Skip to content
Tickwise
WarningPerformanceSIG-PA-001

The server cannot keep up with ticks (Can't keep up!)

Platforms
Paper · Purpur · single server
Verified
Verified
Confidence
Confidence 95%

What it looks like in the log

Any one of these lines is enough. The agent matches them locally, on your server.

logs/latest.log
  • [WARN] Can't keep up! Is the server overloaded?

Symptoms

  • Players see lag: movement rubber-banding, delayed block breaking, mobs “teleporting”
  • The log repeatedly shows “Can't keep up! Is the server overloaded? Running Nms or N ticks behind”

Cause

The server's main thread fell more than 2 seconds behind the tick schedule: one or more ticks took far longer than 50 ms. The message is printed by vanilla, Spigot, forks (Leaf, Purpur) and Paper before the tick loop rewrite; current Paper no longer prints it — there, overload shows up in MSPT.

Possible causes

  • Heavy plugins or their tasks on the main thread (synchronous database/HTTP requests, large operations)
  • World load: many entities/redstone/hoppers, chunk generation (pre-generation, players flying fast)
  • Resource shortage: the host CPU is overloaded by neighbours, long GC pauses due to a small heap

How to fixLow risk

  1. A single message right after startup (world loading) is normal; it is a concern if it keeps recurring during play.
  2. Profile during the lag: /spark profiler start, then /spark profiler stop after 2–5 minutes — see which plugins/tasks are occupying the main thread.
  3. Check /spark tps and /spark healthreport: MSPT, CPU load, GC pauses; with frequent long GCs, increase the heap or look for a leak.
  4. If pre-generation is running (Chunky and similar), slow it down or run it with no players online; limit entity farms and view-distance/simulation-distance.

Translated from the Russian original; log lines are quoted verbatim.

Sources