A batch run skips the nodes waiting on a failed one and ends error; a run whose nodes cannot load fails with the load error before anything executes; cancelling a run stops its pipeline without announcing a pause of the live flow; batch_issues refuses a delay node's interval; the client accepts an engine URL with or without /api/v1; RedisState's compare-and-swap records the names it writes

This commit is contained in:
2026-11-04 21:38:21 +01:00
parent eb17528ce6
commit d028bf326a
11 changed files with 221 additions and 32 deletions
+7 -2
View File
@@ -18,9 +18,14 @@ A batch flow does not need activation and does not open subscriptions,
schedules or webhooks. A run executes each node once after its dependencies
finish. Missing parameters use their declared initial values.
When a node fails, the nodes that wait on it are recorded as `skipped` and
the run ends `error`. A node that cannot be loaded fails the run before
anything executes, with the node's load error as the reason.
Non-streaming ports cannot use `interval` in a batch flow: there is no timer
to deliver a delayed value. Streaming ports can use it to reduce display
updates while keeping every metric point.
to deliver a delayed value. The same goes for a delay node's `interval`
setting. Streaming ports can use it to reduce display updates while keeping
every metric point.
## Submitting