Command line options

Use --help to get the full usage info:

$ escalator --help
usage: escalator --nodegroups=NODEGROUPS [<flags>]

Flags:
      --help                   Show context-sensitive help (also try --help-long and --help-man).
  -v, --loglevel=4             Logging level passed into logrus. 4 for info, 5 for debug.
      --logfmt=ascii           Set the format of logging output. (json, ascii)
      --address=":8080"        Address to listen to for /metrics
      --scaninterval=60s       How often cluster is reevaluated for scale up or down
      --kubeconfig=KUBECONFIG  Kubeconfig file location
      --nodegroups=NODEGROUPS  Config file for nodegroups
      --drymode                master drymode argument. If true, forces drymode on all nodegroups
      --cloud-provider=aws     Cloud provider to use. Available options: (aws)
      --aws-assume-role-arn=AWS-ASSUME-ROLE-ARN
                               AWS role arn to assume. Only usable when using the aws cloud provider. Example: arn:aws:iam::111111111111:role/escalator
      --leader-elect           Enable leader election
      --leader-elect-lease-duration=15s
                               Leader election lease duration
      --leader-elect-renew-deadline=10s
                               Leader election renew deadline
      --leader-elect-retry-period=2s
                               Leader election retry period
      --leader-elect-config-namespace="kube-system"
                               Leader election lease object namespace
      --leader-elect-config-name="escalator-leader-elect"
                               Leader election lease object name

Options

-v, --loglevel

Determines the log level for Escalator. logrus is being used to handle log format and logging levels. You can see the logrus logging levels here.

In some situations it may be helpful to disable debug logging as it can be quite verbose.

Examples:

• -v 5 show all log levels, including debug
• -v 4 show info log level and above
• -v 1 only show fatal log level and above

--logfmt

Defines the log format

ascii

INFO[0000] Starting with log level info                 
INFO[0000] Validating options: [PASS]                    nodegroup=shared
INFO[0000] Registered with drymode false                 nodegroup=shared
INFO[0000] Using in cluster config          

json

{"level":"info","msg":"Starting with log level debug","time":"2018-03-09T16:53:33+11:00"}
{"level":"info","msg":"Validating options: [PASS]","nodegroup":"shared","time":"2018-03-09T16:53:33+11:00"}
{"level":"info","msg":"Registered with drymode false","nodegroup":"shared","time":"2018-03-09T16:53:33+11:00"}
{"level":"info","msg":"Using in cluster config","time":"2018-03-09T16:53:33+11:00"}

--address

Address to listen on for /metrics and /healthz. Must be in a format that http.ListenAndServe can interpret.

--scaninterval

How often to perform a scan or run. It is recommended to have this configured between 30 seconds to 60 seconds. Too long of a scan interval can lead to Escalator reacting too slow to scaling up the cluster. Too short of a scan interval can lead to to Escalator scaling too quickly and imprecisely.

--kubeconfig

The path to the config that client-go uses for connecting to Kubernetes. Note: this isn't required when running Escalator inside the cluster as Escalator will get it's credentials from the Kubernetes environment variables.

--nodegroups

The path to the nodegroups yaml config file that defines the node groups and options. Full nodegroups configuration can be found here.

--drymode

Master drymode flag to force "dry mode" on all node groups. Dry mode will log the actions that Escalator will perform without actually running them.

--cloud-provider

The cloud provider to use. Cloud provider configuration can be found here.

--aws-assume-role-arn

Provides an option to specify an AWS IAM role to assume when Escalator starts. Only works with AWS Cloud Provider.

--leader-elect

Enable leader election behaviour. Note that Escalator uses a ConfigMap for the leader lock, not an Endpoint.

--leader-elect-lease-duration

Sets how long a nonleader will wait before it attempts to require the leadership. Measured against time of last observed ack.

--leader-elect-renew-deadline

Sets how long an acting leader will retry refreshing leadership before giving up.

--leader-elect-retry-period

Sets how long all the clients will wait in between attempts of any action.

--leader-elect-config-namespace

Sets the namespace where the lease object used for locking will be created or looked for.

--leader-elect-config-name

Sets the name of the lease object used for locking.