Network Automation , Operation
How to increase the provisioning and change speed
From Provisioning to Operation
Network engineers used to build simple tools using Python, ASP, PHP , etc to generate configuration for a switch before provisioning and configuring the switch by copy and pasting the generated configuration to the switch or router.
Now a days, with dedicated, flexible Automation tools, network engineers can easily perform quick and fast deployments and reconfiguration.
Dev-Ops and Automation tools such as Ansible, Puppet and Chef are available tools for network administrators to automate their repetitive tasks.
Type Of Tools
The available tools are mainly 2 categories :
1- Agent-Based (Such as Puppet, Chef)
2- Agent Less (Such as Ansible)
Agent-Based tools provide better communication with the tool-master server. they can be used as monitoring tool to find a change or status or an event. they are very fast and reliable polling mechanism.
however agent-based tools require and agent to be installed on the device and it is limited to the available APIs of the agent.
Agent-Less tools just use ssh to connect to a device and execute specific commands. they don't require any agent to be installed on the switch.
User can build any configuration script to execute any commands and parse the results. there is no limitation of APIs.
agent-less tools are very helpful for provisioning new switches in your network. they can connect to multiple switches and configure them.
What they can do for your Network?
New Switch Configuration
Configuring new switches is a tricky job. especially when it is not considered as a project and you are only adding few switches to your network.
Configuring the NTP, DNS, ARP timeout values, OSPF Timers, access lists are all examples of pieces which normally missed out.
Using Automation tool, you only need to configure the out of band management IP address on each new switch. the rest of the configuration will be carried out by automation tool for you without any human mistake.
Proactive Monitoring and Action
We spend lots of time and money for deploying monitoring platforms. Monitoring, ticketing are all like magnifying eyes looking at your network to ind a fault.
But what will happen after a fault has been identified by these tools? They will inform the administrator. Admin will receive emails, tickets, SMS and WhatsApp messages on their phones.
Using the Dev-Ops tools, you can proactively monitor the important parameters of your network and automatically trigger actions. for example if CRC counters on an interface is increasing, shut-down the interface, or
Simple Example : You have 200 switches in your network.
New VLAN 123 needs to be created on all the switches. and Spanning tree needs to be reconfigured.
You are adding new 10G links from each server to the leaf switches. all link aggregations needs to get modified.
How you will execute the above changes? most of admins use manual configuration. however rate of human error is also high which causes re-doing the same.
Using Dev-Ops tools, you can force the tool to execute the jobs on your behalf and ensure that they are executed properly by adding verification tasks.