| 1 | # Health alerts for Cisco Nexus SNMP monitoring. |
| 2 | # These alerts fire on metrics from the cisco-nexus SNMP profile. |
| 3 | # Chart contexts use the pattern: snmp.device_prof_<metric_name> |
| 4 | |
| 5 | # --- vPC (Virtual Port Channel) alerts --- |
| 6 | |
| 7 | template: snmp_cisco_nexus_vpc_peer_keepalive |
| 8 | on: snmp.device_prof_cVpcPeerKeepAliveStatus |
| 9 | class: Errors |
| 10 | type: Switch |
| 11 | component: VPC |
| 12 | lookup: average -1m unaligned percentage of alive |
| 13 | units: % |
| 14 | every: 10s |
| 15 | warn: $this < 100 |
| 16 | delay: down 1m multiplier 1.5 max 5m |
| 17 | summary: vPC peer-keepalive not alive (domain ${label:vpc_domain_id}) |
| 18 | info: vPC peer-keepalive link is not in 'alive' state on domain ${label:vpc_domain_id} |
| 19 | to: sysadmin |
| 20 | |
| 21 | template: snmp_cisco_nexus_vpc_dual_active |
| 22 | on: snmp.device_prof_cVpcDualActiveDetectionStatus |
| 23 | class: Errors |
| 24 | type: Switch |
| 25 | component: VPC |
| 26 | lookup: average -1m unaligned percentage of detected |
| 27 | units: % |
| 28 | every: 10s |
| 29 | crit: $this > 0 |
| 30 | delay: down 15s multiplier 1.5 max 5m |
| 31 | summary: vPC dual-active (split-brain) detected (domain ${label:vpc_domain_id}) |
| 32 | info: vPC dual-active condition detected on domain ${label:vpc_domain_id} — both switches may be forwarding independently, risking loops |
| 33 | to: sysadmin |
| 34 | |
| 35 | template: snmp_cisco_nexus_vpc_member_consistency |
| 36 | on: snmp.device_prof_cVpcStatusHostLinkConsistencyStatus |
| 37 | class: Errors |
| 38 | type: Switch |
| 39 | component: VPC |
| 40 | lookup: average -1m unaligned percentage of failed |
| 41 | units: % |
| 42 | every: 10s |
| 43 | warn: $this > 0 |
| 44 | delay: down 1m multiplier 1.5 max 5m |
| 45 | summary: vPC member consistency check failed (domain ${label:vpc_domain_id} vPC ${label:vpc_id}) |
| 46 | info: vPC member ${label:vpc_id} on domain ${label:vpc_domain_id} has a configuration consistency check failure |
| 47 | to: sysadmin |
| 48 | |
| 49 | # --- HSRP (Hot Standby Router Protocol) alerts --- |
| 50 | |
| 51 | template: snmp_cisco_nexus_hsrp_not_active_standby |
| 52 | on: snmp.device_prof_cHsrpGrpStandbyState |
| 53 | class: Errors |
| 54 | type: Switch |
| 55 | component: HSRP |
| 56 | lookup: average -1m unaligned percentage of active,standby |
| 57 | units: % |
| 58 | every: 10s |
| 59 | warn: $this < 100 |
| 60 | delay: down 1m multiplier 1.5 max 5m |
| 61 | summary: HSRP group not active/standby (${label:interface} group ${label:hsrp_group}) |
| 62 | info: HSRP group ${label:hsrp_group} on ${label:interface} is not in active or standby state |
| 63 | to: sysadmin |
| 64 | |
| 65 | # --- BFD (Bidirectional Forwarding Detection) alerts --- |
| 66 | |
| 67 | template: snmp_cisco_nexus_bfd_session_down |
| 68 | on: snmp.device_prof_ciscoBfdSessState |
| 69 | class: Errors |
| 70 | type: Switch |
| 71 | component: BFD |
| 72 | lookup: average -1m unaligned percentage of down |
| 73 | units: % |
| 74 | every: 10s |
| 75 | warn: $this > 0 |
| 76 | delay: down 30s multiplier 1.5 max 5m |
| 77 | summary: BFD session down (session ${label:bfd_session_index}) |
| 78 | info: BFD session ${label:bfd_session_index} is in 'down' state — sub-second failure detection triggered |
| 79 | to: sysadmin |
| 80 | |
| 81 | # --- NTP alerts --- |
| 82 | |
| 83 | template: snmp_cisco_nexus_ntp_not_synchronized |
| 84 | on: snmp.device_prof_cntpSysSrvStatus |
| 85 | class: Errors |
| 86 | type: Switch |
| 87 | component: NTP |
| 88 | lookup: average -5m unaligned percentage of sync_to_refclock,sync_to_remote_server,sync_to_local |
| 89 | units: % |
| 90 | every: 10s |
| 91 | warn: $this < 100 |
| 92 | delay: down 5m multiplier 1.5 max 15m |
| 93 | summary: NTP service not synchronized |
| 94 | info: Cisco NTP service is not synchronized to any reference clock or remote server |
| 95 | to: sysadmin |
| 96 | |
| 97 | # --- NX-OS Feature Control alerts --- |
| 98 | |
| 99 | template: snmp_cisco_nexus_feature_enabled_not_running |
| 100 | on: snmp.device_prof_cfcFeatureCtrlOpStatus2 |
| 101 | class: Errors |
| 102 | type: Switch |
| 103 | component: NX-OS |
| 104 | lookup: average -1m unaligned percentage of enabled_not_running |
| 105 | units: % |
| 106 | every: 10s |
| 107 | warn: $this > 0 |
| 108 | delay: down 2m multiplier 1.5 max 10m |
| 109 | summary: NX-OS feature enabled but not running (${label:feature_name}) |
| 110 | info: NX-OS feature '${label:feature_name}' is enabled but not running — the feature process may have crashed |
| 111 | to: sysadmin |