# DDoS Guide: 10 Essential Checks for Safer Website Recovery

Source: https://hostreta.com/2026/10/ddos-response-guide/

In this guide: DDoS attacks explained: what a DDoS attack is, the main types, signs you’re under attack, protection services, and a simple response plan to keep your site…

Imagine thousands of cars suddenly blocking the road to your store so real customers can’t get in. That’s essentially a DDoS attack — and it can take down websites of any size, from small blogs to major companies.

Table of Contents

[Toggle](#)

- [What a DDoS attack is](#What_a_DDoS_attack_is)
- [Types of attacks](#Types_of_attacks)
- [Signs you’re under attack](#Signs_youre_under_attack)
- [Protection services](#Protection_services)
- [Your response plan](#Your_response_plan)
- [Key Takeaways: DDoS Attacks](#Key_Takeaways_DDoS_Attacks)
- [1. DDoS triage starts with the actual failure](#1_DDoS_triage_starts_with_the_actual_failure)
- [2. Map DDoS protection to the service it covers](#2_Map_DDoS_protection_to_the_service_it_covers)
- [3. Understand the DDoS response boundary at the origin](#3_Understand_the_DDoS_response_boundary_at_the_origin)
- [4. Preserve DDoS evidence without collecting unnecessary data](#4_Preserve_DDoS_evidence_without_collecting_unnecessary_data)
- [5. Coordinate DDoS mitigation with the host](#5_Coordinate_DDoS_mitigation_with_the_host)
- [6. Tune DDoS application controls around legitimate journeys](#6_Tune_DDoS_application_controls_around_legitimate_journeys)
- [7. Use caching without exposing personalized responses](#7_Use_caching_without_exposing_personalized_responses)
- [8. Keep a DDoS communication plan independent of the outage](#8_Keep_a_DDoS_communication_plan_independent_of_the_outage)
- [9. Verify DDoS recovery before removing the incident label](#9_Verify_DDoS_recovery_before_removing_the_incident_label)
- [10. Maintain DDoS readiness through ordinary changes](#10_Maintain_DDoS_readiness_through_ordinary_changes)
- [Illustrative DDoS incident record](#Illustrative_DDoS_incident_record)
- [Testing and procurement boundaries](#Testing_and_procurement_boundaries)
- [Relevant monitoring and infrastructure services](#Relevant_monitoring_and_infrastructure_services)
- [DDoS readiness review for a planned campaign](#DDoS_readiness_review_for_a_planned_campaign)

## What a DDoS attack is

DDoS stands for Distributed Denial of Service. Attackers use many devices — often a botnet of hacked computers, servers, routers, and IoT devices — to flood a target with traffic or requests. The goal is to overwhelm your server, network, or application so legitimate visitors can’t access it.

Why would someone attack your site?

- Extortion: “Pay us or we keep your site offline.”

- Competition or grudges.

- Hacktivism.

- Distraction while attempting another attack.

- Random targeting — sometimes sites are hit simply because they’re reachable.

## Types of attacks

1. Volumetric attacks

Flood your network with massive amounts of traffic (e.g., UDP floods, amplification attacks), saturating bandwidth.

2. Protocol attacks

Exploit weaknesses in network protocols to exhaust server or firewall resources (e.g., SYN floods).

3. Application-layer (Layer 7) attacks

Send many seemingly legitimate HTTP requests — like repeatedly loading search results, login pages, or uncached pages. They require less traffic but are harder to distinguish from real users, and they hit the most resource-intensive parts of your site.

For WordPress sites, common application-layer targets include wp-login.php, xmlrpc.php, search pages, and dynamic WooCommerce endpoints.

## Signs you’re under attack

- Site becomes very slow or unreachable suddenly.

- Traffic spikes from unusual countries, IP ranges, or user agents.

- Many requests to one URL (login, search, a specific page).

- Server resource usage (CPU, memory, bandwidth) hits limits.

- Your host alerts you or temporarily suspends your account to protect others.

- Low engagement may accompany unusual activity, but analytics alone cannot establish that requests are bots or a denial-of-service attack.

Not every spike is an attack — a viral post or campaign can look similar. Check where traffic comes from and whether visitors behave like humans.

## Protection services

CDN with DDoS protection

Services like Cloudflare (including free plans) absorb large volumes of attack traffic across their global network before it reaches your server. See [Cloudflare Free vs Pro](https://hostreta.com/2026/10/cloudflare-free-vs-pro-is-upgrading-worth-it/).

Web Application Firewall (WAF)

Blocks malicious HTTP requests and bad bots at the application layer. See [Web Application Firewalls Explained](https://hostreta.com/2026/10/waf-guide/).

Rate limiting

Limits how many requests one IP can make in a given time — especially useful for login and search pages.

Bot management and challenges

Challenge suspicious visitors with JavaScript checks or CAPTCHAs (e.g., Cloudflare Turnstile or managed challenges).

Hosting-level protection

Many hosts provide network DDoS mitigation. Ask what’s included.

Hardening your site

- Use page caching so most requests don’t hit PHP and the database.

- Disable XML-RPC if unused.

- Protect login pages — see [How to Change the WordPress Login URL](https://hostreta.com/2026/10/how-to-change-the-wordpress-login-url/).

- Hide your origin server IP behind the CDN (and avoid exposing it via DNS records or email headers).

## Your response plan

Prepare before an attack:

- Put your site behind a CDN/WAF now, not during an emergency.

- Know your host’s emergency contact and procedures.

- Set up monitoring so you know immediately — see [Best Uptime Monitoring Tools](https://hostreta.com/2026/10/best-uptime-monitoring-tools/).

- Document your DNS settings and CDN login (with 2FA).

During an attack:

- Confirm the attack: check analytics, server logs, and CDN dashboards.

- Evaluate provider-supported mitigation with the responsible administrator, accounting for legitimate visitors and machine clients.

- Add rate limiting and firewall rules blocking the targeted URLs, countries, or user agents (if clearly malicious).

- Contact your host — they may have additional mitigation.

- Communicate with customers if the outage is visible — see [Crisis Communication](https://hostreta.com/2026/10/crisis-communication/).

- Don’t pay extortion demands — it doesn’t guarantee the attack stops and encourages more attacks.

After an attack:

- Review what was targeted and strengthen those areas.

- Review temporary restrictions and remove or narrow them when justified by current evidence; do not leave unnecessary blocks indefinitely.

- Document the incident and your response.

Further reading: [WordPress Advanced Administration Handbook](https://developer.wordpress.org/advanced-administration/)

## Key Takeaways: DDoS Attacks

- A DDoS attack floods your site with traffic from many sources.

- Volumetric, protocol, and application-layer attacks target different layers.

- A CDN with WAF, rate limiting, and caching is your best defense.

- Prepare a response plan before you need it.

## 1. DDoS triage starts with the actual failure

Identify what users cannot do and which layer appears affected. A website may fail because of an application error, exhausted database connections, an upstream outage or unwanted traffic. Slow pages and a traffic spike are observations, not a confirmed cause. Record the affected hostname, time, error messages and available provider alerts. Avoid announcing an attack before the responsible technical team has evidence to support that diagnosis.

Separate reachability, application correctness and capacity. A homepage might remain cached while checkout fails, or the origin might respond while one network route has a problem. Compare permitted monitoring observations and service logs. Do not generate extra traffic to reproduce a suspected DDoS event without an approved testing plan. Ordinary troubleshooting should not become an unplanned load test against an already stressed system.

## 2. Map DDoS protection to the service it covers

Review which hostnames, protocols and routes actually pass through the provider’s protection. A proxied website and a separate mail or administration service may have different coverage. A control available for HTTP traffic should not be assumed to protect every exposed network service. Keep a current diagram and ask the provider to clarify the division between application and network mitigation.

The [Cloudflare explanation of its protection systems](https://developers.cloudflare.com/ddos-protection/about/how-ddos-protection-works/) is a product-specific reference. Confirm onboarding and configuration for the service you actually use rather than relying on a general feature statement. A dashboard showing protection for one application does not establish that an alternate origin path or unrelated hostname is covered.

## 3. Understand the DDoS response boundary at the origin

If traffic can reach the application through another path, an edge control may not see those requests. Assess origin restrictions with the host using its supported method. Hiding an address is not a complete access control. At the same time, an overly broad restriction can block health checks, administration or legitimate proxy traffic. Plan authorized recovery access before changing the boundary.

Record the intended route and the exceptions that are necessary for operation. Keep those exceptions owned and reviewed. Do not paste a generic firewall command into production merely because another site uses the same CDN. The [server isolation guide](https://hostreta.com/2026/10/server-isolation/) distinguishes this network boundary from process and account separation; they solve different parts of the infrastructure problem.

## 4. Preserve DDoS evidence without collecting unnecessary data

Useful evidence can include provider event identifiers, request rates, relevant error logs and resource observations. Record the source, time window and metric definition. Requests, unique clients and sessions are different measurements. An analytics report may omit traffic blocked before page scripts execute, while an edge event count may include repeated requests from the same client.

Keep raw logs restricted and redact sensitive information when escalating. Request bodies, cookies and authorization headers may contain private data. Do not publish them in an open support forum to prove that an incident occurred. Provide the host with a concise timeline and sanitized identifiers so its team can correlate the issue without requiring a broad dump of customer information.

## 5. Coordinate DDoS mitigation with the host

Know the emergency support route before an incident. Ask what the plan includes, who can activate additional controls and whether some actions require the account owner. Clarify billing and service limitations during normal planning rather than accepting unexpected financial commitments during a stressful event. Keep the contact details accessible outside the website that may be unavailable.

During an incident, provide the verified impact, affected services and relevant event references. Avoid simultaneously making several unrelated configuration changes while support investigates. Assign one change owner and record each action. A shared record helps distinguish a successful mitigation from an unrelated recovery and makes it easier to reverse an adjustment that harms legitimate traffic.

## 6. Tune DDoS application controls around legitimate journeys

A browser challenge may work for an ordinary visitor but fail for a payment callback, mobile application or API client. Identify important machine traffic before applying broad restrictions. Rate controls should reflect the endpoint and the product’s supported counting model. A shared address can represent many legitimate people, while a distributed attack may stay below a simple per-address threshold.

The [HTTP DDoS ruleset documentation](https://developers.cloudflare.com/ddos-protection/managed-rulesets/http/) describes one provider’s application-layer system. Use the actual available controls and event evidence. Do not block whole countries or user-agent strings merely because they appear unfamiliar. Review the effect on real business journeys and use the narrowest justified change that the responsible administrator can validate.

## 7. Use caching without exposing personalized responses

Serving eligible public content from cache can reduce origin work, but not every response is safe to cache. Account pages, carts, checkout and personalized information need the application’s supported treatment. Avoid applying a cache-everything rule indiscriminately during an emergency. An availability improvement that exposes private data or serves the wrong customer’s state creates a different serious problem.

Review the [WordPress cache guide](https://hostreta.com/2026/10/best-wordpress-caching-plugins-compared/) and the site’s actual commerce or membership requirements. Measure what the origin is doing rather than assuming that installing another plugin fixes a network-level flood. Cache behavior, resource capacity and request filtering are complementary tools, with separate configuration and validation responsibilities.

## 8. Keep a DDoS communication plan independent of the outage

Prepare an authorized update channel that remains usable when the main website fails. Assign who verifies facts and who publishes them. State the affected service, current action and next update when known. Do not promise a restoration deadline or claim that orders, payments and data are unaffected until the relevant checks support that statement.

Use the [crisis communication workflow](https://hostreta.com/2026/10/crisis-communication/) to keep messages consistent. Avoid public speculation about the attacker or motive. A denial-of-service claim should not distract the team from checking whether another incident is also present. If a threat or extortion message arrives, preserve it and follow the organization’s incident and appropriate reporting process through trusted contacts.

## 9. Verify DDoS recovery before removing the incident label

Check important user journeys after the service appears reachable. Public pages loading does not establish that login, checkout or integrations are working. Compare application errors, resource behavior and provider events over a defined period. Keep any remaining degraded functionality visible in the status record. Do not report complete recovery solely because one uptime probe succeeds.

Review temporary rules and exceptions with the responsible owner. Some may need narrowing or removal after the event; others may become maintained controls if justified. Record why each remains. The [security review checklist](https://hostreta.com/2026/10/website-security-review/) helps preserve unresolved questions without pretending that a successful mitigation makes the site immune to future DDoS incidents.

## 10. Maintain DDoS readiness through ordinary changes

Revisit coverage after migrations, DNS changes, new APIs and added services. Update the request-path diagram and emergency contacts. Confirm that the business still controls the provider accounts and recovery methods. A protection design that worked before a move can lose coverage if a hostname is no longer routed through the expected service.

Keep [uptime monitoring](https://hostreta.com/2026/10/best-uptime-monitoring-tools/) and [tested backup recovery](https://hostreta.com/2026/10/automatic-backups/) in the wider plan. Backups do not stop a traffic flood, but they remain relevant to separate operational failures. A larger VPS may help a legitimate capacity problem while still being inadequate against a network attack. Choose infrastructure changes based on evidence rather than a blanket promise that more RAM defeats DDoS.

### Illustrative DDoS incident record

A small business can record when failures began, which journeys failed, the current diagnosis, provider ticket, relevant metrics and each approved change. Include expected and observed results. Mark the cause as unconfirmed until the evidence supports it. This is an example of a record to complete, not a report that Hostreta detected an attack against your site or measured a provider’s mitigation capacity.

If a new challenge rule restores public-page availability but breaks a sandbox callback, note both outcomes. Work with the integration owner and provider to adjust the specific scope. Do not describe the rule as fully successful while an important business function remains broken. Clear evidence makes the next action more practical than celebrating a lower traffic graph without checking what legitimate users can do.

### Testing and procurement boundaries

Any deliberate resilience or load test needs explicit authorization, provider coordination and a defined scope. Do not hire an unknown service to send attack traffic at a production site as an informal demonstration. The test must account for shared infrastructure, customer impact and provider rules. Ordinary content guidance is not permission to run a DDoS simulation against systems outside an agreed assessment.

Before buying additional protection, ask what specific gap it addresses and how success will be validated. Compare current terms, onboarding needs and support responsibilities. Keep claims of unlimited capacity or guaranteed availability in their contractual context rather than repeating them as a personal assurance. A sustainable response plan includes ownership and recovery, not only a product subscription.

### Relevant monitoring and infrastructure services

Promotional recommendation: review [KAWEDMA’s 360 Monitoring offering](https://kawedma.com/store/360monitoring) if you need monitoring to evaluate. Confirm the actual checks and alerts. Monitoring can report symptoms; it is not itself a guarantee of DDoS mitigation or automatic incident resolution.

Affiliate disclosure: Hostreta may earn a commission if you purchase through this link. Where a separate capacity and support review justifies infrastructure changes, compare [hosting.com’s managed VPS service](https://my.hosting.com/aff/a4acdd3ba5260a0812f3b4249da227e940d132e7). Confirm current network protection, support scope and limits directly. A VPS purchase should not be presented as a complete DDoS defense or a substitute for protecting the actual request path.

DDoS triage: identify the failure layer. Original explanatory diagram.

DDoS response: coordinate narrow changes. Original explanatory diagram.

DDoS recovery: validate real availability. Original explanatory diagram.

### DDoS readiness review for a planned campaign

Before an expected marketing campaign, tell the responsible technical team about the timing and the legitimate traffic paths. Review the landing page, form and checkout behavior under the site’s normal approved testing process. Confirm that monitoring and support contacts are ready. Do not interpret a successful campaign as an attack merely because requests arrive quickly or from unfamiliar networks.

Keep a baseline of relevant application and infrastructure observations so an incident can be compared with ordinary activity. Record the measurement source and time window. If a request-rate increase coincides with failures, investigate whether the bottleneck is the application, database, network or an external integration. The appropriate action may be performance work rather than an emergency block.

During the campaign, communicate any restriction that might affect legitimate visitors to the business owner. A broad challenge can reduce traffic while also preventing customers from completing the intended action. Check actual outcomes before calling the change successful. Afterward, review the record and update the plan with evidence rather than assuming that more restrictive settings should remain forever.

This preparation supports both capacity planning and DDoS response without confusing them. It also gives the host a clearer account of what the business expected, what it observed and which changes were approved. Clear context can save time when a service is already under pressure.
