Catch a failing service before your users do.

Group the entities and log sources you operate into services, define their health as KPIs you write in LPQL, map how they depend on one another, and let LogPulse watch each KPI for anomalies, all on the same engine as search.

Your services, with a pulse.

Group your logs into the services you actually run, define health in LPQL, and let LogPulse watch every KPI for the drift that turns into downtime.

Service health

Define “healthy” in your own terms.

Write each KPI in LPQL (error rate, p95 latency, queue depth) and LogPulse rolls them into one health state per service, on the same engine as search.

checkout-svc · healthLPQL
Error rate0.4%
p95 latency180 ms
Queue depth1.2k
Service health98 · healthy
Dependencies

See what breaks when one thing does.

Map how services depend on one another, then trace a degradation to its source and to everything downstream it puts at risk.

Dependenciesblast radius
checkout-svc
payments-svc
orders-svc
auth-svc
payments-svc degraded1 path at risk
Anomaly detection

Know it's drifting before a ticket lands.

LogPulse learns each KPI's normal rhythm and flags the deviation, so a slow leak shows up as an anomaly, not an outage.

p95 latency · orders-svc
Anomaly · 14:02+312% vs baseline

Service-level observability, no extra pipeline

KPIs are just LPQL searches, so there's no separate metrics store to feed: health comes straight from the logs you already ship.

Services from your fleet

Assemble a service from the hosts, containers, and log sources behind it. Scope by entity labels or by log source; membership resolves automatically and stays current.

KPIs from LPQL

Turn any LPQL search into a health signal. A chosen field becomes the metric, and thresholds map it to a severity over 1h to 7d windows.

Rolled-up health

Each service gets one status rolled up from its KPIs, with estate-wide counts of healthy, warning, and critical so you know where to look first.

Dependency mapping

Record upstream and downstream relationships and view them as a graph, so a degraded service shows what it relies on and what relies on it.

KPI anomaly detection

Every KPI is baselined for daily and weekly seasonality and watched for drift, catching trouble before it breaches a static threshold.

Entity 360

Every service member has a unified view that combines its observability signals with its security risk, reachable from both the SRE and analyst sides.

See your services' health today

Group your first service and turn an LPQL search into a health signal in minutes.

Start free
100 MB/day · 7-day retention · No credit card

We use cookies to analyze site traffic and improve your experience. No cookies are placed without your consent. Privacy Policy