def add(ctx, self_signed):
"""Add creates application and self signed(optional) certs
Args:
ctx ([type]): Click context
self_signed ([type]): add self_signed certs. Defaults to False.
"""
run_playbook(certs_cmd(ctx.obj, self_signed))
def verify_sanity(ctx, namespace):
# check if KUBECONFIG is set else find kubeconfig file and set the
# environment variable
constants = ctx.obj
# set kubeconfig
kubeconfig = os.environ.get('KUBECONFIG')
if not kubeconfig:
kubeconfigs = glob.glob(constants['project_dir'] + "/kubeconfig_*")
if len(kubeconfigs) != 1:
if len(kubeconfigs) == 0:
print_success_msg('No kubeconfig found!!!')
else:
print_error_msg("multiple kubeconfigs found %s!!!" %
repr(kubeconfigs))
return
kubeconfig = kubeconfigs[0]
set_kubeconfig_environ(kubeconfig)
# check if we have a valid namespace
all_namespaces = get_all_namespaces(kubeconfig)
while namespace not in all_namespaces:
namespace = click.prompt('Provide orc8r namespace',
type=click.Choice(all_namespaces))
rc = run_playbook(verify_cmd(ctx.obj, namespace))
if rc != 0:
print_error_msg("Post deployment verification checks failed!!!")
sys.exit(1)
print_success_msg("Post deployment verification ran successfully")
def raw_cleanup(
constants: dict,
override_dict: dict = None,
dryrun: bool = False,
max_retries: int = 2):
"""Perform raw cleanup of resources using internal commands
Args:
constants (dict): config dict
overrides (dict): overide dict
dryrun (bool): flag to indicate dryrun. Defaults to False.
max_retries (int): maximum number of retries
Returns:
list: command list
"""
if not override_dict and not constants.get('cleanup_state'):
backup_fn = tf_backup_fn(constants['project_dir'])
if Path(backup_fn).exists():
constants['cleanup_state'] = backup_fn
if override_dict:
constants.update(override_dict)
# sometimes cleanups might not fully happen due to timing related
# resource dependencies. Run it few times to eliminate all resources
# completely
for i in range(max_retries):
rc = run_playbook(cleanup_cmd(constants, dryrun))
if rc != 0:
print_error_msg("Failed cleaning up resources!!!")
def precheck(ctx):
"""
Performs various checks to ensure successful upgrade
"""
rc = run_playbook(precheck_cmd(ctx.obj))
if rc != 0:
print_error_msg("Upgrade prechecks failed!!!")
sys.exit(1)
print_success_msg("Upgrade prechecks ran successfully")
def destroy_gateways(self, cluster_config):
""" destroy the AWS gateways instantiated through test cluster
Args:
cluster_config (dict): Test cluster configuration
"""
if not cluster_config.gateways:
print("Gateway information not found in cluster configs")
return
project_dir = self.constants["project_dir"]
playbook_dir = self.constants["cloudstrapper_playbooks"]
# cloudstrapper expects secrets.yaml to be present
Path(f"{project_dir}/secrets.yaml").touch()
aws_configs = get_aws_configs()
cstrap_dict = {
"testClusterStacks":
[gw.gateway_id for gw in cluster_config.gateways],
"dirLocalInventory": self.constants["project_dir"],
"idSite": "TestCluster",
"awsAccessKey": aws_configs["aws_access_key_id"],
"awsSecretKey": aws_configs["aws_secret_access_key"],
"dirSecretsLocal": self.constants["secret_dir"],
}
cluster_cleanup = AnsiblePlay(
playbook=f"{playbook_dir}/cluster-provision.yaml",
tags=['clusterCleanup'],
extra_vars=cstrap_dict,
)
network_cleanup = AnsiblePlay(
playbook=f"{playbook_dir}/agw-provision.yaml",
tags=['cleanupBridge', 'cleanupNet'],
skip_tags=['attachIface'],
extra_vars=cstrap_dict,
)
for playbook in [cluster_cleanup, network_cleanup]:
print(f"Running playbook {playbook}")
rc = run_playbook(playbook)
if rc != 0:
raise ClusterDestroyError(f"Failed destroying cluster")
def create_orc8r(
self,
constants: dict,
cluster_uuid: str,
template: ClusterTemplate = None,
skip_certs=False,
skip_precheck=False,
) -> Dict[str, Any]:
""" Create an orc8r instance in the test cluster
Args:
constants (dict): Constants dictionary
template (ClusterTemplate, optional): Cluster template definition. Defaults to None.
skip_certs (bool, optional): Skip certs creation. Defaults to False.
skip_precheck (bool, optional): Skip prechecks. Defaults to False.
Raises:
ClusterCreateError: Exception raised when cluster creation fails
"""
# configure deployment
template.orc8r.infra['magma_uuid'] = cluster_uuid
template.orc8r.infra.update(get_aws_configs())
# set elastic deploy role based on current state
k = 'deploy_elasticsearch_service_linked_role'
template.orc8r.platform[k] = check_elastic_role_not_exists()
mgr = ConfigManager(constants)
template_dict = template.orc8r.to_dict()
for component, configs in template_dict.items():
for k, v in configs.items():
mgr.set(component, k, v)
mgr.commit(component)
# run playbooks in order
if not skip_certs:
logging.debug("Adding self signed and application certs")
rc = run_playbook(certs_cmd(constants, self_signed=True))
if rc != 0:
raise ClusterCreateError(f"Failed running adding certs")
if not skip_precheck:
logging.debug("Running installation prechecks")
rc = run_playbook(precheck_cmd(constants))
if rc != 0:
raise ClusterCreateError(f"Failed running prechecks")
# create the orc8r cluster
rc = tf_install(constants, warn=False)
if rc != 0:
raise ClusterCreateError(f"Failed installing cluster")
# update dns record for parent domain
dns_dict = {
"domain_name": template.orc8r.infra["orc8r_domain_name"],
}
dns_dict.update(constants)
rc = run_playbook(
AnsiblePlay(
playbook=f"{constants['playbooks']}/main.yml",
tags=['update_dns_records'],
extra_vars=dns_dict,
), )
if rc != 0:
raise ClusterCreateError(
f"Failed updating dns records for parent domain", )
cluster_config_dict = {
"uuid": cluster_uuid,
"cluster_type": ClusterType.AWS,
"template": template,
}
return cluster_config_dict
def create_gateways(
self,
constants: dict,
cluster_uuid: str,
template: ClusterTemplate = None,
) -> Dict[str, Any]:
""" Create AGW gateways in the test cluster
Args:
constants (dict): Constants dictionary
template (ClusterTemplate, optional): Cluster template definition. Defaults to None.
skip_certs (bool, optional): Skip certs creation. Defaults to False.
skip_precheck (bool, optional): Skip prechecks. Defaults to False.
Raises:
ClusterCreateError: Exception raised when cluster creation fails
Returns:
ClusterConfig: Returns the cluster configuration
"""
# create edge network
project_dir = constants["project_dir"]
playbook_dir = constants["cloudstrapper_playbooks"]
# cloudstrapper expects secrets.yaml to be present
Path(f"{project_dir}/secrets.yaml").touch()
aws_configs = get_aws_configs()
cluster_stack = AWSClusterFactory.generate_cluster_stack(
template.gateway.prefix,
template.gateway.count,
)
cstrap_dict = {
"clusterUuid": cluster_uuid,
"dirLocalInventory": constants["project_dir"],
"idSite": "TestCluster",
"testClusterStacks": cluster_stack,
"awsAccessKey": aws_configs["aws_access_key_id"],
"awsSecretKey": aws_configs["aws_secret_access_key"],
"idGw": "dummy_gateway",
"dirSecretsLocal": constants["secret_dir"],
"awsAgwAmi": template.gateway.ami,
"awsCloudstrapperAmi": template.gateway.cloudstrapper_ami,
"awsAgwRegion": template.gateway.region,
"awsAgwAz": template.gateway.az,
"orc8rDomainName": template.orc8r.infra['orc8r_domain_name'],
}
key_create = AnsiblePlay(
playbook=f"{playbook_dir}/aws-prerequisites.yaml",
tags=['keyCreate'],
extra_vars=cstrap_dict,
)
bridge_gw_create = AnsiblePlay(
playbook=f"{playbook_dir}/agw-provision.yaml",
tags=['createNet', 'createBridge', 'inventory'],
skip_tags=['attachIface'],
extra_vars=cstrap_dict,
)
# create test instances
test_inst_create = AnsiblePlay(
playbook=f"{playbook_dir}/cluster-provision.yaml",
tags=['clusterStart'],
extra_vars=cstrap_dict,
)
jump_config_dict = {"agws": f"tag_Name_TestClusterBridge"}
jump_config_dict.update(cstrap_dict)
test_ssh_configure = AnsiblePlay(
playbook=f"{playbook_dir}/cluster-provision.yaml",
tags=['clusterJump'],
extra_vars=jump_config_dict,
)
# configure test instances
agws_config_dict = {
"agws": f"tag_Name_{template.gateway.prefix}*",
}
agws_config_dict.update(template.gateway.service_config)
agws_config_dict.update(cstrap_dict)
test_inst_configure = AnsiblePlay(
inventory=f"{project_dir}/common_instance_aws_ec2.yaml",
playbook=f"{playbook_dir}/cluster-configure.yaml",
tags=['exporter', 'clusterConfigure'],
extra_vars=agws_config_dict,
)
max_retries = 3
for i in range(max_retries):
fail = False
for playbook in [
key_create,
bridge_gw_create,
test_inst_create,
test_ssh_configure,
test_inst_configure,
]:
print(f"Running playbook {playbook}")
rc = run_playbook(playbook)
if rc != 0:
fail = True
print("Failed creating gateway cluster...trying again")
break
# sleep 10 seconds
time.sleep(10)
if not fail:
break
# get the newly instantiated gateways
gateways = []
for gw_info in get_gateways(template.gateway.prefix):
(gateway_id, hostname) = gw_info
gateways.append(
GatewayConfig(
gateway_id=gateway_id,
hostname=hostname,
hardware_id="",
), )
internal_config = ClusterInternalConfig(bastion_ip=get_bastion_ip(), )
cluster_config_dict = {
"uuid": cluster_uuid,
"internal_config": internal_config,
"cluster_type": ClusterType.AWS,
"template": template,
"gateways": gateways,
}
return cluster_config_dict
