Linux Server / Performance, Capacity & Containers
Inspect server TCP listen queues during connection spikes
Clients may time out while a daemon is listening because the application cannot accept connections fast enough. Observe queue behavior during the spike.
Environment and inputs: Linux server baseline · Ubuntu 24.04 LTS / RHEL 9-style systems. Use the specified distribution, init system and installed tool; service names, privileges and supported options can differ.
Start with this check
Confirm the host, shell and access role. Replace sample names, paths, domains and interfaces with approved targets. These are diagnostic commands; assess policy and workload impact before making changes.
ss -lntRead the result
For listening TCP sockets, queue fields have listener-specific semantics documented by the platform. A snapshot is not a count of all active application requests.
Resolve and verify
Correlate queue growth with worker availability, CPU, backend latency and connection rates. Fix the demonstrated accept or worker bottleneck before adjusting queue capacity.
Before changing production
Do not raise every backlog limit without reviewing application concurrency and memory capacity.