You are browsing the docs for the snapshot version of Nexus, the latest release is available here
Nexus Delta
Running and monitoring
CPU: It suggests heavy indexing and read operations. It is likely to happen when reindexing a large amount of views at the same time but if it happens regularly:
- Review which indexing process is going on by querying the
projection_offsetsandcomposite_offsetstable - Allocate more resources to Delta
Memory and garbage collection: Delta will use the available RAM in 2 ways, JVM heap and the file system cache so like Elasticsearch, the JVM garbage collection frequency and duration are also important to monitor.
JVM and host metrics (including those related to the heap) are collected by OpenTelemetry
Storage: The different types of storage (local, S3) need to be monitored for availability/performance/utilization
Metrics: Nexus Delta relies on OpenTelemetry to also collect metrics about the execution of most operations (fetching a resource, query Blazegraph, etc…)
For more details, see the monitoring section
Logs: Nexus Delta relies on Logback for logs which one of the popular logging frameworks on the JVM.
Logback provides:
- Reloading the configuration while the application is running
- Control the output of the logs, opting for JSON helps for the integration with Filebeats and Elasticsearch for log aggregation
An example of logback configuration writing logs as json in files is available here