hashicorp/nomad · error

failed to write bytes to eventstream.json; bytes written: %d

Error message

failed to write bytes to eventstream.json; bytes written: %d, Err: %w

What it means

After successfully marshaling an event, captureEventStream writes the JSON bytes to the open eventstream.json file handle. If fh.Write fails (bytes may be partially written, n is reported in the message), the error is appended to the multierror and the event loop breaks out. This indicates the capture destination became unwritable mid-capture.

Source

Thrown at command/operator_debug.go:992

			if event.IsHeartbeat() {
				heartbeatCount++
				continue
			}

			for _, e := range event.Events {
				eventCount++
				c.verboseOutf("Event: %4d, Index: %d, Topic: %-10s, Type: %s, FilterKeys: %s", eventCount, e.Index, e.Topic, e.Type, e.FilterKeys)

				bytes, err := json.Marshal(e)
				if err != nil {
					errCount++
					mErrs = multierror.Append(mErrs, fmt.Errorf("failed to marshal json from Topic: %s, Type: %s, Err: %w", e.Topic, e.Type, err))
				}

				n, err := fh.Write(bytes)
				if err != nil {
					errCount++
					mErrs = multierror.Append(mErrs, fmt.Errorf("failed to write bytes to eventstream.json; bytes written: %d, Err: %w", n, err))
					break
				}
				n, err = fh.WriteString("\n")
				if err != nil {
					errCount++
					mErrs = multierror.Append(mErrs, fmt.Errorf("failed to write string to eventstream.json; chars written: %d, Err: %w", n, err))
				}
			}
		case <-c.ctx.Done():
			c.verboseOutf("Event stream captured %d events, %d frames, %d heartbeats, %d errors", eventCount, channelEventCount, heartbeatCount, errCount)
			return mErrs.ErrorOrNil()
		}
	}
}

// collectAgentHosts calls collectAgentHost for each selected node
func (c *OperatorDebugCommand) collectAgentHosts(client *api.Client) {
	for _, n := range c.nodeIDs {

View on GitHub (pinned to 482b49bf1a)

Solutions

1. Free disk space or point the capture at a directory with room (df -h; choose a larger -output target)
2. Check the partial byte count n in the message to assess how much of the event was written
3. Verify the capture directory still exists and is writable during the run (ls -ld, no concurrent cleanup jobs)
4. Re-run the capture after fixing storage; the multierror lists how many events were affected via errCount

// preflight
if st, err := os.Stat(outputDir); err != nil || !st.IsDir() {
	return fmt.Errorf("output dir %s unavailable", outputDir)
}
if du, err := checkFreeSpace(outputDir); err != nil || du < minBytes {
	return fmt.Errorf("insufficient space in %s", outputDir)
}

n, err := fh.Write(bytes)
if err != nil {
	errs = multierror.Append(errs, fmt.Errorf("failed to write bytes to eventstream.json; bytes written: %d, Err: %w", n, err))
	break // stop consuming a broken file handle
}

When it happens

Trigger: fh.Write returns an error: disk full, capture directory removed while capturing, file descriptor closed, permission changed, or I/O error on the underlying filesystem.

Common situations: /tmp or the -output dir filling up during long captures; someone deleting the capture directory while debug runs; running on a read-only or quota-limited filesystem; container tmpfs size limits.

Understand the failure class

Background: "failed to write file", "Could not save figure", "Error saving remote file" — file write failed: causes and fixes across languages and libraries — this error's family across 38 libraries.

Related errors

• failed to write string to eventstream.json; chars written: %
• unable to read rooted allocation directory
• can't seek to offset %d: %w
• Couldn't copy %q to %q: %w
• error writing to file %q: %w

AI-assisted analysis of hashicorp/nomad@482b49bf1a (2026-09-04). Data as JSON: /api/errors/d96eae8c5f563ceb. Report an issue: GitHub.