risingwavelabs/risingwave · warning · DashboardError

CPU profiling duration must be greater than zero

Error message

CPU profiling duration must be greater than zero

What it means

The dashboard `cpu_profile` handler validates that the requested profiling `duration_secs` path parameter is strictly greater than zero. A zero-second profile is meaningless (no samples could be collected), so the request is rejected before starting the profiler on the worker.

Solutions

  1. Pass a positive duration in seconds, e.g. /api/v1/cpu_profile/1/10 for a 10-second profile.
  2. Guard the calling script: skip the request when the computed duration is <= 0.
  3. Check the URL path segments are in the right order — worker_id first, duration second.

Example fix

// before
curl http://localhost:5691/api/v1/cpu_profile/1/0
// after
curl http://localhost:5691/api/v1/cpu_profile/1/15
Defensive patterns

Strategy: validation

Validate before calling

const duration = Number(opts.durationSecs);
if (!Number.isInteger(duration) || duration <= 0) throw new Error('CPU profiling duration must be a positive integer of seconds');

Type guard

function isValidDurationSecs(n) { return Number.isInteger(n) && n > 0; }

Try / catch

try { const res = await fetch(`/api/v1/cpu_profile/${workerId}/${duration}`); if (!res.ok) throw new Error(await res.text()); } catch (e) { /* surface duration validation to the user */ }

Prevention

When it happens

Trigger: Calling GET /api/v1/cpu_profile/{worker_id}/{duration_secs} with duration_secs=0, e.g. `curl http://localhost:5691/api/v1/cpu_profile/1/0`.

Common situations: Templated scripts computing a duration from an empty/unset variable so it defaults to 0; automated tooling that divides a budget into zero-second slices; hand-typed URLs testing the endpoint.

Understand the failure class

Background: "value must be between 0 and 1" / "out of range" / "must not be negative" errors: fixing range-validation failures across open-source libraries — this error's family across 42 libraries.

Related errors

AI-assisted analysis of risingwavelabs/risingwave@6469eb736d (2026-09-11). Data as JSON: /api/errors/bd280294496a0754. Report an issue: GitHub.

Appendix: source

Thrown at src/meta/src/dashboard/mod.rs:570

            .get_worker_by_id(worker_id)
            .await
            .map_err(err)?
            .context("worker node not found")
            .map_err(err)?;

        let client = srv.monitor_clients.get(&worker_node).await.map_err(err)?;

        let result = client.heap_profile("".to_owned()).await.map_err(err)?;

        Ok(result.into())
    }

    pub async fn cpu_profile(
        Path((worker_id, duration_secs)): Path<(WorkerId, u64)>,
        Extension(srv): Extension<Service>,
    ) -> Result<Response> {
        if duration_secs == 0 {
            return Err(err(anyhow!(
                "CPU profiling duration must be greater than zero"
            )));
        }

        let flamegraph = if worker_id == crate::manager::META_NODE_ID {
            srv.profile_service
                .profiling(Request::new(ProfilingRequest {
                    sleep_s: duration_secs,
                }))
                .await
                .map_err(err)?
                .into_inner()
                .result
        } else {
            let worker_node = srv
                .metadata_manager
                .get_worker_by_id(worker_id)
                .await

View on GitHub (pinned to 6469eb736d)