Project archived, due to the number of ad hoc features/behavior needed for our custom requirements. So it is no longer worth putting it as open source. The project is still working.
This project contains a probe aiming to monitor latencies and avaibilities for Couchbase and Memcached nodes.
The targeted clusters can either be discovered through Consul, or via a node (other nodes and buckets would then me discoverred automatically)
Metrics are gathered periodically and exposed through a prometheus endpoint available at Config:app:httpServerPort
/metrics
Metric | Description | Labels |
---|---|---|
memcached_up | Availability of a given bucket on a given memcached node | cluster, bucket, instance |
memcached_latency | Get and Set latencies | cluster, bucket, instance, command, quantile |
memcached_stats | Memcached global stats (ex: total_items, get_misses,..etc.) | cluster, bucket, instance, name |
memcached_items | Memcached slab specific metrics (ex: moves_to_warm, evicted,..etc.) | cluster, bucket, instance, slabsizerange, name |
couchbase_up | Availability of a given bucket on a given Couchbase node | cluster, bucket, instance |
couchbase_latency | Disk persistence latency for a set operation | cluster, bucket, instance, command, quantile |
couchbase_operations | Ongoing rebalance operation on the cluster | cluster, operation |
couchbase_stats | Couchbase specific stats that are listed in the config file in the couchbaseStats/bucket section |
cluster, bucket, instance, name |
couchbase_xdcr | Couchbase replication stats that are listed in the config file in the couchbaseStats/xdcr section |
cluster, bucket, remotecluster, name |
couchbase_membership | Cluster membership status | cluster, membership |
- Clone the repository
- Ensure you have Java 8 and gradle installed
- Run
gradle build
at the root of the directory - Artifacts can be founds inside the
build/libs
directory
- Ensure you have a valid config file (see configuration)
- Run
java -jar my_uber_jar.jar my_config.yml
- That's all, Enjoy !!