300-535 Sample Questions & Answers

The candidate datastore is specifically designed for this purpose. It acts as a staging area where configurations can be built, modified, and validated without affecting the live network (the running datastore). Once the changes in the candidate datastore are complete and validated, they can be committed to the running datastore in a single, atomic transaction. This perfectly matches the company's requirement for isolated changes and transactional commits. The running datastore is the live configuration, and startup is the configuration loaded on boot.

The standard RESTCONF URL structure for accessing data is /restconf/data/{module}:{container}. For operational data (state data), you still use the /data resource path. The distinction between config and state data is handled by the config keyword within the YANG model itself (config true vs config false). Therefore, the correct URL uses /restconf/data/ followed by the module name (Cisco-IOS-XR-ipv4-bgp-oper) and the top-level container (bgp).

YDKOperationError is the specific exception raised by YDK when a NETCONF operation (like or ) fails on the server side and returns an element. This is the correct exception to catch for issues like invalid configuration. YDKServiceProviderError is for connection issues, and YDKModelError is for issues with the YANG model itself (e.g., trying to set a value of the wrong type). YDKError is the base class and is too generic.

Protocol Buffers (Protobuf) is a binary serialization format that is highly efficient and strongly typed. gRPC is a high-performance RPC framework that uses HTTP/2 for transport and Protobuf as its default data format. This combination is specifically designed for high-volume, low-latency, machine-to-machine communication, making it ideal for modern telemetry systems. JSON and XML are text-based and less efficient, and YAML is not typically used for high-speed streaming.

NetMiko uses the paramiko library underneath for SSH connections. To use a bastion host or SSH jump host, you need to pass proxy information to paramiko. The correct way to do this in NetMiko's ConnectHandler is by using the ssh_proxy parameter. This parameter takes a dictionary that defines the proxy command, which is typically ssh -W %h:%p user@bastion_host.

Cisco WAE's Simulation Analysis tool is the correct feature for performing 'what-if' scenarios. It allows engineers to simulate failures of network elements like nodes, circuits, or SRLGs and observe the impact on traffic, utilization, and service paths (including SR-TE policies) without affecting the live network. Demand Deduction is used to infer traffic demands, and Changeover Analysis is for comparing two network plans.

A for loop (or while loop) combined with a try...except block is the standard pattern for implementing retry logic. The loop controls the number of attempts. Inside the loop, the try block contains the API call. If the call fails and raises an exception (e.g., for a 503 status), the except block catches it, waits for a specified delay, and allows the loop to continue to the next attempt. Asynchronous callbacks and multithreading are for concurrency, not for sequential retries.

pyang is a command-line tool used to work with YANG models. Its primary functions are to validate the syntax and semantics of YANG modules and to transform them into various other formats. One of its most common uses is to generate a human-readable tree structure of the data model, which helps developers understand the model's hierarchy and available nodes. It does not generate YDK bindings (that's ydk-gen) or act as a NETCONF client.

This XPath expression correctly navigates the YANG tree from the root. It correctly identifies the list neighbor and uses a predicate [neighbor-address='192.168.10.1'] to filter for the specific neighbor instance where the key neighbor-address has the value '192.168.10.1'. Finally, it selects the connection-state leaf within that specific neighbor instance. The other options use incorrect syntax for list predicates or pathing.

Python virtual environments (created with tools like venv or virtualenv) are the standard way to isolate project-specific dependencies. They allow a project to have its own set of installed packages and Python interpreter, independent of other projects and the system-wide Python installation. This prevents version conflicts and ensures that the CI/CD pipeline runs with the exact same library versions every time, as defined in a requirements.txt file.