Notes on observability, platform engineering and running your own infrastructure.https://dusansusic.com/enhttps://dusansusic.com/writing/session-10-lessons-learned/https://dusansusic.com/writing/session-10-lessons-learned/Ten sessions in: what I would do the same, what I would do differently, and whether it was worth it.Tue, 19 May 2026 00:00:00 GMTretrospectiveobservabilitylessonshttps://dusansusic.com/writing/session-09-the-cost-breakdown/https://dusansusic.com/writing/session-09-the-cost-breakdown/What the self-hosted stack costs per month, line by line, against the bill it replaced.Thu, 14 May 2026 00:00:00 GMTcostfinopsawshttps://dusansusic.com/writing/session-08-migrating-monitors/https://dusansusic.com/writing/session-08-migrating-monitors/Porting hundreds of alerts into a new query language, and the remote-write we forgot to move.Tue, 12 May 2026 00:00:00 GMTalertingmigrationgrafanahttps://dusansusic.com/writing/session-07-gitops-deployment-model/https://dusansusic.com/writing/session-07-gitops-deployment-model/Terraform modules, configuration in Git, automated sync - running an observability stack you can rebuild from scratch.Thu, 07 May 2026 00:00:00 GMTgitopsterraformautomationhttps://dusansusic.com/writing/session-06-otel-instrumentation/https://dusansusic.com/writing/session-06-otel-instrumentation/Wiring the OpenTelemetry SDK into real services: resource attributes, batch processors, and the settings that actually matter.Tue, 05 May 2026 00:00:00 GMTopentelemetryinstrumentationpythonhttps://dusansusic.com/writing/session-05-metrics-mimir-over-prometheus/https://dusansusic.com/writing/session-05-metrics-mimir-over-prometheus/Prometheus scrapes well but stores poorly. Why Mimir's S3 block storage replaced a fleet of unreplicated Prometheus instances.Thu, 30 Apr 2026 00:00:00 GMTmetricsmimirprometheushttps://dusansusic.com/writing/session-04-traces-and-span-filtering/https://dusansusic.com/writing/session-04-traces-and-span-filtering/Getting trace volume under control with tail sampling and span filtering, without losing the traces worth keeping.Tue, 28 Apr 2026 00:00:00 GMTtracestemposamplinghttps://dusansusic.com/writing/session-03-logs-from-aws-cloudwatch-to-loki/https://dusansusic.com/writing/session-03-logs-from-aws-cloudwatch-to-loki/Moving logs off CloudWatch onto Loki: label design, chunk storage on S3, and what queries cost once you own them.Thu, 23 Apr 2026 00:00:00 GMTlogslokiawscosthttps://dusansusic.com/writing/session-02-the-architecture/https://dusansusic.com/writing/session-02-the-architecture/Why we ran the LGTM stack on EC2 with dedicated node pools per signal, and how we designed out the single points of failure.Tue, 21 Apr 2026 00:00:00 GMTarchitectureawslgtmhigh-availabilityhttps://dusansusic.com/writing/session-01-opentelemetry-changed-the-game/https://dusansusic.com/writing/session-01-opentelemetry-changed-the-game/Why vendor-neutral instrumentation turned the monitoring backend from a marriage into a decision we can revisit.Thu, 16 Apr 2026 00:00:00 GMTopentelemetryobservabilityvendor-lock-in