micro/go-micro · error

ErrPoolClosed

ErrPoolClosed

Error message

connection pool is closed

What it means

ErrPoolClosed is a sentinel error returned by pool.Get(), TestConnectionPool_Close, and TestTransportConnectionPool_Close when an operation attempts to acquire a connection from a pool that has already been closed. Once Close() is called the pool stops handing out connections permanently.

Source

Thrown at broker/nats/pool.go:15

package nats

import (
	"errors"
	"sync"
	"time"

	natsp "github.com/nats-io/nats.go"
)

var (
	// ErrPoolExhausted is returned when no connections are available in the pool
	ErrPoolExhausted = errors.New("connection pool exhausted")
	// ErrPoolClosed is returned when trying to use a closed pool
	ErrPoolClosed = errors.New("connection pool is closed")
)

// connectionPool manages a pool of NATS connections
type connectionPool struct {
	mu          sync.RWMutex
	connections chan *pooledConnection
	factory     func() (*natsp.Conn, error)
	size        int
	idleTimeout time.Duration
	closed      bool
}

// pooledConnection wraps a NATS connection with metadata
type pooledConnection struct {
	conn      *natsp.Conn
	createdAt time.Time
	lastUsed  time.Time
	mu        sync.Mutex

View on GitHub (pinned to 24529f1404)

Solutions

1. Call broker.Connect again to recreate the pool before subscribing after a Disconnect.
2. Fix lifecycle ordering: stop all consumers before closing the broker.
3. Check the broker's connection state before use in long-running services.

Example fix

// before
b.Disconnect()
sub, _ := b.Subscribe("events", handler) // ErrPoolClosed
// after
b.Disconnect()
if err := b.Connect(); err != nil { return err }
sub, _ := b.Subscribe("events", handler)

if err == natsbroker.ErrPoolClosed {
    // pool was closed; reconnect before subscribing
}

sub, err := b.Subscribe(topic, handler)
if errors.Is(err, natsbroker.ErrPoolClosed) {
    if cerr := b.Connect(); cerr != nil { return cerr }
    sub, err = b.Subscribe(topic, handler)
}

When it happens

Trigger: Calling Subscribe (which does pool.Get()) after broker.Disconnect/Close has closed the connection pool; reusing a broker instance after shutdown.

Common situations: Lifecycle ordering bugs — a shutdown handler closes the broker while background workers still try to subscribe/publish; reusing a cached broker client after application teardown.

Related errors

• ErrPoolExhausted
• connection pool is closed
• invalid connection from pool
• invalid connection from pool
• connection pool exhausted

AI-assisted analysis of micro/go-micro@24529f1404 (2026-09-01). Data as JSON: /api/errors/ca1f64b41a96c994. Report an issue: GitHub.