HealthCheck

This is a command-line tool to check the health of a Redis Enterprise cluster.

Click here for more information.

Description

The tool does read-only operations via remote execution (e.g. SSH) and the Redis Enterprise REST-API.

There are 3 categories of checks:

Each check has a result which is indicated by:

[+] check succeeded: The conditions were met.
[-] check failed: The conditions were not met.
[~] check with no result: There were no conditions, this check only output some values.
[*] check had an error: The check could not be executed due to an error.
[ ] check was skipped: The check was omitted because it did not apply in the given context.

Checks are grouped into check suites, currently there are 3 check suites available:

Checks may or may not have parameter maps, i.e. JSON files with parameters.

There are parameter maps given, but you can provide your own.
To provide your own parameters, clone or edit a paramter map in parameter_maps/<SUITE>/<CHECK>/.
Alternatively you can pass the full filename (i.e. with .json at the end) and it will look it up in the current directory.

Fill in the config.ini file with the following configuration data:
- Under a section called http, HTTP access to the REST-API of a Redis Enterprise cluster:
  - Address of the cluster, i.e. FQDN
  - Username of the cluster
  - Password of the cluster
- Under a section called ssh, SSH access to all nodes of the Redis Enterprise cluster:
  - SSH username
  - CSV list of hostnames
  - Path to SSH private key file
- Alternatively to SSH:
  - Under a section called docker, a CSV list of Docker containers (name or ID) can be specified.
  - Under a section called k8s, a CSV list of Kubernetes pods and a namespace can be specified.
- Under a section called renderer, a renderer module name can be specified. Options are:
  - basic The default renderer.
  - json Renders result in JSON format.
  - html Renders result in HTML format.
  - syslog Renders results according to RFC5425 w/o structured data elements.
Alternatively to config.ini you can pass a different configuration filename with -cfg <CONFIG>.
Don't forget to make hc executable, e.g. chmod u+x hc.

To run all check suites, execute ./hc.
To see a list of all checks, execute ./hc -l.
To run one check suite, execute ./hc -s <SUITE>, e.g.
- execute ./hc -s node for node checks.
To run all checks from a category, execute ./hc -c <CATEGORY>, e.g.
- execute /hc -c status for status checks.
To run single checks, execute ./hc -c <CHECK>, e.g.
- execute ./hc -c "network link" to get the network link between all nodes.
- execute ./hc -c DS-001,DS-002 to execute 1st and 2nd database status check.
If a check supports a parameter map, execute ./hc -c <CHECK> -p <PARAMS>, e.g.
- execute ./hc -c "cluster sizing" -p reco for cluster sizing check with recommended HW requirements.
- execute ./hc -c "database config" -p config1 for database configuration check with parameter map config1.
- execute ./hc -c "database config" -p my_config.json for database configuration check with parameters given in my_config.json from the current directory.
For a quick help, execute ./hc -h.

Run Docker container, e.g. docker run -v $(pwd)/myconfig.ini:/myconfig.ini redislabs/healthcheck:latest -cfg /myconfig.ini.

The script exits with the following return code: