--- a/rtr-lax-01.conf+++ b/rtr-lax-01.conf@@ -1,34 +1,36 @@see every change.
know every cause.
last 24 hours
One cell per drift event across the fleet. Colour is severity. Hover for the device.
situation-driven ops
not alerts. situations.
Every signal across config drift, SaaS probes, and device telemetry is correlated into one Situation — with impact, root cause, and the next move. Stop triaging twelve alerts for one outage.
Slack degraded for NYC branch
detected
11:38 · 14m ago
users
47
sites
3
duration
14m
ISP routing change at 11:42 — packets traversing congested transit
↳ BGP route flap detected on edge-nyc-01
anomalies · 3 signals
path
Failover to backup ISP — estimated 30s impact
fleet overview
one pane. every device. every site.
KPI strip, client health cards, device fleet table. See what's up, what's drifted, and what needs attention — across every customer network at once.
saas monitoring
watch every saas your customers depend on.
End-to-end probes run from every collector — measured from your customer's network, not ours. Microsoft 365, Google Workspace, Salesforce, Slack: catch the outage before the tickets land.
slack — degraded across 3 of 4 sitesdrift analysis
read the summary, not the diff.
Each drift event lands with severity, root cause, blast radius, and one remediation step. Written per event, in plain English.
neighbor 10.0.1.1 removed and 10.0.1.5 added in the same commit. traffic will blackhole during reconvergence. affected prefixes: 203.0.113.0/24, 198.51.0.0/22.
→ revert the neighbor change or add the new peer before removing the old one.
the 10.42.0.0/16 static has been replaced with a less-specific 10.0.0.0/8 route. any traffic for 10.42.0.0/16 will follow the new summary.
→ verify that 10.1.1.1 can reach all /16 subnets, or restore the original specific route.
fair-queue + random-detect dscp-based added to the WAN-OUT policy. this improves congestion behaviour without affecting existing VOIP priority class.
→ no action required. annotated for change-log hygiene.
on-prem collectors
deploy once. poll forever.
A lightweight Docker container on the customer network. SSH into every device, encrypt configs with per-site Fernet keys, ship to the cloud. Heartbeats every 60 seconds.
client health
every client. at a glance.
Per-client health cards across your entire book of business. Device status, drift velocity, and time-since-last-change — the monthly review meeting in a single scroll.