Flow based application discovery reconfiguration for large VM counts (>3K) in Aria Operations for Networks
Article ID: 324428
Updated On:
Products
VCF Operations for Networks
Issue/Introduction
- Reconfigure Aria Operations for Networks for cluster deployment to enable flow based discovery on large number of VMs (> 3K)
- Warning/Caution: Increasing VM capacity decreases total flow capacity.
Environment
- VCF Operations for Networks 6.13.0
- VCF Operations for Networks 6.14.x
Cause
Existing cluster deployment supports Flow-Based Application Discovery VM 3000
Resolution
Follow below mentioned procedure for reconfiguration of the deployment.
- Before starting this process take cold snap-shot of the platform nodes. It is recommended to follow KB Best practices to shutdown VCF Operations for Networks Clustered deployments.
- Perform the steps below to validate the current configuration from VCF Operations for Networks GUI.
- Navigate to to
Settings>Infrastructure and support>Infrastructure and updatespages and check the platform capacity section.
Refer to UI as per the below screenshot: - If capacity shows 3K, then go to step 5.
- Take a putty/ssh session to platform1 and login with username
support
Note:
Below steps should be executed only from platform node1 in a cluster deployment. - Switch to ubuntu user by using command
sudo su ubuntu - Execute below command
python3 /home/ubuntu/build-target/deploymentmanager/change_deployment_config.pyAbove command will executes an interactive program with the following options:
1. Enter deployment def ID to apply 2. List deployment defs for current config 3. List deployment defs for another config 4. exit - Choose the option to list available configs (Option 2)
This will list currently applied config as well as configs that can be applied (see sample output)
The list also has a description of the config to help in understanding the status of service and limits in the config - Get the id of config available to be applied (e.g. PX3C3E in sample output)
- Apply the config and wait for the completion (Option 1)
Note:
This process can take up to 40 minutes to complete the reconfiguration.
Logs for this process can be viewed in the following file and location:/home/ubuntu/logs/deployment_change.log
See below a sample output for changing configuration on a 3 Node cluster with XL brick size deployment setup:support@platform1:~$ sudo su ubuntu ubuntu@platform1:/home/support$ python3 /home/ubuntu/build-target/deploymentmanager/change_deployment_config.py ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 1. Enter deployment def ID to apply 2. List deployment defs for current config 3. List deployment defs for another config 4. exit Please choose one option(enter number): 2 ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Following deployment defs are available to apply: ID: PX3C3E Description: Enterprise, App-Discovery and veriflow services running, NodeManager stopped on platform3 numVMs: 12000 activeFlows: 3000000 numDevices: 300 ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Current active def: ID: PX3C1ED Description: Default Role, Enterprise, App-Discovery and Veriflow running in basic mode numVMs: 18000 activeFlows: 6000000 numDevices: 50 ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 1. Enter deployment def ID to apply 2. List deployment defs for current config 3. List deployment defs for another config 4. exit Please choose one option(enter number): 1 Please enter the ID you want to apply: PX3C3E Are you sure you want apply deployment def PX3C3E?(y/n): y active active active active active active inactive inactive inactive active active active active active active inactive inactive inactive active inactive active active active inactive inactive Created symlink /etc/systemd/system/hadoop-yarn-nodemanager.service → /dev/null. inactive active active inactive Created symlink /etc/systemd/system/batch-analytics.service → /dev/null. Removed /etc/systemd/system/batch-analytics.service. enabled inactive inactive active inactive inactive enabled active inactive inactive enabled active inactive inactive inactive Created symlink /etc/systemd/system/samzajobs.service → /dev/null. inactive inactive inactive inactive inactive inactive inactive inactive inactive enabled enabled masked inactive inactive inactive inactive active active enabled masked enabled inactive inactive inactive inactive active active enabled masked enabled deactivating active deactivating active ubuntu@platform1:/home/support$ - Once above is completed the configurations would shows complete, Click on the Option prompted on the GUI.
- At this time a login screen is seen, login to VCF operations for Networks GUI and validate the deployment changes by navigation to Settings>Infrastructure and support>Infrastructure and updates pages and check the platform capacity section.
Additional Information
Increasing the capacity of VMs for Flow Based Application Discovery will decrease the capacity of Total Flows. Review System Recommendations and Requirements before making any changes.
It is generally recommended to narrow the scope of Flow Based Application Discovery rather than using the above script to increase the capacity. See Discover Applications using Flows for more information.
Feedback
Yes
No
Powered by
