CorefinityCorefinityDocs
Open the Control PanelControl Panel
Deployments · 9 of 65

corestack_deployment_retry

Retry a deployment.

Scope writeRuns as a taskNo secrets returnedCoreCluster
On this page

When to use it

Run a failed deployment again.

Retrying a deployment that is not the site's newest puts its older commit live over the newer one. The tool refuses it with confirm and a message naming both deployments. Show the person that message, and only once they agree, call again with confirm_older: true. Retrying the newest deployment needs no confirmation.

Ask for it

Prompt
Using the Corefinity MCP server, call corestack_deployment_retry with site_id 4211, deployment_id 6610. (Retry a deployment.) Tell me what it returned.

Inputs

confirm_olderboolean

Required, as true, to retry a deployment that is not the site's newest

deployment_idintegerRequired

At least 1.

site_idintegerRequired

At least 1.

Example call

tools/call
{
  "name": "corestack_deployment_retry",
  "arguments": {
    "site_id": 4211,
    "deployment_id": 6610
  }
}

If it is refused

ReasonWhat to do
forbiddenThe person's role lacks the Retry Deployments permission. The message names it and says an owner of the company can grant it in Roles; details.permission is retryDeployments. Nothing is changed.
artifact_prunedThis deployment's archive was removed after 30 days, so it can't be retried. Deploy again from the branch or commit (corestack_deployment_trigger). Archives are kept for 30 days, and always for a site's 10 most recent deployments and its newest successful one.
pipeline_changed_since_buildThis deployment's build finished, and its build lines or the variables the build reads changed after it. A Retry deploys the same built files and would skip the new build: "The pipeline changed after this build. Retry deploys the same built files; use Deploy again to run the new build." Deploy the same commit again instead (corestack_deployment_trigger with its commit as ref), which builds the pipeline as it is now. A change to the database or post-deploy lines alone does not refuse a Retry: those lines run anew.
not_failedOnly a failed deployment can be retried; this one did not fail.
release_not_liveThat release is not live; choose the live release.
confirmThis is not the site's newest deployment, so retrying it puts an older release live. The message names the commit and date of both; ask the person before you send confirm_older: true.
managed_elsewhereAnother control panel manages this site, so use that control panel for it. This is refused before anything else about the site, as in the Control Panel and the API. Nothing was done.
invalid_argumentsThe arguments do not match the tool's input schema, or the Control Panel refused a field. Fix the fields named in the answer and call again.
tenant_mismatchThe object does not exist or the token may not see it. Check the id.
scope_requiredThe token lacks a scope this call needs. Open the link in the answer to add it, then call again.
pin_refusedThe grant is limited to chosen clusters, sites or environments, and this call is outside them. Open the link in the answer to widen it.
rate_limitedToo many calls. Wait for the number of seconds the answer names.
temporarily_unavailableThe Control Panel could not answer just now. Nothing was started; try again shortly.
internal_errorSomething went wrong on our side. Quote the trace id to support.

The same task elsewhere

Was this helpful?
Open in ClaudeOpen in ChatGPT