Contrail Networking Alarms

Table 1 lists the default alarms in Contrail Networking and their severity levels.

An alarm with severity level 0 (zero) is critical, 1 (one) is major, and 2 (two) is minor.

Table 1: Contrail Networking Alarms and Severity Level
Alarm Name	Severity	Description	Steps to Resolve This Alarm
system-defined-address-mismatch-compute	1	Compute Node IP Address mismatch.	The compute node IP address provided in the configuration file and the IP address provided as part of creating (provisioning) vrouter-agent do not match.
system-defined-address-mismatch-control	1	Control Node IP Address mismatch.	IP address for control node is different in config node and control node.
system-defined-bgp-connectivity	1	BGP peer mismatch. Not enough BGP peers are up.	Total number of BGP peers is different from the configured number of BGP peers.
system-defined-bottle-request-size-limit	-	Bottle request size limit exceeded.	Request Size received by API server is too large. In most cases, this can be resolved by increasing the value set for the variable `max_request_size` in the /etc/contrail/contrail-api.conf file in config API Docker container. However, as a good practice, investigate as to why such a huge request is being sent to the Config API server.
system-defined-conf-incorrect	1	ContrailConfig missing or incorrect. Configuration pushed to Ifmap as ContrailConfig is missing or incorrect.	Config node did not send ContrailConfig for this node. This could be due to name mismatch between the node configured compared to actual node.
system-defined-disk-usage-high	1	Disk usage exceeds high threshold limit.	Corresponding disk is filled between 70%-90% capacity. Delete some files to create disk space.
system-defined-disk-usage-critical	0	Disk usage crosses critical threshold limit.	Corresponding disk is filled up > 90%. Delete some files to create disk space.
system-defined-node-status	0	Node Failure. NodeStatus UVE not present.	NodeStatus UVE is not present or process is non-functional for this node. Verify that the `process` and `nodemgr` is up.
system-defined-partial-sysinfo	1	System Info Incomplete.	build_info is not present in NodeStatus. Cause unknown at this time.
system-defined-process-connectivity	0	Process(es) reporting as non-functional.	One or more processes have connections missing.
system-defined-process-status	0	Process Failure.	Review the docker logs to understand the reason for process failure.
system-defined-prouter-connectivity	1	Prouter connectivity to controlling tor agent does not exist. Contrail looks for non-empty value for connected_agent_list	Check for OVSDB connectivity status on the physical device. Debug for link failures between physical device or OVSDB connection failure between the vrouter-agent and physical router.
system-defined-prouter-tsn-connectivity	1	Prouter connectivity to controlling TSN agent does not exist. Contrail looks for non-empty value for tsn_agent_list.	Check for OVSDB connectivity status on the physical device. Debug for link failures between physical device or OVSDB connection failure between the vrouter-agent and physical router.
system-defined-storage-cluster-state	1	Storage Cluster warning or errors.	Since Contrail is not provisioning storage this alarm is not generated.
system-defined-vrouter-interface	1	vRouter interface(s) down.	This alarm is raised if forwarding and bridging is disabled or if health check has failed. Other reasons for this alarm include the following: no IP or subnet assignment admin state is down parent interface is down VLAN is down oper state is down config is missing Resolve above items based on information available from the introspect page for interface.
system-defined-xmpp-connectivity	1	XMPP peer mismatch.	Number of XMPP peers is different from configured XMPP peers.
system-defined-xmpp-close-reason	1	XMPP connection closed towards peer. Alarm has reason to close.	This alarm is deprecated.
system-defined-core-files	0	A core file has been generated on the node.	There is some core file in the node.
system-defined-pending-cassandra-compaction-tasks	1	Pending compaction tasks in cassandra crossed the configured threshold.	This alarm is raised when disk space is insufficient. Check Cassandra system logs to understand the reason for pending compaction.
system-defined-package-version-mismatch	0	There is a mismatch between installed and running package version.	Package version for the package mentioned in the alarm is not matching with the required version.
system-defined-vrouter-limit-exceeded	1	Agent resource usage exceeded configured watermark for resource.	This alarm is raised when the next hop count or the used mpls label count crosses the high watermark. The alarm is reset when the next hop count or the used MPLS label count becomes less than the low watermark. To reset alarm, delete the `nexthop` and `mpls` label, which can be achieved by deleting virtual machines on the compute. Alarm can also be cleared by increasing the default watermark, which is 80 (80% of the maximum number of nexthops configured in vRouter after which alarm is raised). For this, you need to change the configuration in the contrail-vrouter-agent.conf file and restart the vRouter agent.
system-defined-vrouter-table-limit-exceeded	0	Agent resource usage exceeded table size for resource in vRouter.	This alarm is raised when the next hop count reaches the `nexthop count` configured in vRouter, or when the maximum number of MPLS labels on the compute are used. This alarm is cleared when the next hop count goes below 95% of the next hop count in vRouter, or the number of used MPLS label count becomes 95 % of the maximum labels or less. To reset the alarm, delete the nexthop and MPLS labels, which can be achieved by deleting virtual machines on the compute for which alarm is raised. This alarm can also be reset by increasing the maximum number of `nexthop` and MPLS labels configured in vRouter, if it is not already configured to the maximum supported limit.