Free 200-901 practice test: 30 questions with answers

Version-controlled templates applied by automation give repeatable changes and an audit trail per device. Manual typing, random copies and complaint-driven changes are inconsistent and untraceable.

YAML is a human-readable structured format that tools such as Ansible use for variables and inventories. RAID, CSMA/CD and ARP are not data formats.

Parsing the JSON into a data structure and reading the key is reliable and repeatable. Manual reading, image conversion and assumptions are error-prone.

POST is used to create resources, while GET reads, DELETE removes and HEAD retrieves headers only.

401 means the request is unauthenticated or the credentials are invalid. A 503 indicates an unavailable service, 201 indicates creation, and 301 indicates a permanent redirect.

A loop over the devices with programmatic queries and error handling checks all devices consistently and quickly. Manual logins, guessing from names and waiting for reports are slow or unreliable.

Automation applies consistent configuration and makes changes faster and more repeatable with fewer manual mistakes. It does not remove the need for testing, it does not guarantee zero failures, and it reduces manual logins.

The Content-Type header identifies the format of the response body, which tells the client how to parse it. The request method, passwords and DNS suffixes do not describe the body format.

Git tracks changes, supports rollback, and enables collaboration through branches and merges. USB copies, printouts and emailed files lose history and make collaboration difficult.

Automated tests in a lab or staging environment catch errors before production, which is the safe approach. Testing on production, skipping tests and waiting for users all increase the chance of outages.

The exposed password must be rotated, and secrets belong in a secrets manager rather than in code. Comments do not protect secrets, a private repository still keeps the password in code, and publishing passwords is unsafe.

A deterministic function returns the same output for the same input, and repeated calls with the same effect are idempotent. Random output, endless recursion and race conditions are different properties.

A for loop iterates over each item in a sequence and runs the enclosed code for each one. CSS selectors, YAML anchors and printer queues do not repeat actions over a list in Python.

A dictionary maps keys to values, which suits device attributes such as hostname and address. It is not limited to a number, it is not machine code, and it is not restricted to hardware IDs.

Runtime retrieval keeps credentials out of code, and least-privilege service accounts limit the impact of a leak. Committed tokens, printed passwords and shared administrator tokens all expose credentials.

The requests library and the standard urllib and http.client modules send HTTP requests from Python. Pandas is for data analysis, and the macro recorder and PowerPoint plugins are not API clients.

Catalyst Center provides centralized management, policy and APIs for many devices in enterprise networks. Packet Tracer is a simulator, ASDM manages a single firewall, and Webex is a collaboration product.

Intent-based networking describes the desired outcome and uses automation to reconcile devices to it. Manual tickets, ad hoc scripts and paper inventories do not model desired state.

IOS XE supports NETCONF with YANG models, which provide structured configuration and state. Packet Tracer is a simulator, the Meraki dashboard is for cloud-managed devices, and Webex Calling is a telephony product.

Ansible performs agentless configuration management over SSH or APIs, using playbooks that run across many devices. Word, Acrobat and Paint are not automation tools.

A northbound REST API exposes controller functions for programmatic creation, update and deletion of policies. Screen captures, keystrokes and emails are manual and not automation.

A webhook is an HTTP callback that sends event data to a URL, which lets external systems react to events. Routing protocols, cable standards and password policies are unrelated.

Event subscriptions or webhooks trigger scripts or workflows as alarms occur, which is event-driven automation. Annual scripts, manual monitoring and disabled alarms do not respond to events.

The network layer provides logical IP addressing and routing between networks, which the script depends on to reach the device. The physical and data link layers handle local delivery, and the presentation layer does not provide addressing.

HTTPS uses TCP port 443 by default. DNS uses UDP 53, SSH uses TCP 22, and NTP uses UDP 123.

A TCP connection test verifies that the port accepts connections, which is the check needed before scripting. Word processors, spreadsheets and cache clearing do not test reachability.

The default gateway forwards traffic for destinations outside the local subnet, which lets the server reach other subnets through one route. Broadcast, loopback and multicast addresses do not provide that routing function.

SSH and HTTPS both encrypt traffic, and SSH is the standard for command-line device access. HTTPS uses port 443 by default and SSH uses TCP 22, and neither sends plain text.

The default gateway, a router, forwards traffic for destinations outside the local subnet. MAC tables, DNS servers and NTP servers do not route traffic between networks.

DNS resolves hostnames to IP addresses. DHCP assigns addresses, SNMP monitors devices, and Syslog collects log messages.