fix: handle suspended Power Automate flow with retry logic and actionable error

[Copilot]

The send-to-power-automate job was failing because the Power Automate flow trigger was in a Suspended state, causing the webhook to return HTTP 400 (WorkflowTriggerIsNotEnabled). The workflow had no retry logic and no error parsing, so it failed immediately with a generic message.

Changes

• Retry logic: Up to 3 attempts with exponential backoff (5s → 10s → 20s) before giving up, covering transient HTTP failures
• Non-retryable error detection: Parses the JSON response to detect WorkflowTriggerIsNotEnabled and exits immediately with an actionable message instead of retrying uselessly:

  Power Automate flow is suspended and must be re-enabled before it can accept requests. Go to Power Automate, find the flow, and turn it on.

• Improved final error: On exhaustion, reports the last HTTP code and attempt count

ERROR_CODE=$(echo "$RESP_BODY" | jq -r '.error.code // empty' 2>/dev/null || true)
if [ "$ERROR_CODE" = "WorkflowTriggerIsNotEnabled" ]; then
  echo "::error::Power Automate flow is suspended and must be re-enabled..."
  exit 1
fi

The Power Automate flow itself still needs to be re-enabled in the portal — the suspended trigger is a configuration issue on the Power Automate side that must be resolved there.

Original prompt

Fix the failing GitHub Actions workflow send-to-power-automate
Analyze the workflow logs, identify the root cause of the failure, and implement a fix.
Job ID: 65542890559
Job URL: https://github.com/golclinics/discussions/actions/runs/22620048385/job/65542890559

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.