#!/bin/bash
# Forgejo Runner healthcheck — managed by Ansible (roles/forgejo_runner)
#
# Answers "is forgejo-runner healthy" and records it two ways: this log, and the
# exit code. The exit code is the durable artefact — systemd keeps it, so
#     systemctl is-failed {{ healthcheck_service_name }}.service
# answers the question with no monitoring system involved.
#
# Reporting is optional and generic: if a push URL is configured it also pings
# it. Nothing here knows or cares which monitoring product is on the other end.

LOG_FILE="{{ healthcheck_log_file }}"
PUSH_URL="{{ healthcheck_push_url }}"

log_message() {
    echo "$(date '+%Y-%m-%d %H:%M:%S') - $1" >> "$LOG_FILE"
}

main() {
    if ! systemctl is-active --quiet forgejo-runner; then
        log_message "ERROR: forgejo-runner is not active"
        exit 1
    fi

    if [ -z "$PUSH_URL" ]; then
        # Healthy, and nothing to report to. Not an error: the exit code below
        # is still a complete answer for anything reading unit state.
        log_message "forgejo-runner is active (no push URL configured)"
        exit 0
    fi

    log_message "forgejo-runner is active, sending ping"
    response=$(curl -s -w "\n%{http_code}" "$PUSH_URL?status=up&msg=forgejo-runner%20is%20active" 2>&1)
    http_code=$(echo "$response" | tail -n1)
    if [ "$http_code" = "200" ] || [ "$http_code" = "201" ]; then
        log_message "Ping sent successfully (HTTP $http_code)"
    else
        log_message "ERROR: Failed to send ping (HTTP $http_code)"
        exit 1
    fi
}

main