failure

Built-in failure-oriented update policies.

Modules

availability_flap(down_probability, *[, ...])	Toggle node availability up and down according to separate probabilities.
brownout([probability, factor, node_assets, ...])	Apply partial service degradation without a hard failure.
correlated_failure(probability, *, group_key)	Fail all selected nodes sharing a group value together.
edge_availability_flap(down_probability, *)	Toggle edge availability up and down according to probabilities.
kill_edges(probability, *[, ...])	Mark selected edges as unavailable according to probability.
kill_nodes(probability, *[, ...])	Randomly mark selected nodes as unavailable, with optional revival.
latency_spike(probability, *[, ...])	Inject random latency spikes on selected edges.
network_partition(groups, *[, ...])	Partition node groups by disabling or removing cross-group edges.
resource_exhaustion([probability, factor, ...])	Reduce selected capacity-like assets according to factor.
revive_edges(probability, *[, ...])	Mark selected edges as available according to probability.
revive_nodes(probability, *[, availability, ...])	Randomly restore selected unavailable nodes.