Neutron

This section lists the alerts for Neutron:

NeutronAgentDisabled

Severity

Critical

Summary

{{ $labels.binary }} agent is disabled.

Description

The {{ $labels.binary }} agent is disabled in the {{ $labels.zone }} zone on all hosts.

NeutronAgentDown

Severity

Critical

Summary

{{ $labels.binary }} agent is down.

Description

The {{ $labels.binary }} agent is in the down state in the {{ $labels.zone }} zone on the {{ $labels.host }} host.

NeutronAgentOutage

Severity

Critical

Summary

{{ $labels.binary }} agent outage.

Description

The {{ $labels.binary }} agent is down in the {{ $labels.zone }} zone on all hosts where it is enabled.

NeutronSNATDestinationConnectionsHigh

Severity

Critical

Summary

Single destination consumes 50% or more of SNAT ports.

Description

Connections from the {{ $labels.router }} router ({{ $labels.netns }} namespace) on the {{ $labels.node }} node ({{ $labels.machine_name }} machine) to the single destination {{ $labels.dst_ip }}:{{ $labels.dst_port }} consume {{ $value | humanizePercentage }} of the usable SNAT source port range. A single destination, for example, a mis-routed floating IP, is about to exhaust the SNAT source ports of the namespace.

NeutronSNATExporterFailure

Severity

Critical

Summary

SNAT Exporter fails to discover router network namespaces.

Description

The SNAT Exporter on the {{ $labels.node }} node ({{ $labels.machine_name }} machine) fails to discover router network namespaces, for example, because the /run/netns host mount is missing or unreadable. SNAT monitoring of the entire node is unavailable.

NeutronSNATExporterTargetDown

Severity

Critical

Summary

SNAT Exporter Prometheus target is down.

Description

Prometheus fails to scrape metrics from the SNAT Exporter endpoint on the {{ $labels.node }} node ({{ $labels.machine_name }} machine).

NeutronSNATExporterTargetsOutage

Severity

Critical

Summary

SNAT Exporter Prometheus targets outage.

Description

Prometheus fails to scrape metrics from all SNAT Exporter endpoints.

NeutronSNATNamespaceCollectFailure

Severity

Critical

Summary

Failure to gather SNAT conntrack metrics for a network namespace.

Description

The SNAT Exporter fails to read the conntrack table of the {{ $labels.netns }} namespace ({{ $labels.router }} router) on the {{ $labels.node }} node ({{ $labels.machine_name }} machine) for 2 minutes.

NeutronSNATPortRangeMismatch

Severity

Warning

Summary

SNAT exporter port range does not match the NAT pool.

Description

The SNAT source port range configured on the {{ $labels.node }} node ({{ $labels.machine_name }} machine) does not match the NAT pool of the {{ $labels.router }} router ({{ $labels.netns }} namespace): a significant share of SNAT-ed connections use source ports outside the range. Align the exporter --snat.port-min and --snat.port-max options with the NAT pool of the cluster to make the snat_src_port_utilization_ratio metric reliable again.

NeutronSNATPortUsageCritical

Severity

Critical

Summary

SNAT source port utilization is 85% or above.

Description

SNAT source port utilization of the {{ $labels.router }} router ({{ $labels.netns }} namespace) on the {{ $labels.node }} node ({{ $labels.machine_name }} machine) is {{ $value | humanizePercentage }} for 2 minutes. Source port exhaustion is imminent, new outbound connections will start failing.

NeutronSNATPortUsageWarning

Severity

Warning

Summary

SNAT source port utilization is 70% or above.

Description

SNAT source port utilization of the {{ $labels.router }} router ({{ $labels.netns }} namespace) on the {{ $labels.node }} node ({{ $labels.machine_name }} machine) is {{ $value | humanizePercentage }} for 5 minutes. Growing connection count to a single destination may lead to source port exhaustion and degraded tenant traffic.