+*** 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 %{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
-
-Variables %{ROBOT_DEVOPS_FOLDER}/resources/sa_02-vnf_with_vim_metrics_and_autoscaling_data.py
+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
-Force Tags sa_02 cluster_sa daily regression sanity
+Test 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 Scale-Out Threshold From VNF
-
- ${rc} ${stdout}= Run and Return RC and Output osm vnfpkg-show ${vnfd_name} --literal | yq '.df[0]."scaling-aspect"[0]."scaling-policy"[0]."scaling-criteria"[0]."scale-out-threshold"' | tr -d \\"
- Should Be Equal As Integers ${rc} ${success_return_code} msg=${stdout} values=False
+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
${scaleout_threshold}= Convert To Number ${stdout}
- Set Suite Variable ${metric_threshold} ${scaleout_threshold}
- Log ${metric_threshold}
-
+ 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 ${THRESHOLD_TIME}
Create NS Descriptor
-
- Create NSD '%{PACKAGES_FOLDER}/${nsd_pkg}'
-
+ [Documentation] Upload NS package for the testsuite.
+ Create NSD '%{PACKAGES_FOLDER}/${NSD_PKG}'
Instantiate Network Service
-
- ${id}= Create Network Service ${nsd_name} %{VIM_TARGET} ${ns_name} ${ns_config} ${publickey}
- Set Suite Variable ${ns_id} ${id}
-
+ [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}
Get VNF Id
-
- @{vnfr_list}= Get Ns Vnfr Ids ${ns_id}
+ [Documentation] Retrieve VNF instance id to be used later on.
+ @{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
-
- ${ip_addr}= Get Vnf Management Ip Address ${ns_id} ${vnf_member_index}
+ [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}
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
-
- 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_value}= Wait Until Keyword Succeeds ${prometheus_poll_retries} ${prometheus_poll_timeout} Get Metric ${prometheus_host} ${prometheus_port} ${prometheus_user} ${prometheus_password} ${metric_name}
- 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
-
+ [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
Increase VIM-based Metric To Force Auto-scaling
-
- 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
-
+ [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
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
+ 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}
- Run Keyword If ${vdurs} <= 1 Fail msg=There is no new VDU after auto-scaling
-
+ 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_value}= Get Metric ${prometheus_host} ${prometheus_port} ${prometheus_user} ${prometheus_password} ${metric_name}
- Run Keyword If ${metric_value} <= ${metric_threshold} Fail msg=The metric '${metric_name}' value is '${metric_value}' which is lower than '${metric_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