-*** Comments ***
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# See the License for the specific language governing permissions and
# limitations under the License.
-
*** Settings ***
Documentation [SA-02] VNF with VIM-based metrics and auto-scaling.
Library Collections
Library SSHLibrary
-Resource ../lib/vnfd_lib.resource
-Resource ../lib/nsd_lib.resource
-Resource ../lib/ns_lib.resource
-Resource ../lib/ssh_lib.resource
-Resource ../lib/prometheus_lib.resource
+Resource %{ROBOT_DEVOPS_FOLDER}/lib/vnfd_lib.robot
+Resource %{ROBOT_DEVOPS_FOLDER}/lib/nsd_lib.robot
+Resource %{ROBOT_DEVOPS_FOLDER}/lib/ns_lib.robot
+Resource %{ROBOT_DEVOPS_FOLDER}/lib/ssh_lib.robot
+Resource %{ROBOT_DEVOPS_FOLDER}/lib/prometheus_lib.robot
-Test Tags sa_02 cluster_sa daily regression sanity azure
+Force Tags sa_02 cluster_sa daily regression sanity azure
Suite Setup Run Keyword And Ignore Error Suite Preparation
Suite Teardown Run Keyword And Ignore Error Suite Cleanup
*** Variables ***
# NS instantiation parameters
-${NS_CONFIG} {vld: [ {name: mgmtnet, vim-network-name: %{VIM_MGMT_NET}} ] }
+${ns_config} {vld: [ {name: mgmtnet, vim-network-name: %{VIM_MGMT_NET}} ] }
# NS and VNF descriptor package folder and ids
-${VNFD_PKG} hackfest_basic_metrics_vnf
-${VNFD_NAME} hackfest_basic_metrics-vnf
-${NSD_PKG} hackfest_basic_metrics_ns
-${NSD_NAME} hackfest_basic-ns-metrics
+${vnfd_pkg} hackfest_basic_metrics_vnf
+${vnfd_name} hackfest_basic_metrics-vnf
+${nsd_pkg} hackfest_basic_metrics_ns
+${nsd_name} hackfest_basic-ns-metrics
# NS instance name and id
-${NS_ID} ${EMPTY}
-${NS_NAME} sa_02
+${ns_id} ${EMPTY}
+${ns_name} sa_02
# SSH keys and username to be used
-${PUBLICKEY} %{HOME}/.ssh/id_rsa.pub
-${PRIVATEKEY} %{HOME}/.ssh/id_rsa
-${USERNAME} ubuntu
-${PASSWORD} osm4u
+${publickey} %{HOME}/.ssh/id_rsa.pub
+${privatekey} %{HOME}/.ssh/id_rsa
+${username} ubuntu
+${password} osm4u
# Prometheus polling interval and retries
-${PROMETHEUS_POLL_RETRIES} 15 times
-${PROMETHEUS_POLL_TIMEOUT} 1 minute
+${prometheus_poll_retries} 15 times
+${prometheus_poll_timeout} 1 minute
# Prometheus metrics to retrieve
-${METRIC_NAME} osm_cpu_utilization
+${metric_name} osm_cpu_utilization
# VNF Variables
-${VNF_MEMBER_INDEX} vnf
-${VNF_IP_ADDR} ${EMPTY}
-${VNF_ID} ${EMPTY}
+${vnf_member_index} vnf
+${vnf_ip_addr} ${EMPTY}
+${vnf_id} ${EMPTY}
-${SUCCESS_RETURN_CODE} 0
+${success_return_code} 0
*** Test Cases ***
Create VNF Descriptor
- [Documentation] Upload VNF package for the testsuite.
- Create VNFD '%{PACKAGES_FOLDER}/${VNFD_PKG}'
+
+ Create VNFD '%{PACKAGES_FOLDER}/${vnfd_pkg}'
+
Get Thresholds From VNF
- [Documentation] Read metric threshold and threshold time from VNF descriptor.
- ${rc} ${stdout}= Run And Return RC And Output osm vnfpkg-show ${VNFD_NAME} --literal | yq -r '.df[0]."scaling-aspect"[0]."scaling-policy"[0]."scaling-criteria"[0]."scale-out-threshold"'
- Should Be Equal As Integers ${rc} ${SUCCESS_RETURN_CODE} msg=${stdout} values=False
+
+ ${rc} ${stdout}= Run and Return RC and Output osm vnfpkg-show ${vnfd_name} --literal | yq -r '.df[0]."scaling-aspect"[0]."scaling-policy"[0]."scaling-criteria"[0]."scale-out-threshold"'
+ Should Be Equal As Integers ${rc} ${success_return_code} msg=${stdout} values=False
${scaleout_threshold}= Convert To Number ${stdout}
- Set Suite Variable ${METRIC_THRESHOLD} ${scaleout_threshold}
- Log ${METRIC_THRESHOLD}
- ${rc} ${stdout}= Run And Return RC And Output osm vnfpkg-show ${VNFD_NAME} --literal | yq -r '.df[0]."scaling-aspect"[0]."scaling-policy"[0]."threshold-time"'
- Should Be Equal As Integers ${rc} ${SUCCESS_RETURN_CODE} msg=${stdout} values=False
+ Set Suite Variable ${metric_threshold} ${scaleout_threshold}
+ Log ${metric_threshold}
+
+ ${rc} ${stdout}= Run and Return RC and Output osm vnfpkg-show ${vnfd_name} --literal | yq -r '.df[0]."scaling-aspect"[0]."scaling-policy"[0]."threshold-time"'
+ Should Be Equal As Integers ${rc} ${success_return_code} msg=${stdout} values=False
${threshold_time_value}= Convert To Number ${stdout}
- Set Suite Variable ${THRESHOLD_TIME} ${threshold_time_value}
- Log ${THRESHOLD_TIME}
+ Set Suite Variable ${threshold_time} ${threshold_time_value}
+ Log ${threshold_time}
+
Create NS Descriptor
- [Documentation] Upload NS package for the testsuite.
- Create NSD '%{PACKAGES_FOLDER}/${NSD_PKG}'
+
+ Create NSD '%{PACKAGES_FOLDER}/${nsd_pkg}'
+
Instantiate Network Service
- [Documentation] Instantiate the NS for the testsuite.
- ${id}= Create Network Service ${NSD_NAME} %{VIM_TARGET} ${NS_NAME} ${NS_CONFIG} ${PUBLICKEY}
- Set Suite Variable ${NS_ID} ${id}
+
+ ${id}= Create Network Service ${nsd_name} %{VIM_TARGET} ${ns_name} ${ns_config} ${publickey}
+ Set Suite Variable ${ns_id} ${id}
+
Get VNF Id
- [Documentation] Retrieve VNF instance id to be used later on.
- @{vnfr_list}= Get Ns Vnfr Ids ${NS_ID}
+
+ @{vnfr_list}= Get Ns Vnfr Ids ${ns_id}
Log List ${vnfr_list}
- Set Suite Variable ${VNF_ID} ${vnfr_list}[0]
+ Set Suite Variable ${vnf_id} ${vnfr_list}[0]
+
Get VNF IP Address
- [Documentation] Get the mgmt IP address of the VNF to be used in later tests.
- ${ip_addr}= Get Vnf Management Ip Address ${NS_ID} ${VNF_MEMBER_INDEX}
+
+ ${ip_addr}= Get Vnf Management Ip Address ${ns_id} ${vnf_member_index}
Log ${ip_addr}
- Set Suite Variable ${VNF_IP_ADDR} ${ip_addr}
+ Set Suite Variable ${vnf_ip_addr} ${ip_addr}
+
Get VNF VIM-based Metric Before Auto-scaling
- [Documentation] Get from Prometheus the VIM metric associated to the VNF auto-scaling group.
- Variable Should Exist ${PROMETHEUS_POLL_RETRIES} msg=Metric polling retries is not available
- Variable Should Exist ${PROMETHEUS_POLL_TIMEOUT} msg=Metric polling timeout is not available
- Variable Should Exist ${PROMETHEUS_HOST} msg=Prometheus address is not available
- Variable Should Exist ${PROMETHEUS_PORT} msg=Prometheus port is not available
- Variable Should Exist ${METRIC_NAME} msg=Prometheus metric name is not available
- ${metric_filter}= Set Variable ns_id=${NS_ID}
- ${metric_value}= Wait Until Keyword Succeeds ${PROMETHEUS_POLL_RETRIES} ${PROMETHEUS_POLL_TIMEOUT} Get Metric ${PROMETHEUS_HOST} ${PROMETHEUS_PORT} ${PROMETHEUS_USER} ${PROMETHEUS_PASSWORD} ${METRIC_NAME} ${metric_filter}
- IF ${metric_value} <= 0
- Fail msg=The metric '${METRIC_NAME}' value is '${metric_value}'
- END
- IF ${metric_value} >= ${METRIC_THRESHOLD}
- Fail msg=The metric '${METRIC_NAME}' value is higher than '${METRIC_THRESHOLD}' before scaling
- END
+
+ Variable Should Exist ${prometheus_poll_retries} msg=Metric polling retries is not available
+ Variable Should Exist ${prometheus_poll_timeout} msg=Metric polling timeout is not available
+ Variable Should Exist ${prometheus_host} msg=Prometheus address is not available
+ Variable Should Exist ${prometheus_port} msg=Prometheus port is not available
+ Variable Should Exist ${metric_name} msg=Prometheus metric name is not available
+ ${metric_filter}= Set Variable ns_id=${ns_id}
+ ${metric_value}= Wait Until Keyword Succeeds ${prometheus_poll_retries} ${prometheus_poll_timeout} Get Metric ${prometheus_host} ${prometheus_port} ${prometheus_user} ${prometheus_password} ${metric_name} ${metric_filter}
+ Run Keyword If ${metric_value} <= 0 Fail msg=The metric '${metric_name}' value is '${metric_value}'
+ Run Keyword If ${metric_value} >= ${metric_threshold} Fail msg=The metric '${metric_name}' value is higher than '${metric_threshold}' before scaling
+
Increase VIM-based Metric To Force Auto-scaling
- [Documentation] Connect to the VNF via SSH and force VIM metric to increase.
- Variable Should Exist ${PRIVATEKEY} msg=SSH private key not available
- Execute Remote Command Check Rc Return Output ${VNF_IP_ADDR} ${USERNAME} ${PASSWORD} ${PRIVATEKEY} for i in {1..9}; do yes &> /dev/null & done
+
+ Variable Should Exist ${privatekey} msg=SSH private key not available
+ Execute Remote Command Check Rc Return Output ${vnf_ip_addr} ${username} ${password} ${privatekey} for i in {1..9}; do yes &> /dev/null & done
+
Wait VIM-based Metric To Exceed Threshold
- [Documentation] Wait until the VIM metric exceeds threshold, which should lead to a scale up.
- Variable Should Exist ${PROMETHEUS_POLL_RETRIES} msg=Metric polling retries is not available
- Variable Should Exist ${PROMETHEUS_POLL_TIMEOUT} msg=Metric polling timeout is not available
- Wait Until Keyword Succeeds ${PROMETHEUS_POLL_RETRIES} ${PROMETHEUS_POLL_TIMEOUT} Check VIM-based Metric Exceeds Threshold
+
+ Variable Should Exist ${prometheus_poll_retries} msg=Metric polling retries is not available
+ Variable Should Exist ${prometheus_poll_timeout} msg=Metric polling timeout is not available
+ Wait Until Keyword Succeeds ${prometheus_poll_retries} ${prometheus_poll_timeout} Check VIM-based Metric Exceeds Threshold
+
Wait Threshold Time
- [Documentation] Wait until the VIM metric has exceeded threshold for the defined period in the descriptor.
- ${threshold_time_sleep}= Evaluate ${THRESHOLD_TIME} + 1
+
+ ${threshold_time_sleep}= Evaluate ${threshold_time} + 1
Sleep ${threshold_time_sleep} minutes Wait scale-out threshold time plus 1 minute
+
Check VIM-based Metric Exceeds Threshold After Threshold-time
- [Documentation] Check that the VIM metric has exceeded the threshold.
+
Check VIM-based Metric Exceeds Threshold
+
Get VDUs After Auto-scaling
- [Documentation] Check that the VNF has scaled up and the humber of VDUs has increased.
+
Sleep 2 minutes Wait for auto-scale to take place
- @{vdur_list}= Get Vnf Vdur Names ${VNF_ID}
+ @{vdur_list}= Get Vnf Vdur Names ${vnf_id}
Log List ${vdur_list}
${vdurs}= Get Length ${vdur_list}
- IF ${vdurs} <= 1 Fail msg=There is no new VDU after auto-scaling
+ Run Keyword If ${vdurs} <= 1 Fail msg=There is no new VDU after auto-scaling
+
Delete NS Instance
- [Documentation] Delete NS instance.
[Tags] cleanup
- Delete NS ${NS_NAME}
+
+ Delete NS ${ns_name}
+
Delete NS Descriptor
- [Documentation] Delete NS package from OSM.
[Tags] cleanup
- Delete NSD ${NSD_NAME}
+
+ Delete NSD ${nsd_name}
+
Delete VNF Descriptor
- [Documentation] Delete VNF package from OSM.
[Tags] cleanup
- Delete VNFD ${VNFD_NAME}
+
+ Delete VNFD ${vnfd_name}
*** Keywords ***
Suite Preparation
[Documentation] Test Suite Preparation: Setting Prometheus Testsuite Variables
+
Set Testsuite Prometheus Variables
+
Suite Cleanup
[Documentation] Test Suite Cleanup: Deleting descriptors and NS instance
- Run Keyword If Any Tests Failed Delete NS ${NS_NAME}
- Run Keyword If Any Tests Failed Delete NSD ${NSD_NAME}
- Run Keyword If Any Tests Failed Delete VNFD ${VNFD_NAME}
+
+ Run Keyword If Any Tests Failed Delete NS ${ns_name}
+ Run Keyword If Any Tests Failed Delete NSD ${nsd_name}
+ Run Keyword If Any Tests Failed Delete VNFD ${vnfd_name}
+
Check VIM-based Metric Exceeds Threshold
[Documentation] Auxiliar keyword to check if metric exceeds threshold
- Variable Should Exist ${PROMETHEUS_HOST} msg=Prometheus address is not available
- Variable Should Exist ${PROMETHEUS_PORT} msg=Prometheus port is not available
- Variable Should Exist ${METRIC_NAME} msg=Prometheus metric name is not available
- ${metric_filter}= Set Variable ns_id=${NS_ID}
- ${metric_value}= Get Metric ${PROMETHEUS_HOST} ${PROMETHEUS_PORT} ${PROMETHEUS_USER} ${PROMETHEUS_PASSWORD} ${METRIC_NAME} ${metric_filter}
- IF ${metric_value} <= ${METRIC_THRESHOLD}
- Fail msg=The metric '${METRIC_NAME}' value is '${metric_value}' which is lower than '${METRIC_THRESHOLD}'
- END
+
+ Variable Should Exist ${prometheus_host} msg=Prometheus address is not available
+ Variable Should Exist ${prometheus_port} msg=Prometheus port is not available
+ Variable Should Exist ${metric_name} msg=Prometheus metric name is not available
+ ${metric_filter}= Set Variable ns_id=${ns_id}
+ ${metric_value}= Get Metric ${prometheus_host} ${prometheus_port} ${prometheus_user} ${prometheus_password} ${metric_name} ${metric_filter}
+ Run Keyword If ${metric_value} <= ${metric_threshold} Fail msg=The metric '${metric_name}' value is '${metric_value}' which is lower than '${metric_threshold}'