Skip to content

Commit

Permalink
use cluster_id instead of cluster_name with ocm command (#1237)
Browse files Browse the repository at this point in the history 
➜ ocm list machinepools --cluster ods-qe-rosa-01
Error: Failed to get cluster 'ods-qe-rosa-01': There are 3 subscriptions
with cluster identifier or name 'ods-qe-rosa-01'

➜ ocm list idps --cluster ods-qe-rosa-01
Error: Failed to get cluster 'ods-qe-rosa-01': There are 3 subscriptions
with cluster identifier or name 'ods-qe-rosa-01'

➜ ocm --v=0 delete user xxx --cluster ods-qe-rosa-01
--group=cluster-admins
Error: Failed to get cluster 'ods-qe-rosa-01': There are 3 subscriptions
with cluster identifier or name 'ods-qe-rosa-01'

This fix using the cluster_id instead of the cluster_name

followup of #1056

related to jira ticket: RHOAIENG-3878
  • Loading branch information
@kobihk
kobihk authored Feb 29, 2024
2 parents 4cc011c + 0397a44 commit 7e4a6e8
Showing 1 changed file with 31 additions and 36 deletions.
67 changes: 31 additions & 36 deletions ods_ci/utils/scripts/ocm/ocm.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -119,7 +119,7 @@ def ocm_describe(self, jq_filter=""):
cmd += f" {jq_filter}"
ret = execute_command(cmd)
if ret is None or "Error: Can't retrieve cluster for key" in ret:
log.info(f"ocm describe for cluster {self.cluster_name} failed")
log.info(f"ocm describe for cluster {self.cluster_name} with id: {cluster_id} failed")
return None
return ret

Expand Down Expand Up @@ -309,7 +309,7 @@ def wait_for_osd_cluster_to_be_ready(self, timeout=7200):
while count <= timeout:
cluster_state = self.get_osd_cluster_state()
if cluster_state == "ready":
log.info("{} is in ready state".format(self.cluster_name))
log.info(f"{self.cluster_name} is in ready state")
check_flag = True
break
elif cluster_state == "error":
Expand Down Expand Up @@ -343,18 +343,19 @@ def is_addon_installed(self, addon_name="managed-odh"):

addon_state = self.get_addon_state(addon_name)
if addon_state == "not installed":
log.info("Addon {} not installed in cluster {}".format(addon_name, self.cluster_name))
log.info(f"Addon {addon_name} not installed in cluster {self.cluster_name}")
return False
log.info("Addon {} is installed in cluster {}".format(addon_name, self.cluster_name))
log.info(f"Addon {addon_name} is installed in cluster {self.cluster_name}")
return True

def get_addon_state(self, addon_name="managed-odh"):
"""Gets given addon's state"""

cmd = "ocm list addons --cluster {} --columns id,state | grep {} ".format(self.cluster_name, addon_name)
cluster_id = self.get_osd_cluster_id()
cmd = f"ocm list addons --cluster {cluster_id} --columns id,state | grep {addon_name} "
ret = execute_command(cmd)
if ret is None:
log.info("Failed to get {} addon state for cluster {}".format(addon_name, self.cluster_name))
log.info(f"Failed to get {addon_name} addon state for cluster {self.cluster_name}")
return None
match = re.search(addon_name + "\s*(.*)", ret)
if match is None:
Expand All @@ -363,10 +364,10 @@ def get_addon_state(self, addon_name="managed-odh"):
return match.group(1).strip()

def check_if_machine_pool_exists(self):
"""Checks if given machine pool name already
exists in cluster"""
"""Checks if given machine pool name already exists in cluster"""

cmd = "/bin/ocm list machinepools --cluster {} | grep -w {}".format(self.cluster_name, self.pool_name)
cluster_id = self.get_osd_cluster_id()
cmd = f"ocm list machinepools --cluster {cluster_id} | grep -w {self.pool_name}"
ret = execute_command(cmd)
if not ret:
return False
Expand All @@ -375,24 +376,15 @@ def check_if_machine_pool_exists(self):
def add_machine_pool(self):
"""Adds machine pool to the given cluster"""
if bool(self.reuse_machine_pool) and self.check_if_machine_pool_exists():
log.info(
"MachinePool with name {} exists in cluster "
"{}. Hence "
"reusing it".format(self.pool_name, self.cluster_name)
)
log.info(f"MachinePool with name {self.pool_name} exists in cluster {self.cluster_name}. Hence reusing it")
else:
cmd = (
"/bin/ocm --v={} create machinepool --cluster {} "
"--instance-type {} --replicas {} "
"--taints {} "
"{}".format(
self.ocm_verbose_level,
self.cluster_name,
self.pool_instance_type,
self.pool_node_count,
self.taints,
self.pool_name,
)
cmd = "ocm --v={} create machinepool --cluster {} --instance-type {} --replicas {} --taints {} {}".format(
self.ocm_verbose_level,
self.cluster_name,
self.pool_instance_type,
self.pool_node_count,
self.taints,
self.pool_name,
)
ret = execute_command(cmd)
if ret is None:
Expand Down Expand Up @@ -441,7 +433,8 @@ def wait_for_addon_uninstallation_to_complete(self, addon_name="managed-odh", ti
def list_idps(self):
"""Lists IDPs for the cluster"""

cmd = "ocm list idps --cluster {} --columns name".format(self.cluster_name)
cluster_id = self.get_osd_cluster_id()
cmd = f"ocm list idps --cluster {cluster_id} --columns name"
ret = execute_command(cmd)
if ret is None:
return []
Expand Down Expand Up @@ -688,7 +681,7 @@ def install_managed_starburst_addon(self, license, exit_on_failure=True):
# else:
# self.wait_for_addon_installation_to_complete(addon_name="managed-starburst")
else:
log.info("managed-api-service is already installed on {}".format(self.cluster_name))
log.info(f"managed-api-service is already installed on {self.cluster_name}")

def uninstall_managed_starburst_addon(self, exit_on_failure=True):
"""Uninstalls RHOAM addon"""
Expand Down Expand Up @@ -780,7 +773,8 @@ def create_idp(self):
def delete_idp(self):
"""Deletes Identity Provider"""

cmd = f"ocm --v={self.ocm_verbose_level} delete idp -c {self.cluster_name} {self.idp_name}"
cluster_id = self.get_osd_cluster_id()
cmd = f"ocm --v={self.ocm_verbose_level} delete idp -c {cluster_id} {self.idp_name}"
ret = execute_command(cmd)
if ret is None:
log.info(f"Failed to delete identity provider of type {self.idp_name}")
Expand All @@ -806,7 +800,8 @@ def delete_user(self, user="", group="cluster-admins"):

if user == "":
user = self.htpasswd_cluster_admin
cmd = f"ocm --v={self.ocm_verbose_level} delete user {user} --cluster {self.cluster_name} --group={group}"
cluster_id = self.get_osd_cluster_id()
cmd = f"ocm --v={self.ocm_verbose_level} delete user {user} --cluster {cluster_id} --group={group}"
ret = execute_command(cmd)
if ret is None:
log.info(f"Failed to delete user {user} of group {group}")
Expand Down Expand Up @@ -933,7 +928,7 @@ def delete_cluster(self):
cmd = f"ocm --v={self.ocm_verbose_level} delete cluster {cluster_id}"
ret = execute_command(cmd)
if ret is None:
log.error(f"Failed to delete osd cluster {self.cluster_name}")
log.error(f"Failed to delete osd cluster '{self.cluster_name}' with id: {cluster_id}")
sys.exit(1)
self.wait_for_osd_cluster_to_get_deleted()

Expand All @@ -945,7 +940,7 @@ def wait_for_osd_cluster_to_get_deleted(self, timeout=5400):
while count <= timeout:
cluster_exists = self.is_osd_cluster_exists()
if not cluster_exists:
log.info(f"{self.cluster_name} is deleted")
log.info(f"Cluster '{self.cluster_name}' was deleted")
check_flag = True
break

Expand Down Expand Up @@ -976,14 +971,14 @@ def wait_for_osd_cluster_to_get_hibernated(self, timeout=1800):
while count <= timeout:
cluster_state = self.get_osd_cluster_state()
if cluster_state == "hibernating":
log.info("{} is in hibernating state".format(self.cluster_name))
log.info(f"{self.cluster_name} is in hibernating state")
check_flag = True
break

time.sleep(60)
count += 60
if not check_flag:
log.error(f"{self.cluster_name} not in hibernating state even after 30 mins. EXITING")
log.error(f"{self.cluster_name} not in hibernating state even after {timeout / 60} minutes. EXITING")
sys.exit(1)

def resume_cluster(self):
Expand All @@ -1007,14 +1002,14 @@ def wait_for_osd_cluster_to_get_resumed(self, timeout=3600):
while count <= timeout:
cluster_state = self.get_osd_cluster_state()
if cluster_state == "ready":
log.info("{} is in ready state".format(self.cluster_name))
log.info(f"{self.cluster_name} is in ready state")
check_flag = True
break

time.sleep(60)
count += 60
if not check_flag:
log.error(f"{self.cluster_name} not in ready state even after 30 mins. EXITING")
log.error(f"{self.cluster_name} not in ready state even after {timeout / 60} minutes. EXITING")
sys.exit(1)

def update_notification_email_address(self, addon_name, email_address, exit_on_failure=True):
Expand Down

0 comments on commit 7e4a6e8

Please sign in to comment.