Glossary

Plain definitions of the technical terms, monitoring concepts and DevOps vocabulary you meet while keeping services online.

133Terms
17Categories

Type to filter terms instantly

A

10 terms
A

A Record

The DNS record that maps a hostname to an IPv4 address. It is what most browsers ultimately look up before connecting to a website.

A

A/B Testing

Showing two versions of a page or feature to different groups of users and measuring which one performs better against a chosen goal.

A

AAAA Record

The IPv6 counterpart of an A record: it maps a hostname to an IPv6 address.

A

Agent

A small program installed on a server or inside a network that runs checks or collects data and reports it to a central monitoring service.

A

Alert Fatigue

The state where people receive so many alerts - many of them unimportant or false - that they start ignoring them, including the ones that matter.

A

Alerting

Automatically notifying people when a monitored system crosses a condition that needs attention, such as being down or too slow.

A

Ansible

An open-source automation tool that configures servers and deploys software over SSH using human-readable YAML playbooks, without an agent on the target machines.

A

API

Application Programming Interface: a defined way for programs to talk to each other, for example by sending HTTP requests to a set of URLs and receiving JSON responses.

A

API Gateway

A single entry point in front of several backend services that routes requests and handles shared concerns such as authentication, rate limiting and logging.

A

Availability

The share of time a service is usable by the people who depend on it, usually written as a percentage over a period.

B

3 terms
B

Backup

A copy of data kept separately so it can be restored after deletion, corruption, hardware failure or an attack. A backup is only proven once a restore has been tested.

B

Bandwidth

The maximum amount of data a network link can carry per second, often confused with the actual throughput achieved.

B

Blue-Green Deployment

Running two identical production environments and switching traffic from the old (blue) to the new (green) one in a single step, so a bad release can be rolled back just as fast.

C

15 terms
C

Cache

A store of recently used data kept closer to where it is needed, so repeated requests can be answered faster without redoing the work.

C

Canary Release

Rolling a new version out to a small share of users or servers first and watching its error rate and latency before releasing it to everyone.

C

Capacity Planning

Estimating the compute, storage and network resources a service will need in the future, so it can grow without running out of headroom.

C

CDN

Content Delivery Network: a network of servers in many locations that cache and serve a site's content from near the visitor, reducing latency and load on the origin.

C

Certificate Authority

A trusted organisation that issues digital certificates, vouching that a public key belongs to a particular domain or organisation.

C

Chaos Engineering

Deliberately injecting failures - killed processes, slow networks, lost servers - into a system to learn how it behaves and fix weaknesses before real incidents expose them.

C

CI/CD

Continuous Integration and Continuous Delivery/Deployment: automatically building and testing every change, and shipping it to production through a repeatable pipeline.

C

CIDR

Classless Inter-Domain Routing notation, such as 192.168.1.0/24, which writes an IP network as an address plus the number of bits in its network prefix.

C

Cloud Computing

Renting computing resources - servers, storage, databases, networking - on demand from a provider instead of owning and running the hardware yourself.

C

Cluster

A group of servers that work together and appear as one system, sharing load and taking over from each other when one fails.

C

CNAME Record

A DNS record that makes one hostname an alias of another, so lookups for the alias follow the target's records.

C

Configuration Management

Keeping servers and software in a known, consistent state by describing their configuration in code and applying it automatically.

C

Container

A lightweight, isolated package that bundles an application with its libraries and settings, so it runs the same way on any host with a container runtime.

C

Cron

The Unix scheduler that runs commands at fixed times or intervals, configured through crontab files of five time fields plus a command.

C

Cron Expression

The schedule part of a cron job, such as */15 * * * *, made of minute, hour, day-of-month, month and day-of-week fields.

D

11 terms
D

Dashboard

A screen that brings together the most important charts and statuses of a system so its health can be understood at a glance.

D

Database Replication

Copying data from one database server to others continuously, for redundancy, failover and spreading read traffic.

D

DDoS Attack

Distributed Denial of Service: flooding a service with traffic from many sources so that real users can no longer reach it.

D

Degraded Performance

A state where a service still responds but more slowly or with partial errors - not fully down, but not healthy either.

D

DevOps

A way of working that joins software development and operations, using automation, shared ownership and fast feedback to ship and run software reliably.

D

Disaster Recovery

The plans and tooling for restoring service after a major failure such as losing a data center, including backups, standby systems and tested runbooks.

D

DNS

The Domain Name System: the internet's directory that turns names like example.com into the IP addresses computers use to connect.

D

DNS Propagation

The delay before a DNS change is seen everywhere, because resolvers keep serving cached answers until their TTL runs out.

D

Docker

A widely used platform for building container images and running containers on a single host.

D

Domain Expiration

The date a domain registration lapses. If it isn't renewed, the website and email stop working and the name can be lost.

D

Downtime

Any period when a service is unavailable or not working correctly for its users, whether planned or unplanned.

E

4 terms
E

Edge Computing

Running computation close to where data is produced or users are located, instead of in a distant central data center, to cut latency.

E

Endpoint

A specific URL of an API or service that accepts requests, such as https://api.example.com/v1/orders.

E

Error Budget

The amount of unreliability an SLO allows - for 99.9% over 30 days, about 43 minutes. Teams spend it on releases and risk, and slow down when it runs out.

E

Escalation Policy

Rules that decide who is notified about an alert and who is notified next if nobody responds within a set time.

F

5 terms
F

Failover

Automatically switching to a standby server, database or site when the primary one fails.

F

False Positive

An alert that reports a problem that didn't really happen, for example a site marked down because of a brief network blip at one check location.

F

Fault Tolerance

A system's ability to keep working, possibly at reduced capacity, when some of its components fail.

F

Feature Flag

A switch in the code that turns a feature on or off at runtime, so it can be released gradually or disabled without a new deploy.

F

Firewall

A network control that allows or blocks traffic based on rules about addresses, ports and protocols.

G

3 terms
G

Git

A distributed version control system that records every change to a set of files and lets many people work on them in parallel branches.

G

GitOps

Managing infrastructure and deployments by declaring the desired state in a Git repository and letting automation make the live system match it.

G

Grafana

An open-source tool for building dashboards and alerts on top of time-series data from sources such as Prometheus.

H

6 terms
H

Health Check

A lightweight endpoint or test that reports whether a service and its key dependencies are working, used by load balancers and monitors.

H

Heartbeat Monitoring

Monitoring in reverse: a job or device sends a regular signal to a unique URL, and an alert fires when the signal stops arriving on time.

H

High Availability

Designing a system with redundancy and no single point of failure so it stays available through individual component failures.

H

HTTP

The Hypertext Transfer Protocol: the request-and-response protocol browsers and APIs use to fetch pages and data on the web.

H

HTTP Status Code

The three-digit number in every HTTP response that says how a request went: 2xx success, 3xx redirect, 4xx client error, 5xx server error.

H

HTTPS

HTTP carried over an encrypted TLS connection, protecting data in transit and proving the server's identity with a certificate.

I

6 terms
I

ICMP

The Internet Control Message Protocol, used for network diagnostics and error messages; ping sends ICMP echo requests.

I

Incident

An unplanned event that disrupts or degrades a service and needs a response, from detection to resolution.

I

Incident Management

The process of detecting, communicating about, fixing and learning from incidents, with clear roles and timelines.

I

Infrastructure as Code

Defining servers, networks and other infrastructure in version-controlled files, so environments can be created and changed reproducibly.

I

IP Address

The numeric address of a device on a network - four numbers like 203.0.113.10 for IPv4, or a longer hexadecimal form for IPv6.

I

IPv6

The newer version of the Internet Protocol with 128-bit addresses, created because IPv4's roughly 4.3 billion addresses ran out.

J

2 terms
J

Jitter

Variation in latency between packets. High jitter hurts real-time traffic such as voice and video even when the average latency is fine.

J

JSON

JavaScript Object Notation: a light text format for structured data, used by most web APIs.

K

3 terms
K

Keyword Monitoring

Checking that a page contains (or doesn't contain) a given word, which catches error pages and broken content that still return HTTP 200.

K

KPI

Key Performance Indicator: a measurable value that shows how well a team or system is meeting an important objective.

K

Kubernetes

An open-source platform that schedules containers across a cluster of machines and keeps them running, scaled and connected.

L

5 terms
L

Latency

The time it takes for data to travel from one point to another and back, or for a request to get its first response.

L

Load Balancer

A component that spreads incoming requests across several servers and stops sending traffic to servers that fail their health checks.

L

Load Testing

Sending a realistic volume of simulated traffic to a system to see how it performs and where it breaks before real users do it.

L

Log

A timestamped record of events written by an application or system, used to understand what happened and when.

L

Log Aggregation

Collecting logs from many servers and services into one searchable place.

M

8 terms
M

Maintenance Window

A scheduled period for planned work, during which monitoring alerts are usually paused and downtime is not counted against uptime targets.

M

Mean Time to Detect (MTTD)

The average time between a problem starting and someone noticing it. Frequent external checks shorten it.

M

Mean Time to Recovery (MTTR)

The average time it takes to restore a service after a failure, from detection to full recovery.

M

Metric

A numeric measurement recorded over time, such as response time, CPU usage or requests per second.

M

Microservices

An architecture that splits an application into small, independently deployable services that talk over the network.

M

Monitoring

Continuously collecting and checking data about systems to know whether they are working, and alerting people when they aren't.

M

Multi-Location Monitoring

Checking a service from several geographic locations and confirming failures across them, which separates real outages from local network problems.

M

MX Record

The DNS record that names the mail servers accepting email for a domain, with priorities that set the order they are tried in.

N

2 terms
N

Nameserver

A server that holds a domain's DNS records and answers queries about them; NS records point to it.

N

NAT

Network Address Translation: rewriting addresses as traffic passes a router, letting many private devices share one public IP.

O

4 terms
O

Observability

How well you can understand a system's internal state from its outputs - metrics, logs and traces - including problems you didn't predict.

O

On-Call

A rotation in which a named person is responsible for responding to alerts outside normal working hours.

O

OpenTelemetry

An open standard and set of SDKs for producing and exporting traces, metrics and logs in a vendor-neutral way.

O

Outage

A period when a service is completely unavailable to its users.

P

8 terms
P

Packet Loss

The share of network packets that never reach their destination, causing retransmissions, slowness or dropped connections.

P

Ping

A basic network test that sends ICMP echo requests to a host and measures whether and how fast it replies.

P

Port

A number from 0 to 65535 that identifies a specific service on a host, such as 443 for HTTPS or 22 for SSH.

P

Port Monitoring

Regularly checking that a TCP port accepts connections, which shows whether services such as mail, databases or SSH are reachable.

P

Postmortem

A written review after an incident describing what happened, why, how it was handled and what will change - ideally without blaming individuals.

P

Private Location

A check location you run inside your own network, so internal services can be monitored and compared with checks from outside.

P

Prometheus

An open-source monitoring system that scrapes metrics from services, stores them as time series and evaluates alert rules.

P

Proxy

An intermediary server that forwards requests between clients and other servers, often adding caching, filtering or anonymity.

R

8 terms
R

Rate Limiting

Capping how many requests a client may make in a period, to protect a service from overload and abuse.

R

Redirect

An HTTP response (301, 302, 307, 308) telling the client to fetch a different URL instead.

R

Redundancy

Having extra copies of critical components so that one can fail without taking the service down.

R

Response Time

How long a server takes to answer a request, usually measured from sending the request to receiving the full response.

R

Reverse Proxy

A server in front of application servers that receives client requests and forwards them, often handling TLS, caching and load balancing.

R

Rollback

Returning a system to its previous known-good version after a faulty change.

R

Root Cause Analysis

Investigating an incident until the underlying reason is found, rather than stopping at the first visible symptom.

R

Runbook

Step-by-step instructions for handling a known task or alert, so anyone on call can respond consistently.

S

12 terms
S

Scalability

A system's ability to handle more load by adding resources, either bigger machines (vertical) or more machines (horizontal).

S

Serverless

A cloud model where you deploy functions or containers and the provider runs and scales them on demand, billing per use.

S

SLA

Service Level Agreement: a contract that promises a level of service, such as 99.9% monthly uptime, and states what happens if it is missed.

S

SLI

Service Level Indicator: the measured value behind a reliability goal, such as the share of successful requests.

S

SLO

Service Level Objective: an internal target for an SLI, such as 99.95% of requests succeeding over 30 days.

S

SMTP

The Simple Mail Transfer Protocol, used to send email between servers, usually on ports 25, 465 or 587.

S

SNMP

The Simple Network Management Protocol, used to read status and counters from routers, switches and other network devices.

S

SSH

Secure Shell: an encrypted protocol for logging in to and running commands on remote machines, normally on port 22.

S

SSL Certificate

A digital certificate (today technically a TLS certificate) that proves a site's identity and enables HTTPS. It has an expiry date and must be renewed.

S

Status Page

A public or private page showing the current health of services and any ongoing incidents or maintenance.

S

Subnet Mask

A value such as 255.255.255.0 that splits an IP address into its network and host parts.

S

Synthetic Monitoring

Running scripted requests or user journeys against a service on a schedule to test it continuously, even when no real users are active.

T

9 terms
T

TCP

The Transmission Control Protocol: a connection-based protocol that delivers data reliably and in order; HTTP, SSH and email run over it.

T

Throughput

How much work a system completes per unit of time, such as requests per second or megabytes transferred per second.

T

Time to First Byte (TTFB)

The time from sending a request until the first byte of the response arrives, covering network, TLS and server processing time.

T

Timeout

The longest a client waits for a response before giving up and treating the request as failed.

T

TLS

Transport Layer Security: the encryption protocol that secures HTTPS and many other connections; the successor to SSL.

T

Traceroute

A diagnostic tool that lists the routers a packet passes through on its way to a destination, showing where delays or losses happen.

T

Tracing

Following a single request as it moves through multiple services, recording each step's timing to find where time and errors come from.

T

TTL

Time To Live: in DNS, how many seconds resolvers may cache a record before asking again; it controls how fast changes spread.

T

TXT Record

A DNS record holding free text, widely used for domain verification and email authentication such as SPF and DKIM.

U

3 terms
U

UDP

The User Datagram Protocol: a connectionless protocol that sends packets without delivery guarantees; DNS, VoIP and games rely on it.

U

Uptime

The share of time a system is running and reachable, usually expressed as a percentage such as 99.9%.

U

Uptime Monitoring

Checking from outside, at regular intervals, that a website, server or API is reachable and responding correctly, and alerting when it isn't.

V

2 terms
V

Virtual Machine

A software-emulated computer with its own operating system, running on shared physical hardware through a hypervisor.

V

VPN

Virtual Private Network: an encrypted tunnel that connects a device or office to a private network over the internet.

W

2 terms
W

Webhook

An HTTP request a service sends automatically to a URL you choose when an event happens, used to connect systems without polling.

W

WHOIS

A lookup service that returns registration details of a domain, such as its registrar and expiry date.

Z

2 terms
Z

Zero Downtime Deployment

Releasing a new version without any interruption for users, typically with rolling, blue-green or canary techniques.

Z

Zero Trust

A security model that trusts no user or device by default, even inside the network, and verifies every access request.