Usually harmless but can accumulate.

• Action: Check the specific vehicle in-game. Remove it if it's glitched.
• Prevention: Avoid placing vehicles in impossible locations (e.g., inside other objects).

How to Share Logs

Never paste the entire 10MB log file into Discord.

1. Open the file.
2. Scroll to the bottom (most recent events).
3. Copy the last 50 lines.
4. Look for the phrase "Error:" or "Warning:".

Log Backup & Rotation

Automatic Rotation

Satisfactory creates new log files periodically:

• Current: `FactoryGame.log` (active session).
• Previous: `FactoryGame_1.log`, `FactoryGame_2.log` (up to 5 backups).
• Retention: Logs are kept for approximately 7 days before automatic deletion.

Manual Backup Strategy

For recurring issues, archive logs for analysis:

1. Copy `FactoryGame.log` to a safe location with timestamp (e.g., `FactoryGame_crash_2024-01-09_1030.log`).
2. Document what you were doing before the crash (e.g., "Crash during autosave with 5 players online").
3. Store these archives for at least 30 days to identify patterns.

Zero downtime. Host your Satisfactory server with Supercraft and let our automated watchdog system monitor your logs and restart the server if it hangs.