Monitoring etcd
The etcd sensor is automatically deployed and installed after you install the Instana agent.
Supported Versions
Etcd sensor supports etcd API 3.3.x and 3.4.x with /v3 endpoints.
Configuration
If the etcd node is configured with SSL, the agent requires the CA certificate, client certificate and key to collect metrics. The agent uses the trusted-ca-file argument value (directory) as a location to find
these files, unless overridden in the agent configuration file:
com.instana.plugin.etcd:
caCertificate: '' # path to CA certificate (eg. /etc/etcd/ca.crt)
clientCertificate: '' # path to client certificate (eg. /etc/etcd/client.crt)
clientKey: '' # path to client key (eg. /etc/etcd/client.key)
If the etcd instance is running in a container, these paths point to files in the container's filesystem.
When the agent starts and discovers an etcd that has no certificates, it logs the following message:
Cannot get content from https://192.168.155.186:2379/v2/stats/self.
Received fatal alert: bad_certificate.
Metrics collection
To view the metrics, select Infrastructure in the sidebar of the Instana User interface, click a specific monitored host, and then you can see a host dashboard with all the collected metrics and monitored processes.
Configuration data
- ID
- Name
- Leader ID
- State
- Cluster Version
- Server Version
- Start time
Performance metrics for etcd API 2.x
| Metric | Description | Granularity |
|---|---|---|
| Received requests | The number of received requests. | 1 second |
| Sent requests | The number of sent requests. | 1 second |
| Received traffic | Incoming traffic in bytes. | 1 second |
| Sent traffic | Outgoing traffic in bytes. | 1 second |
| Expire count | The number of expired keys per second. | 1 second |
| Watchers | The number of watchers. | 1 second |
| Operations (Success vs Fail) | The number of compare and delete, compare and swap, create, delete, gets, sets, update operations. | 1 second |
Health signatures for etcd API 2.x
For each sensor, there is a curated knowledgebase of health signatures that are evaluated continuously against the incoming metrics and are used to raise issues or incidents depending on user impact.
Performance metrics for etcd API 3.x
| Metric | Description | Granularity |
|---|---|---|
| Received traffic | Incoming traffic in bytes. | 1 second |
| Sent traffic | Outgoing traffic in bytes. | 1 second |
| Expire count | The number of expired keys per second. | 1 second |
| Watchers | The number of watchers. | 1 second |
| Operations (Success versus Fail) | The number of operations, such as compare and delete, compare and swap, create, delete, gets, sets, and update. | 1 second |
Built-in events trigger issues or incidents based on failing health signatures on entities, and custom events trigger issues or incidents based on the thresholds of an individual metric of any given entity.
For information about built-events for the etcd sensor, see the Built-in events reference.