NVMeoFMaxGatewaysPerCluster: 32,
NVMeoFHighGatewayCPU: 80,
NVMeoFMaxSubsystemsPerGateway: 128,
- NVMeoFMaxNamespaces: 2048,
+ NVMeoFMaxNamespaces: 4096,
NVMeoFHighClientCount: 128,
NVMeoFHostKeepAliveTimeoutTrackDurationHours: 24,
NVMeoFHighHostCPU: 80,
type: "ceph_default"
- alert: "NVMeoFTooManyNamespaces"
annotations:
- description: "NVMeoF gateway {{ $labels.gateway_host }} has reached or exceeded the supported maximum of 2048 namespaces. Current count: {{ $value }}."
+ description: "NVMeoF gateway {{ $labels.gateway_host }} has reached or exceeded the supported maximum of 4096 namespaces. Current count: {{ $value }}."
summary: "The number of namespaces defined to the NVMeoF gateway reached or exceeded supported values on cluster {{ $labels.cluster }}"
- expr: "sum by(gateway_host, cluster) (label_replace(ceph_nvmeof_subsystem_namespace_count,\"gateway_host\",\"$1\",\"instance\",\"(.*?)(?::.*)?\")) >= 2048.00"
+ expr: "sum by(gateway_host, cluster) (label_replace(ceph_nvmeof_subsystem_namespace_count,\"gateway_host\",\"$1\",\"instance\",\"(.*?)(?::.*)?\")) >= 4096.00"
for: "1m"
labels:
severity: "warning"
- interval: 1m
input_series:
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn1",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn2",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn3",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn4",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn5",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn6",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn7",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn8",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn9",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn10",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
- series: 'ceph_nvmeof_subsystem_namespace_count{instance="node-1:10008",nqn="nqn11",cluster="mycluster"}'
- values: '200+0x10'
+ values: '400+0x10'
promql_expr_test:
- - expr: sum by(gateway_host, cluster) (label_replace(ceph_nvmeof_subsystem_namespace_count,"gateway_host","$1","instance","(.*):.*")) > 2048
+ - expr: sum by(gateway_host, cluster) (label_replace(ceph_nvmeof_subsystem_namespace_count,"gateway_host","$1","instance","(.*):.*")) > 4096
eval_time: 1m
exp_samples:
- labels: '{gateway_host="node-1", cluster="mycluster"}'
- value: 2200
+ value: 4400
alert_rule_test:
- eval_time: 5m
alertname: NVMeoFTooManyNamespaces
type: ceph_default
exp_annotations:
summary: "The number of namespaces defined to the NVMeoF gateway reached or exceeded supported values on cluster mycluster"
- description: "NVMeoF gateway node-1 has reached or exceeded the supported maximum of 2048 namespaces. Current count: 2200."
+ description: "NVMeoF gateway node-1 has reached or exceeded the supported maximum of 4096 namespaces. Current count: 4400."
# NVMeoFVersionMismatch
- interval: 1m