Full observability across your entire infrastructure
Standing up the infrastructure is only half the job: you also need to see what's happening inside it. We set up a fully open source observability stack, with Prometheus, Grafana, Loki, and Promtail, covering metrics and logs on both bare metal and AWS.
Why monitor with Apsis
Observability built to act before the incident
Real-time metrics
Periodic scraping of CPU, memory, disk, network, and application metrics, stored in a time-series database.
Centralized, searchable logs
Logs from all your applications aggregated automatically, with search and correlation by service, time range, or error pattern.
Unified dashboards
Metrics and logs side by side in one place, with panels designed for each team to see what matters.
Alerts before the incident
Configurable rules via email, Slack, or other channels, notifying your team before an issue turns into an incident.
Cost under control
A fully open source stack, with retention and scrape intervals tuned to keep everything lean, with no noise or billing surprises.
Complete coverage
The same observability stack, whether your infrastructure runs on bare metal, on AWS, or both.
Metrics
Prometheus at the center of metrics collection
Prometheus periodically scrapes endpoints exposed by your applications and by the infrastructure itself, storing everything in a time-series database and letting you configure alerts based on rules.
Periodic scraping: automatic collection of CPU, memory, disk, and network from every server or cluster.
Application metrics: latency, error rate, and throughput straight from your code.
Time series: queryable history to investigate trends and spikes.
Rule-based alerts: configurable triggers based on how metrics actually behave.
Logs
Promtail and Loki, logs without the high cost
Promtail runs alongside your applications and automatically ships logs to Loki, which indexes them in a lightweight, low-cost way, since it only indexes metadata instead of the full content of every line.
Automatic collection: Promtail follows your applications with no manual setup.
Lightweight indexing: Loki indexes metadata, not full content, cutting storage cost.
Search and correlation: find logs by service, time range, or error pattern in seconds.
Tuned retention: retention periods designed to keep the stack relevant, without noise.
Dashboards & Alerts
Grafana, unifying metrics and logs
Grafana brings everything together in one place: dashboards with Prometheus metrics side by side with Loki logs, plus configurable alerts that notify your team before an issue turns into an incident.
Dashboards per team: panels designed for infrastructure, application, or business needs.
Metrics and logs together: correlate what happened with why, without switching tools.
Configurable alerts: email, Slack, or other channels when something goes out of range.
Fine-tuned: no panels or alerts that turn into daily distractions.
Use cases
What we've already solved with this stack
The same observability stack covers everything from infrastructure health to how your applications behave in production.
Infrastructure Observability
Full visibility into CPU, memory, disk, and network for every server or cluster.
Infrastructure dashboards
CPU, memory, disk, and network per server or cluster, in real time.
Configurable alerts
Via email, Slack, or other channels when metrics go out of expected range.
Frequently asked questions
Why do I need monitoring beyond just standing up the infrastructure?
Without observability, performance problems, error spikes, and capacity limits only surface once your customers have already felt the impact. Monitoring is what lets you act before the incident, not after.
How does metrics collection with Prometheus work?
Prometheus periodically scrapes endpoints exposed by your applications and by the infrastructure itself, storing everything in a time-series database and letting you configure alerts based on rules.
How do you keep logs from blowing up the storage budget?
We use Loki, which only indexes metadata instead of the full content of every log line, cutting storage costs significantly. We also tune retention periods to keep the stack lean and relevant.
Does monitoring work on both bare metal and AWS?
Yes. It's the same observability stack regardless of where your infrastructure runs, which makes it easy to compare and correlate metrics even across hybrid environments.