Behind the Screens: What happens when you type a URL in a browser. Nice illustration.

 

Basic F5 interview questions

 Q1: – What is Server Load Balancing?

Server Load Balancing (SLB) provides network performance and content delivery by

implementing a series of algorithms and priorities to respond to the specific requests made to

the network. In simple terms, SLB distributes clients to a group of servers and ensures that

clients are not sent to failed servers. 

Q2: – What information needs to be provided in order to set up my appliance to do

cookie load balancing?

You need the cookie name, cookie values, the VIP to bind to, and the groups to balance to.

Q3: – What is Health Check in refer to load balancer ?

The Health Check feature of the load balancer that allows you to set parameters to perform

diagnostic observations on the performance of web servers and web server farms associated

with each appliance. 

Health checking allows you to determine if a particular server or service is running or has failed. When a service fails health checks, the SLB algorithm will stop sending clients to that server until the service passes health checks again.

Q4: – When load balancing to a real server, which server will be accessed first?

This depends on the load balancing method that you select. Here are a few examples:

1. Least connections method: The real server with the lowest number of concurrent connections will receive the first connection. 
2. Round robin method: The real server with the lowest entry index will get the first connection. 
3. Shortest response: The loadbalancer or appliance will establish connections with each server and calculate the round trip time. The client connection will go to the real server with the lowest response time.

Q5: -What is the difference between a Persistent Cookie policy and a QoS Cookie policy in array network loadbalancer ?

Persistent Cookie policy selects a group based on the cookie name. A QoS Cookie policy

selects a server group based on the cookie name and value assigned to that group.

Q6: -What is Global Server Load Balancing (GSLB)?

GSLB operates very similarly to SLB, but on a global scale. It allows you to load balance

VIPs from different geographical locations as a single entity. This provides geographical site

fault tolerance and scalability.

Q7: – Does Clustering need to be turned on in order to use GSLB?

Yes, Clustering must be configured and turned on in order to use GSLB. Each proxy within

the site (or cluster) must have the same configuration. Hence, each appliance can act as a

DNS server if it becomes a master for the site. 

Each site will have a unique cluster/SLB/GSLB configuration, and you will use the gslb site overflow command to add the remote GSLB site to the local appliance.

Q8: – What load balancing methods are supported with array network GSLB ?

The Array appliance supports the following methods for GSLB:

1. Least connections method: Least connections sends clients to the site that has the least number of current connections.
2. Round robin method: Round robin simply sends client to each site in round robin succession.
3. Overflow: Overflow allows requests to be sent to another (remote) site when the local site is 80% loaded.

Q9: – What is Reverse Proxy Cache?

Reverse Proxy Cache is a cache that is in front of the origin servers, hence the use of the term reverse in the name. If a client requests a cached object, the proxy will service the request from the cache instead of the origin server.

Q10: – What is meant by dynamic and static content? Can my Array appliance cache

dynamic content?

When a client requests a web page, a web server returns the requested content. If the content

is retrieved from disk/memory and returned to the client without further processing, the

content is said to be static. 

If the content is generated on-the-fly by the server (e.g., based on database data), it is referred to as dynamic content.

Q11: – What is recursion depth?

structure, it refers to the number of levels that can be traversed below the main or parent

page. This is often used to specify how much of a web site (how many levels deep) to preload

into the cache.

Q12: – How does the Cache decide what to cache?

The HTTP Cache-Control header determines the cacheability of the object and can also

determine how long the object should be cached. The cache can be configured to override the

cacheability of an object by specifying the host-name and a regular expression that matches

within the URL of the object.

Q13: – What algorithms are used for cache content replacement?

The Array(Hardware loadbalancer) Cache uses several algorithms for determining cache

replacement behavior, depending on the context. In most cases, the Array will use LRU

(Least Recently Used) to remove older content when space is needed.

Q14: – What is Real Clustering?

Real clustering allows all configurations from all appliances in the cluster to be synchronized.

Of course, only global parameters such as the SLB configuration are synchronized. Local

parameters such as interface IP addresses are not synchronized.

Q15: -What is Virtual Clustering?

Virtual clustering provides fault tolerance for VIPs among cluster members. For example, if

an appliance is handling traffic and it fails, another appliance in the cluster will take over

traffic processing.

Q16: – How does virtual clustering work?

In a cluster of Array appliances(h/w load balancer), one appliance becomes Master for a

particular VIP and handles all traffic related to that VIP. All others stay in Backup status. If

the appliance containing the Master fails, one of the appliances in Backup status changes to

Master status. 

Note: If multiple VIPs are configured, then Masters can be configured so that they are distributed among the appliances in the cluster, providing additional performance.

Q17 : – What parameters need to be defined in the cluster configuration?

Cluster ID, VIP address, interface (Outside or Inside), priorities, and authorization method

(password or none) must be defined. Preempt and mrthresh are optional.

Q18: – What is the purpose of Content Rewrite?

The Content Rewrite function allows CDN (Content Distribution Network) reference strings

to be dynamically inserted into URLs that are embedded in web pages. With respect to one

particular CDN, the term "Linux" refers to the rewriting of embedded URLs to point to

Ubuntu-Linux network.

Q19: – What is meant by dynamic and static content? Can my Array appliance cache

dynamic content?

When a client requests a web page, a web server returns the requested content. If the content

is retrieved from disk/memory and returned to the client without further processing, the content is said to be static.

If the content is generated on-the-fly by the server (e.g., based on database data), it is referred to as dynamic content.

Q20: – What is recursion depth?

In a tree structure, it refers to the number of levels that can be traversed below the main or

parent page. This is often used to specify how much of a web site (how many levels deep) to

preload into the cache.

Q21: – How does the Cache decide what to cache?

The HTTP Cache-Control header determines the cacheability of the object and can also

determine how long the object should be cached. The cache can be configured to override the

cacheability of an object by specifying the host-name and a regular expression that matches

within the URL of the object.

Q22: – What algorithms are used for cache content replacement?

The Array(Hardware loadbalancer) Cache uses several algorithms for determining cache

replacement behavior, depending on the context. In most cases, the Array will use LRU

(Least Recently Used) to remove older content when space is needed.