Dell EMC Failover

Dell EMC Failover

Certified
Enterprise Edition

Fail over a Dell EMC PowerStore replication session

Fails over a replication session via POST /api/rest/replication_session/{id}/failover, promoting the destination volume(s) to read-write. PowerStore typically accepts this asynchronously (HTTP 202 with a job reference); see wait and waitTimeout.

yaml
type: io.kestra.plugin.ee.dellemc.powerstore.replication.Failover

Fail over a replication session to the remote site.

yaml
id: powerstore_failover
namespace: company.team

tasks:
  - id: failover
    type: io.kestra.plugin.ee.dellemc.powerstore.replication.Failover
    host: "{{ secret('POWERSTORE_HOST') }}"
    username: "{{ secret('POWERSTORE_USERNAME') }}"
    password: "{{ secret('POWERSTORE_PASSWORD') }}"
    sessionId: "{{ inputs.session_id }}"
Properties

PowerStore management address

Hostname or IP address of the PowerStore appliance management interface, optionally including a port if non-default (for example: powerstore.example.com or 192.168.1.20: 8443). The REST API is reached at https://{host}/api/rest.

PowerStore password

Password for username. Store this value as a Kestra secret and reference it via {{ secret('POWERSTORE_PASSWORD') }}.

Replication session ID

ID of the PowerStore replication session to fail over.

PowerStore username

Username of a PowerStore local or LDAP account used to authenticate via HTTP Basic Auth.

Reference (ref) of the pluginDefaults to apply to this task.

Defaultfalse

Skip TLS certificate verification

When true, accepts the PowerStore appliance's TLS certificate without validation. Defaults to false. PowerStore management interfaces are commonly deployed with a self-signed certificate; set this to true to connect to those without importing the certificate into a trust store. Intended for lab/non-production use — leave false in production and use a properly signed certificate instead.

Defaulttrue

Wait for completion

When the PowerStore array accepts the request asynchronously (HTTP 202 with a job reference), whether to poll the job until it completes before returning. Defaults to true so downstream tasks receive a usable result. When false, the task returns immediately with the job ID once accepted, without waiting for it to finish. Polling blocks the worker thread running this task for as long as the wait lasts (up to waitTimeout); size the worker pool accordingly if many job-aware tasks run concurrently, or set this to false and poll separately.

DefaultPT10M

Job wait timeout

Maximum time to wait for an asynchronously accepted job to complete when wait is true. Defaults to PT10M. Capped at PT30M: values above that are clamped down to PT30M (with a warning logged) to avoid tying up a worker thread indefinitely. If the timeout elapses before the job completes, the task returns successfully without confirmed completion — the job ID is still available in the output so its status can be checked later. Killing the execution while this task is waiting stops it from polling immediately, but PowerStore's job REST resource has no cancel/abort endpoint: the array-side operation itself (resize, failover, synchronize, restore) keeps running to completion on the appliance regardless.

Whether the failover is still in progress

True when the job did not confirm completion within waitTimeout, or wait was false.

Session state after failover

For example: Failed_Over. Null if still pending.