DevOpsArk + Loki
Read existing log streams, add pattern analysis
Vendor: Grafana Labs
What does the DevOpsArk Loki integration do?
The DevOpsArk Loki integration reads log streams from existing Loki deployments, applies pattern extraction and novelty detection, and correlates the results with metrics, traces and deployments.
What the Loki integration provides
- Query existing Loki instances across clusters through one interface.
- Apply pattern extraction so log volume becomes a readable list of recurring events.
- Detect new and unusually frequent patterns without hand-written rules.
- Correlate log patterns with metrics, traces and deployment events.
- Retain selected log classes beyond Loki retention for audit purposes.
Exactly what is read, and what can be written
Read and write are separate grants. Nothing in the write column happens unless you explicitly allow it.
- Log streams via the Loki HTTP API
- Stream labels and cardinality
- Retention configuration
- No writes to Loki. Ingestion configuration remains yours.
Connecting Loki
- 1Connect Loki
Provide the Loki endpoint and credentials, in-cluster or external.
- 2Map labels
Map stream labels to the DevOpsArk service model.
- 3Enable analysis
Turn on pattern extraction and novelty detection for the streams you care about.
Loki integration: frequently asked questions
It can collect logs directly, or read from an existing Loki deployment. Teams that have already invested in Loki generally keep it and use DevOpsArk for cross-cluster search, pattern analysis and correlation.
LogQL answers questions you know to ask. Pattern extraction surfaces the log line that appeared for the first time this morning, which nobody would have queried for.
Yes. Selected classes (typically audit-relevant logs) can be retained in DevOpsArk on their own schedule independent of Loki retention.
Integrations that commonly go with this one
Connect Loki and see your own data
Read-only connection takes minutes. Most demos start with the customer estate rather than a sample one.