Skip to contents

Combines .diag timestamps, SLURM sacct times, and event store data to produce a chronological view of what happened during a pipeline run.

Usage

failure_timeline(x, around_failure = 60, ...)

Arguments

x

A parade_deferred object or collected results data.frame

around_failure

Seconds of context to show around each failure event

...

Additional arguments (unused)

Value

A tibble with columns: offset, offset_str, event, chunk_id, stage, context, class

Examples

# \donttest{
grid <- data.frame(x = 1:4, group = rep(c("A", "B"), 2))
fl <- flow(grid) |>
  stage("calc", function(x) x^2, schema = returns(result = dbl())) |>
  distribute(dist_local(by = "group"))
d <- submit(fl)
deferred_await(d, timeout = 60)
tl <- failure_timeline(d)
# }