Monitoring cbft

cbft provides the following monitoring features:

Logs

The web admin UI Logs screen shows the most recent 1000 log messages for the cbft node. It is a convenience feature so that an administrator does not have to log into the machine to see recent logs.

Note: the Logs screen does not show cluster-wide logs, but only the logs from the cbft node that the web browser is pointed at.

Recommended practice: an administrator should consider capturing cbft's stdout/stderr output to rotated files or to a centralized log service.

Node monitoring

The web admin UI of cbft provides a Monitor screen that shows node-related memory and GC (garbage collection utlization).

Node index monitoring

The web admin UI provides a dropdown on the Monitor screen to control graphs that can allow an administrator to focus in on the index partition performance of the current cbft node.

Index partition specific graphs include:

Updates (count of updates)
Deletes (count of deletes)
Analysis/Index Time
Searches (count of searches)
Search Time

The web admin UI also includes more numeric, per index monitoring statistics. Using a web browser...

Navigate to the Indexes screen.
Click on the index name link of the index you're interested in.
Click on the Monitor sub-tab to see per-index specific stats and counters.

The displayed stats and counters are only for the current cbft node.

The Refresh button will request the latest information from the current cbft node.

Selecting the Show all stats checkbox will display even more stats and counters.

By default, aggregated stats across index partitions for the current cbft node are shown.

Selecting the Show details and timings for each index partition checkbox will display non-aggregated details from every index partition from the current cbft node.

Memory

TBD

GC pauses

TBD

CPU

TBD

Storage

TBD

Networking

TBD

Performance

TBD