bstein/titan-iac
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

monitoring: fix gpu pie data and network panels

Browse Source
This commit is contained in:
Brad Stein 2025-11-18 00:31:51 -03:00
parent ac62387e07
commit 5144bbe1f2
5 changed files with 19 additions and 27 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
scripts/render_dashboards.py
View File

@ -178,10 +178,7 @@ def namespace_ram_share_expr():
def namespace_gpu_share_expr():
selected = (
f"(( {NAMESPACE_GPU_RAW} ) and on(namespace) ( {NAMESPACE_COMBINED_FILTER} ))"
f" or on(namespace) ( {NAMESPACE_COMBINED_FILTER} * 0)"
)
selected = f"(( {NAMESPACE_GPU_RAW} ) or on(namespace) ( {NAMESPACE_COMBINED_FILTER} * 0)) and on(namespace) ( {NAMESPACE_COMBINED_FILTER} )"
total = f"clamp_min(sum( {NAMESPACE_GPU_RAW} ), 1)"
return f"100 * ( {selected} ) / {total}"
@ -226,7 +223,8 @@ NAMESPACE_RAM_RAW = (
'sum(container_memory_working_set_bytes{namespace!="",pod!="",container!=""}) by (namespace)'
)
NAMESPACE_GPU_RAW = (
'sum(kube_pod_container_resource_requests{namespace!="",resource="nvidia.com/gpu"}) by (namespace)'
'sum(kube_pod_container_resource_requests{namespace!="",resource="nvidia.com/gpu"} '
'or kube_pod_resource_request{namespace!="",resource="nvidia.com/gpu"}) by (namespace)'
)
NAMESPACE_GPU_WEIGHT = (
"(( "
@ -245,14 +243,8 @@ NAMESPACE_COMBINED_FILTER = (
+ " * 10))"
)
TRAEFIK_ROUTER_EXPR = "sum by (router) (rate(traefik_router_requests_total[5m]))"
NET_INGRESS_EXPR = (
'sum(rate(container_network_receive_bytes_total{namespace!="" ,pod!=""}[5m])) '
"or on() vector(0)"
)
NET_EGRESS_EXPR = (
'sum(rate(container_network_transmit_bytes_total{namespace!="" ,pod!=""}[5m])) '
"or on() vector(0)"
)
NET_INGRESS_EXPR = 'sum(rate(node_network_receive_bytes_total{device!~"lo"}[5m])) or on() vector(0)'
NET_EGRESS_EXPR = 'sum(rate(node_network_transmit_bytes_total{device!~"lo"}[5m])) or on() vector(0)'
# ---------------------------------------------------------------------------
# Panel factories

4
services/monitoring/dashboards/atlas-network.json
View File

@ -20,7 +20,7 @@
},
"targets": [
{
"expr": "sum(rate(container_network_receive_bytes_total{namespace!=\"\" ,pod!=\"\"}[5m])) or on() vector(0)",
"expr": "sum(rate(node_network_receive_bytes_total{device!~\"lo\"}[5m])) or on() vector(0)",
"refId": "A"
}
],
@ -80,7 +80,7 @@
},
"targets": [
{
"expr": "sum(rate(container_network_transmit_bytes_total{namespace!=\"\" ,pod!=\"\"}[5m])) or on() vector(0)",
"expr": "sum(rate(node_network_transmit_bytes_total{device!~\"lo\"}[5m])) or on() vector(0)",
"refId": "A"
}
],

10
services/monitoring/dashboards/atlas-overview.json
View File

@ -722,7 +722,7 @@
},
"targets": [
{
"expr": "100 * ( ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ), 1)",
"expr": "100 * ( ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -764,7 +764,7 @@
},
"targets": [
{
"expr": "100 * ( (( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) )) or on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) * 0) ) / clamp_min(sum( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ), 1)",
"expr": "100 * ( (( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) * 0)) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -806,7 +806,7 @@
},
"targets": [
{
"expr": "100 * ( ( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ), 1)",
"expr": "100 * ( ( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -1016,7 +1016,7 @@
},
"targets": [
{
"expr": "sum(rate(container_network_receive_bytes_total{namespace!=\"\" ,pod!=\"\"}[5m])) or on() vector(0)",
"expr": "sum(rate(node_network_receive_bytes_total{device!~\"lo\"}[5m])) or on() vector(0)",
"refId": "A",
"legendFormat": "Ingress"
}
@ -1060,7 +1060,7 @@
},
"targets": [
{
"expr": "sum(rate(container_network_transmit_bytes_total{namespace!=\"\" ,pod!=\"\"}[5m])) or on() vector(0)",
"expr": "sum(rate(node_network_transmit_bytes_total{device!~\"lo\"}[5m])) or on() vector(0)",
"refId": "A",
"legendFormat": "Egress"
}

4
services/monitoring/grafana-dashboard-network.yaml
View File

@ -29,7 +29,7 @@ data:
},
"targets": [
{
"expr": "sum(rate(container_network_receive_bytes_total{namespace!=\"\" ,pod!=\"\"}[5m])) or on() vector(0)",
"expr": "sum(rate(node_network_receive_bytes_total{device!~\"lo\"}[5m])) or on() vector(0)",
"refId": "A"
}
],
@ -89,7 +89,7 @@ data:
},
"targets": [
{
"expr": "sum(rate(container_network_transmit_bytes_total{namespace!=\"\" ,pod!=\"\"}[5m])) or on() vector(0)",
"expr": "sum(rate(node_network_transmit_bytes_total{device!~\"lo\"}[5m])) or on() vector(0)",
"refId": "A"
}
],

10
services/monitoring/grafana-dashboard-overview.yaml
View File

@ -731,7 +731,7 @@ data:
},
"targets": [
{
"expr": "100 * ( ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ), 1)",
"expr": "100 * ( ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -773,7 +773,7 @@ data:
},
"targets": [
{
"expr": "100 * ( (( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) )) or on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) * 0) ) / clamp_min(sum( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ), 1)",
"expr": "100 * ( (( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) * 0)) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -815,7 +815,7 @@ data:
},
"targets": [
{
"expr": "100 * ( ( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ), 1)",
"expr": "100 * ( ( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -1025,7 +1025,7 @@ data:
},
"targets": [
{
"expr": "sum(rate(container_network_receive_bytes_total{namespace!=\"\" ,pod!=\"\"}[5m])) or on() vector(0)",
"expr": "sum(rate(node_network_receive_bytes_total{device!~\"lo\"}[5m])) or on() vector(0)",
"refId": "A",
"legendFormat": "Ingress"
}
@ -1069,7 +1069,7 @@ data:
},
"targets": [
{
"expr": "sum(rate(container_network_transmit_bytes_total{namespace!=\"\" ,pod!=\"\"}[5m])) or on() vector(0)",
"expr": "sum(rate(node_network_transmit_bytes_total{device!~\"lo\"}[5m])) or on() vector(0)",
"refId": "A",
"legendFormat": "Egress"
}