Skip to main content
Reports how much memory the run is holding, and which parts of it, at the point in the flow where you place the node. The run continues unchanged — this node only ever describes a run, it never alters one.

​
When to use it

Put one at the top of a Flow Each body when a run is using more memory than it should. Each iteration logs a reading, and a field that climbs every time is the thing accumulating. Two probes either side of a section tell you what that section held on to.

​
Configuration

  • Label (optional) — names the probe in the log, so several in one flow are easy to tell apart. {assign_name} is replaced with that assign’s value at run time.
  • Fields — how many of the largest parts to list. Defaults to 5.

​
Reading the output

  • Held is the total the run is using right now. It goes up and down as the VM reclaims memory, so a single reading means little; a series from the same probe is what to look at.
  • Change compares against the previous time this same probe ran. A positive number every iteration means the flow is accumulating.
  • The listed fields are the largest contributors, biggest first.

​
Cost

Measuring is not free, and it gets slower as the run’s data grows. Inside a large loop that is noticeable, so take the probe out once you have your answer. The run’s Timings view shows exactly what the probe cost.

​
Ports