feat: peer occupancy panels for the out-peers experiment
opened MINUS closed per direction, plotted against the caps. Occupancy is the measurement that matters and cumulative opens actively mislead: inbound churns constantly (170 opened, 156 closed) while outbound is deliberately maintained (36 opened, 27 closed), so a high inbound open count says nothing about whether its cap is full. Reading the ratio of cumulative opens is how issue #1 came to recommend the wrong knob. Measured now: outbound 9 held against a cap of 8 -- saturated -- while inbound sits at 14 of 32 and the node has discovered 300 peers. So --out-peers binds and --in-peers does not, the reverse of what #1 said. No exporter change needed; both series are already scraped from the node. Adds the discovered-peers count and rounds-per-minute alongside, since those are the other two the experiment compares.
This commit is contained in:
Rob Thijssen
+230
-1
@@ -9,7 +9,7 @@
|
||||
],
|
||||
"timezone": "browser",
|
||||
"schemaVersion": 39,
|
||||
"version": 5,
|
||||
"version": 6,
|
||||
"refresh": "30s",
|
||||
"time": {
|
||||
"from": "now-6h",
|
||||
@@ -2221,6 +2221,235 @@
|
||||
"values": false
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "row",
|
||||
"title": "Peer occupancy (experiment: lair/quantus#1)",
|
||||
"collapsed": false,
|
||||
"gridPos": {
|
||||
"h": 1,
|
||||
"w": 24,
|
||||
"x": 0,
|
||||
"y": 98
|
||||
},
|
||||
"panels": []
|
||||
},
|
||||
{
|
||||
"type": "timeseries",
|
||||
"title": "Connections held vs cap, by direction",
|
||||
"datasource": {
|
||||
"type": "prometheus",
|
||||
"uid": "prometheus"
|
||||
},
|
||||
"gridPos": {
|
||||
"h": 9,
|
||||
"w": 14,
|
||||
"x": 0,
|
||||
"y": 99
|
||||
},
|
||||
"targets": [
|
||||
{
|
||||
"datasource": {
|
||||
"type": "prometheus",
|
||||
"uid": "prometheus"
|
||||
},
|
||||
"editorMode": "code",
|
||||
"expr": "sum(substrate_sub_libp2p_connections_opened_total{job=\"quantus-node\",direction=\"out\"}) - sum(substrate_sub_libp2p_connections_closed_total{job=\"quantus-node\",direction=\"out\"})",
|
||||
"legendFormat": "outbound held",
|
||||
"range": true,
|
||||
"refId": "A"
|
||||
},
|
||||
{
|
||||
"datasource": {
|
||||
"type": "prometheus",
|
||||
"uid": "prometheus"
|
||||
},
|
||||
"editorMode": "code",
|
||||
"expr": "sum(substrate_sub_libp2p_connections_opened_total{job=\"quantus-node\",direction=\"in\"}) - sum(substrate_sub_libp2p_connections_closed_total{job=\"quantus-node\",direction=\"in\"})",
|
||||
"legendFormat": "inbound held",
|
||||
"range": true,
|
||||
"refId": "B"
|
||||
},
|
||||
{
|
||||
"datasource": {
|
||||
"type": "prometheus",
|
||||
"uid": "prometheus"
|
||||
},
|
||||
"editorMode": "code",
|
||||
"expr": "8",
|
||||
"legendFormat": "out cap (default)",
|
||||
"range": true,
|
||||
"refId": "C"
|
||||
},
|
||||
{
|
||||
"datasource": {
|
||||
"type": "prometheus",
|
||||
"uid": "prometheus"
|
||||
},
|
||||
"editorMode": "code",
|
||||
"expr": "32",
|
||||
"legendFormat": "in cap (default)",
|
||||
"range": true,
|
||||
"refId": "D"
|
||||
}
|
||||
],
|
||||
"description": "opened MINUS closed, which is occupancy. Cumulative opens measure churn and mislead: inbound churns constantly while outbound is deliberately maintained, so a high inbound open count says nothing about whether its cap is full. Outbound sitting on its cap while ~300 peers are discovered is the case for raising --out-peers (issue #1).",
|
||||
"fieldConfig": {
|
||||
"defaults": {
|
||||
"custom": {
|
||||
"lineWidth": 2,
|
||||
"fillOpacity": 6,
|
||||
"showPoints": "never"
|
||||
},
|
||||
"color": {
|
||||
"mode": "palette-classic"
|
||||
}
|
||||
},
|
||||
"overrides": [
|
||||
{
|
||||
"matcher": {
|
||||
"id": "byRegexp",
|
||||
"options": ".*cap.*"
|
||||
},
|
||||
"properties": [
|
||||
{
|
||||
"id": "custom.lineStyle",
|
||||
"value": {
|
||||
"dash": [
|
||||
8,
|
||||
8
|
||||
],
|
||||
"fill": "dash"
|
||||
}
|
||||
},
|
||||
{
|
||||
"id": "custom.fillOpacity",
|
||||
"value": 0
|
||||
}
|
||||
]
|
||||
}
|
||||
]
|
||||
},
|
||||
"options": {
|
||||
"legend": {
|
||||
"displayMode": "list",
|
||||
"placement": "bottom",
|
||||
"showLegend": true
|
||||
},
|
||||
"tooltip": {
|
||||
"mode": "multi",
|
||||
"sort": "desc"
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "stat",
|
||||
"title": "Peers discovered",
|
||||
"datasource": {
|
||||
"type": "prometheus",
|
||||
"uid": "prometheus"
|
||||
},
|
||||
"gridPos": {
|
||||
"h": 4,
|
||||
"w": 10,
|
||||
"x": 14,
|
||||
"y": 99
|
||||
},
|
||||
"targets": [
|
||||
{
|
||||
"datasource": {
|
||||
"type": "prometheus",
|
||||
"uid": "prometheus"
|
||||
},
|
||||
"editorMode": "code",
|
||||
"expr": "substrate_sub_libp2p_peerset_num_discovered{job=\"quantus-node\"}",
|
||||
"legendFormat": "discovered",
|
||||
"range": true,
|
||||
"refId": "A"
|
||||
}
|
||||
],
|
||||
"description": "Peers the node knows about. The gap between this and outbound-held is the headroom --out-peers is refusing to use.",
|
||||
"fieldConfig": {
|
||||
"defaults": {
|
||||
"color": {
|
||||
"mode": "thresholds"
|
||||
},
|
||||
"thresholds": {
|
||||
"mode": "absolute",
|
||||
"steps": [
|
||||
{
|
||||
"color": "text",
|
||||
"value": null
|
||||
}
|
||||
]
|
||||
}
|
||||
},
|
||||
"overrides": []
|
||||
},
|
||||
"options": {
|
||||
"colorMode": "value",
|
||||
"graphMode": "area",
|
||||
"textMode": "auto",
|
||||
"reduceOptions": {
|
||||
"calcs": [
|
||||
"lastNotNull"
|
||||
],
|
||||
"fields": "",
|
||||
"values": false
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "timeseries",
|
||||
"title": "Mining rounds per minute",
|
||||
"datasource": {
|
||||
"type": "prometheus",
|
||||
"uid": "prometheus"
|
||||
},
|
||||
"gridPos": {
|
||||
"h": 5,
|
||||
"w": 10,
|
||||
"x": 14,
|
||||
"y": 103
|
||||
},
|
||||
"targets": [
|
||||
{
|
||||
"datasource": {
|
||||
"type": "prometheus",
|
||||
"uid": "prometheus"
|
||||
},
|
||||
"editorMode": "code",
|
||||
"expr": "sum(rate(substrate_proposer_block_constructed_count{job=\"quantus-node\"}[5m])) * 60",
|
||||
"legendFormat": "rounds/min",
|
||||
"range": true,
|
||||
"refId": "A"
|
||||
}
|
||||
],
|
||||
"description": "Stale-tip proxy for the experiment: each new tip triggers a fresh proposal, so this should track the network block rate. Below it means tips are arriving late.",
|
||||
"fieldConfig": {
|
||||
"defaults": {
|
||||
"custom": {
|
||||
"lineWidth": 2,
|
||||
"fillOpacity": 8,
|
||||
"showPoints": "never"
|
||||
},
|
||||
"color": {
|
||||
"mode": "palette-classic"
|
||||
}
|
||||
},
|
||||
"overrides": []
|
||||
},
|
||||
"options": {
|
||||
"legend": {
|
||||
"displayMode": "list",
|
||||
"placement": "bottom",
|
||||
"showLegend": true
|
||||
},
|
||||
"tooltip": {
|
||||
"mode": "multi",
|
||||
"sort": "desc"
|
||||
}
|
||||
}
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user