solaris.pro / Field manuals / Observability

Solaris performance triage

Determine whether CPU, memory, storage or network is limiting throughput

Scope: Oracle Solaris 11.4 and, for LDoms, compatible SPARC hardware with Oracle VM Server for SPARC. Examples are educational: verify release, privileges, target names and backups before running changes.

How it works

Baseline

Collect at normal load so incident values have context.

CPU

Run queue, saturation and per-process activity can tell different stories.

I/O

Separate queue depth and service time from application response time.

Memory

High cache use alone is not evidence of memory pressure.

Worked examples

Each example explains the question it answers, a reproducible starting point, what to look for, and operational precautions.

1. CPU and processes

Use this as a starting point for cpu and processes.

prstat 1

Interpretation: Check command output against the expected host configuration; record the before/after state.

Operational note: Run commands in a test environment first; confirm device names, privileges and release-specific options.

2. Memory and scheduling

Use this as a starting point for memory and scheduling.

vmstat 2

Interpretation: Check command output against the expected host configuration; record the before/after state.

Operational note: Run commands in a test environment first; confirm device names, privileges and release-specific options.

3. Disk latency

Use this as a starting point for disk latency.

iostat -xn 2

Interpretation: Check command output against the expected host configuration; record the before/after state.

Operational note: Run commands in a test environment first; confirm device names, privileges and release-specific options.

4. Pool activity

Use this as a starting point for pool activity.

zpool iostat -v 2

Interpretation: Check command output against the expected host configuration; record the before/after state.

Operational note: Run commands in a test environment first; confirm device names, privileges and release-specific options.

5. Network counters

Use this as a starting point for network counters.

netstat -s

Interpretation: Check command output against the expected host configuration; record the before/after state.

Operational note: Run commands in a test environment first; confirm device names, privileges and release-specific options.

Primary documentation

Read the relevant Oracle documentation for exact version-specific flags, supported platforms and limitations: Oracle primary reference ↗.

This field manual is independent, not affiliated with or endorsed by Oracle. Command examples should be adapted to your environment.