7d44c38c90
This commit introduces dccertmon, a new managed service for DC certificate auditing and management. Currently, platform cert management, DC cert management, and subcloud cert auditing are coupled into a single platform service (certmon). To meet the requirements of DC scalability and portability, DC specific functionality must be decoupled. These changes lay the groundwork for the new service, by: - Creating the necessary service files. - Introducing configs for the service. - Declaring high level methods (Skeleton - lifecycle and manager) DC-specific functionality will be migrated to this dccertmon service and optimized in subsequent changes. Non-DC cert management will continue to be handled by certmon. Overall, this commit introduces: - The OCF file necessary for high availability management of the dccertmon service by SM. - Package configurations to build the service (Package: distributedcloud-dccertmon). - Lifecycle manager for a running DC cert monitor service. - Skeleton/base service application logic - CertificateMonitorManager. - RPC notification handlers for subcloud online/managed. - Configuration for the log folders and log rotation. The logs will be available in /var/log/dccertmon/dccertmon.log. These changes are part of a set of commits to introduce the dccertmon service: [1] https://review.opendev.org/c/starlingx/ha/+/941205 [2] https://review.opendev.org/c/starlingx/stx-puppet/+/941208 Test Plan: - PASS: Build dccertmon package - PASS: Install and bootstrap system with custom ISO containing the newly created dccertmon package - PASS: Verify that the dccertmon.service is loaded - PASS: Verify dccertmon is being properly logged to the correct folder. - PASS: Check logged messages and verify execution of - Cert Watcher thread - Task Executor (Audit thread) - Periodic tasks running at expected intervals - PASS: Configure and provision the service using SM and verify it has correctly started and can be restarted with 'sm-restart'. - PASS: Tox checks running on dccertmon Note: This commit has been tested alongside the related changes and their respective test plans. [1][2] Story: 2011311 Task: 51663 Change-Id: Ic23d8d13e4b292cf0508d23eaae99b8e07f36d31 Signed-off-by: Salman Rana <salman.rana@windriver.com>
27 lines
659 B
Python
27 lines
659 B
Python
#
|
|
# Copyright (c) 2025 Wind River Systems, Inc.
|
|
#
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
#
|
|
|
|
# TODO(srana): Refactor DC role usage due to deprecation.
|
|
DC_ROLE_UNDETECTED = "unknown"
|
|
DC_ROLE_SUBCLOUD = "subcloud"
|
|
DC_ROLE_SYSTEMCONTROLLER = "systemcontroller"
|
|
|
|
DC_ROLE_TIMEOUT_SECONDS = 180
|
|
DC_ROLE_DELAY_SECONDS = 5
|
|
|
|
INVALID_SUBCLOUD_AUDIT_DEPLOY_STATES = [
|
|
# Secondary subclouds should not be audited as they are expected
|
|
# to be managed by a peer system controller (geo-redundancy feat.)
|
|
"create-complete",
|
|
"create-failed",
|
|
"pre-rehome",
|
|
"rehome-failed",
|
|
"rehome-pending",
|
|
"rehoming",
|
|
"secondary",
|
|
"secondary-failed",
|
|
]
|