diff options
Diffstat (limited to 'monitoring/ceph-mixin/prometheus_alerts.yml')
| -rw-r--r-- | monitoring/ceph-mixin/prometheus_alerts.yml | 50 |
1 files changed, 43 insertions, 7 deletions
diff --git a/monitoring/ceph-mixin/prometheus_alerts.yml b/monitoring/ceph-mixin/prometheus_alerts.yml index ba6a6ded0a3..7c0da4d51a4 100644 --- a/monitoring/ceph-mixin/prometheus_alerts.yml +++ b/monitoring/ceph-mixin/prometheus_alerts.yml @@ -765,20 +765,29 @@ groups: labels: severity: "warning" type: "ceph_default" + - alert: "NVMeoFMultipleNamespacesOfRBDImage" + annotations: + description: "Each NVMeoF namespace must have a unique RBD pool and image, across all different gateway groups." + summary: "RBD image {{ $labels.pool_name }}/{{ $labels.rbd_name }} cannot be reused for multiple NVMeoF namespace " + expr: "count by(pool_name, rbd_name) (count by(bdev_name, pool_name, rbd_name) (ceph_nvmeof_bdev_metadata and on (bdev_name) ceph_nvmeof_subsystem_namespace_metadata)) > 1" + for: "1m" + labels: + severity: "warning" + type: "ceph_default" - alert: "NVMeoFTooManyGateways" annotations: - description: "You may create many gateways, but 4 is the tested limit" + description: "You may create many gateways, but 32 is the tested limit" summary: "Max supported gateways exceeded on cluster {{ $labels.cluster }}" - expr: "count(ceph_nvmeof_gateway_info) by (cluster) > 4.00" + expr: "count(ceph_nvmeof_gateway_info) by (cluster) > 32.00" for: "1m" labels: severity: "warning" type: "ceph_default" - alert: "NVMeoFMaxGatewayGroupSize" annotations: - description: "You may create many gateways in a gateway group, but 4 is the tested limit" + description: "You may create many gateways in a gateway group, but 8 is the tested limit" summary: "Max gateways within a gateway group ({{ $labels.group }}) exceeded on cluster {{ $labels.cluster }}" - expr: "count(ceph_nvmeof_gateway_info) by (cluster,group) > 4.00" + expr: "count(ceph_nvmeof_gateway_info) by (cluster,group) > 8.00" for: "1m" labels: severity: "warning" @@ -814,7 +823,16 @@ groups: annotations: description: "Although you may continue to create subsystems in {{ $labels.gateway_host }}, the configuration may not be supported" summary: "The number of subsystems defined to the gateway exceeds supported values on cluster {{ $labels.cluster }}" - expr: "count by(gateway_host, cluster) (label_replace(ceph_nvmeof_subsystem_metadata,\"gateway_host\",\"$1\",\"instance\",\"(.*):.*\")) > 16.00" + expr: "count by(gateway_host, cluster) (label_replace(ceph_nvmeof_subsystem_metadata,\"gateway_host\",\"$1\",\"instance\",\"(.*?)(?::.*)?\")) > 128.00" + for: "1m" + labels: + severity: "warning" + type: "ceph_default" + - alert: "NVMeoFTooManyNamespaces" + annotations: + description: "Although you may continue to create namespaces in {{ $labels.gateway_host }}, the configuration may not be supported" + summary: "The number of namespaces defined to the gateway exceeds 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" for: "1m" labels: severity: "warning" @@ -830,13 +848,31 @@ groups: type: "ceph_default" - alert: "NVMeoFHighClientCount" annotations: - description: "The supported limit for clients connecting to a subsystem is 32" + description: "The supported limit for clients connecting to a subsystem is 128" summary: "The number of clients connected to {{ $labels.nqn }} is too high on cluster {{ $labels.cluster }}" - expr: "ceph_nvmeof_subsystem_host_count > 32.00" + expr: "ceph_nvmeof_subsystem_host_count > 128.00" for: "1m" labels: severity: "warning" type: "ceph_default" + - alert: "NVMeoFMissingListener" + annotations: + description: "For every subsystem, each gateway should have a listener to balance traffic between gateways." + summary: "No listener added for {{ $labels.instance }} NVMe-oF Gateway to {{ $labels.nqn }} subsystem" + expr: "ceph_nvmeof_subsystem_listener_count == 0 and on(nqn) sum(ceph_nvmeof_subsystem_listener_count) by (nqn) > 0" + for: "10m" + labels: + severity: "warning" + type: "ceph_default" + - alert: "NVMeoFZeroListenerSubsystem" + annotations: + description: "NVMeoF gateway configuration incomplete; one of the subsystems have zero listeners." + summary: "No listeners added to {{ $labels.nqn }} subsystem" + expr: "sum(ceph_nvmeof_subsystem_listener_count) by (nqn) == 0" + for: "10m" + labels: + severity: "warning" + type: "ceph_default" - alert: "NVMeoFHighHostCPU" annotations: description: "High CPU on a gateway host can lead to CPU contention and performance degradation" |