What causes "error: can't ping self" in ipfs/kubo?

Ping a different peer: get a remote peer ID from `ipfs swarm peers` or `ipfs dht findprovidercmd` output and use that.

error: can't ping self

Error message

error: can't ping self

What it means

`ErrPingSelf` is a sentinel error returned by the ping command when the target peer ID equals the local node's own identity (`n.Identity`). Pinging yourself is meaningless in the DHT/ping protocol, so kubo rejects it upfront. The doc comment explicitly marks it as the error for attempting to ping yourself.

Source

Thrown at core/commands/ping.go:33

	pstore "github.com/libp2p/go-libp2p/core/peerstore"
	ping "github.com/libp2p/go-libp2p/p2p/protocol/ping"
	ma "github.com/multiformats/go-multiaddr"
)

const kPingTimeout = 10 * time.Second

type PingResult struct {
	Success bool
	Time    time.Duration
	Text    string
}

const (
	pingCountOptionName = "count"
)

// ErrPingSelf is returned when the user attempts to ping themself.
var ErrPingSelf = errors.New("error: can't ping self")

var PingCmd = &cmds.Command{
	Helptext: cmds.HelpText{
		Tagline: "Send echo request packets to IPFS hosts.",
		ShortDescription: `
'ipfs ping' is a tool to test sending data to other nodes. It finds nodes
via the routing system, sends pings, waits for pongs, and prints out round-
trip latency information.
		`,
	},
	Arguments: []cmds.Argument{
		cmds.StringArg("peer ID", true, true, "ID of peer to be pinged.").EnableStdin(),
	},
	Options: []cmds.Option{
		cmds.IntOption(pingCountOptionName, "n", "Number of ping messages to send.").WithDefault(10),
	},
	Run: func(req *cmds.Request, res cmds.ResponseEmitter, env cmds.Environment) error {
		n, err := cmdenv.GetNode(env)

View on GitHub (pinned to 329838acdf)

Solutions

Ping a different peer: get a remote peer ID from `ipfs swarm peers` or `ipfs dht findprovidercmd` output and use that.
Compare the target with `ipfs id --format='<id>'` before pinging and skip when equal (self-check in scripts).
If you meant to test connectivity to yourself, use the daemon's own endpoints instead (e.g. `ipfs id`, API health) rather than `ipfs ping`.
In code, check for `ErrPingSelf` with `errors.Is(err, corecommands.ErrPingSelf)` and treat it as a user mistake, not a network failure.

Example fix

// before (shell)
SELF=$(ipfs id --format='<id>')
ipfs ping "$SELF"
// after
SELF=$(ipfs id --format='<id>')
TARGET="$argv[1]"
[ "$TARGET" != "$SELF" ] && ipfs ping "$TARGET" || echo 'refusing to ping self'

Defensive patterns

Strategy: try-catch

Validate before calling

target, _ := peer.Decode(args[0])
self, _ := node.Identity.Decode()
if target == self {
    return fmt.Errorf("refusing to ping self")
}

Type guard

func isPingSelf(err error) bool {
    return errors.Is(err, ErrPingSelf)
}

Try / catch

pid, err := ipfsPing(ctx, target)
if err != nil {
    if errors.Is(err, ErrPingSelf) {
        // user mistake: skip or report, do not retry
        return fmt.Errorf("target %s is the local node; pick a remote peer", target)
    }
    return err
}

Prevention

Filter the local node's own peer ID out of any peer list before pinging
Resolve multiaddrs first and compare the decoded peer ID against `ipfs id`
Treat ErrPingSelf as a usage error in tooling, distinct from network timeouts
Never copy the peer ID from `ipfs id` when testing local connectivity

When it happens

Trigger: Running `ipfs ping <own-peer-id>` where the argument is the node's own ID from `ipfs id`; scripts that resolve a peer ID via DNSLink/multiaddr and accidentally resolve back to the local node; clusters/tests where a node list accidentally includes self.

Common situations: Copy-pasting the peer ID from `ipfs id` on the same machine you're testing from; loopback multiaddr pings (`ipfs ping /p2p/<self>`); automation iterating over a peer list that includes the local node without filtering.

Related errors

AI-assisted analysis of ipfs/kubo@329838acdf (2026-09-03). Data as JSON: /api/errors/9625a1e8a4511432. Report an issue: GitHub.