Skip to main content
Drupal emergency support for critical production incidents

Drupal emergency support for critical production incidents

Drupal emergency support for organizations whose production platform is down, compromised or failing under load. Engineers experienced in Drupal diagnose the incident, contain it and aim to restore service or a working workaround, then hand over a written account of the cause.

Phone line: Monday to Friday, 08:00 to 16:00 (Madrid time, CET/CEST). Outside those hours, use the incident form.

Common Drupal emergencies and their symptoms

Most Drupal emergencies start with a symptom you can name. The site or a critical feature is down: a white screen, a 500 error, logins and forms that suddenly stop working, or the message "The website encountered an unexpected error. Try again later." Or the platform broke after an update or migration: database updates fail halfway through, a Composer update leaves the site unusable, the site is stuck in maintenance mode, or a PHP version change or a move to new hosting takes it offline.

Some incidents are about security or load. Signs of a compromise include administrator accounts nobody created, spam pages or redirects to other sites, a "Deceptive site ahead" or malware warning in the browser or in Google Search Console, and unexpected outbound traffic. Other sites fail under traffic, with 502 or 504 errors, timeouts, "Allowed memory size exhausted" messages or a checkout that stops working when a campaign, a news peak or a bot surge arrives.

On larger estates, the failure often sits in the connections. Integrations stop syncing: orders no longer reach the ERP, leads no longer reach the CRM, payments fail or search returns nothing across a multisite or multilingual platform. An intranet or member area behind SSO (single sign-on) stops letting people in, or administrators get "Access denied" after an update.

History makes things harder. An inherited platform fails and the previous agency is gone, or the site has no documentation at all. A Drupal 7 platform past end of life starts failing, or custom code, including AI-generated modules, was deployed without review and now breaks in production.

Cosmetic issues, content changes and new features belong in planned maintenance, and the form lets you describe them so they can be scheduled for later.

How Metadrop handles a Drupal incident

Metadrop works through a production incident in a fixed order, so that changes on the live platform follow from the diagnosis. On an inherited platform, the approach is to stabilise and understand first: urgent security fixes come before improvements, and structural changes wait until the incident is over. Security incidents follow the incident-handling practice of Metadrop's ENS (Categoría Media) certified security framework: contain, preserve evidence, remediate, then analyse the cause.

  1. Triage

    An engineer confirms the impact, reviews the priority you reported and agrees one communication channel for the incident, so you are not relaying messages between people.

  2. Diagnosis before changes

    Logs, recent deploys, updates and traffic are reviewed, and the aim is to preserve logs and backups before anything changes in production.

  3. Containment

    The goal is to stop the damage spreading: blocking malicious traffic at the edge, isolating a compromised component, or putting the site into maintenance mode while it is investigated.

  4. Fix or functional workaround

    Where a permanent fix is complex, the priority is a workaround that keeps the critical business process running, for example patching a module or temporarily downgrading it. The permanent fix is planned afterwards.

  5. Root-cause report

    A written account of what happened, why, what was changed and what is recommended next. Where it applies, a functional or visual regression test is added, designed to stop the same failure returning unnoticed.

Europe flag

What we need from you to start

Having a few details ready saves a round of questions at the start. About the incident, that means the site URL, the Drupal version, what fails, since when, and what changed just before: a deploy, an update, a campaign or a hosting change. On impact, say whether it is an outage, a critical feature down, a security incident or a performance failure, and who is affected, whether customers, editors or staff.

Evidence helps the diagnosis: screenshots or a screen recording, the error messages, and the browser or device where the problem appears. For access, the engineer needs the hosting or server (SSH/SFTP), the code repository and a Drupal administrator account. Traffic and security incidents also need access to the CDN, the WAF (web application firewall) and DNS.

Name one contact who decides: someone who can approve a workaround or a maintenance window while the incident is open. Credentials are shared through a secure channel, and rotating them after a security incident is part of the recommendations.

Every member of the Metadrop team works from Spain, so data handled on Metadrop's side stays under EU law.

Drupal Certified Partner: Silver
Logo acreditativo certificación ISO 9001
Logo acreditativo certificación ISO 27001 - ADOK certificación
distintivo_ens_certificacion

Experience with unstable and inherited Drupal platforms

The examples below come from stabilisation, takeover and security work on business-critical Drupal platforms. An emergency draws on the same skills.

Metadrop has spent 15+ years building Drupal platforms, has contributed 100+ modules to drupal.org and holds Drupal Association Silver Certified Partner status, renewed annually. It is also certified under Spain's ENS (Categoría Media) and ISO 9001. Metadrop has published its approach to mitigating bot traffic across Drupal, reverse proxies and CDNs, as bot load is a growing cause of instability and hosting cost.

They've Trusted Metadrop

From manufacturing and chemicals to food, e-commerce and humanitarian work, organizations running Drupal across Europe have relied on Metadrop to keep business-critical platforms stable and secure.

What clients say

Organizations that have worked with Metadrop on their Drupal platforms.

When to Commission a UX/UI Audit

After the emergency

Once service is back, the root-cause report is written for two readers. The IT lead gets the technical detail, while the communications or marketing director and the budget owner get the impact, risk and decisions needed in plain language. It gives the platform owner a written record of what happened and what was done, ready to share upward, and it flags the remaining risks found during the incident, ranked by urgency.

The next step depends on the cause. After a security incident, a Drupal security audit reviews the rest of the platform for similar weaknesses. After a failure under traffic, a Drupal performance audit looks at caching, infrastructure and code under load. Where the cause is architectural or the platform needs a longer plan, Drupal consulting covers those decisions with your team.

Backup cover for in-house Drupal teams

Teams that run Drupal in-house can arrange planned cover before the next incident. A Drupal maintenance contract adds contractual response and resolution times, uptime monitoring and extended availability windows by agreement.

With cover planned in advance, the access, documentation and contacts are already in place when something fails. To discuss cover without an incident in progress, contact Metadrop.

Drupal emergency support FAQ

  • What counts as a Drupal emergency?

    A Drupal emergency is an incident on a production platform that stops a critical business process: the site down, a critical feature failing, a suspected compromise or a failure under traffic. Cosmetic issues, content changes and new features are handled as planned work. Metadrop reviews the priority you report at triage.

  • What should I do when my Drupal site is down?

    When a Drupal site is down, the first goal is to avoid making the cause harder to find. Note what changed just before (a deploy, an update, a hosting or DNS change), save the exact error message and the time it started, and keep the server and Drupal logs. Restoring a backup over the live site or reinstalling modules before anyone has read the logs can erase the evidence of what went wrong. Then report the incident or, during office hours, call.

  • My Drupal site has been hacked: what should I do first?

    On a hacked Drupal site, the priority is to contain the compromise and preserve the evidence before cleaning anything. Change the administrator, hosting and database passwords from a device you trust, restrict access or switch on maintenance mode if the site is serving malicious content, and keep a copy of the files, database and logs as they are. Deleting suspicious files straight away can leave the entry point open and remove the trace of how the attacker got in.

  • Why did my Drupal site break after an update?

    A Drupal site that breaks after an update usually points to one of a few causes: database updates that failed or were never run, a module that is not compatible with the new core or PHP version, a patch that no longer applies, or configuration that no longer matches the code. The Drupal log messages and the output of the update are the starting point. Where the failure blocks a critical process, the usual workaround is to roll back that specific update while the cause is fixed.

  • Why does my Drupal site crash under high traffic?

    Drupal sites that crash under high traffic typically do so because requests bypass the cache and reach PHP and the database: personalised or logged-in pages, uncached views, cache settings changed by a deploy, or bots crawling search and filter pages. PHP workers or database connections running out, and slow calls to external services, are other common causes. During an incident, the first steps are usually to block abusive traffic at the CDN or WAF and restore caching for anonymous visitors; a Drupal performance audit covers the longer-term fixes.

  • How quickly can Metadrop respond to a Drupal emergency?

    Metadrop's response time to a Drupal emergency is not published for organizations without a maintenance contract, because it depends on engineer availability and the severity of the incident. The phone line takes critical production incidents Monday to Friday, 08:00 to 16:00 (Madrid time); outside those hours, the incident form is the route. Maintenance clients have contractual response and resolution times, with extended availability windows by agreement.

  • Does Metadrop help organizations that are not clients?

    Metadrop's Drupal emergency support is open to organizations that are not clients, with the work billed by the hour. Reporting an incident does not require a long-term contract.

  • How much does emergency Drupal support cost?

    The cost of emergency Drupal support depends on the time the incident takes, because work is billed by the hour. The main variables are platform complexity, the state of access and documentation, and whether the incident is contained or still spreading. For predictable costs, a maintenance contract covers incident response under agreed terms.

  • How does Metadrop handle a hacked Drupal site?

    Metadrop handles a hacked Drupal site in four stages: contain the compromise, preserve logs and evidence, remediate, then analyse how the attacker got in. Remediation typically includes patching the entry point, reviewing users and permissions, and rotating credentials. The root-cause report describes the security risk found as a technical assessment and does not constitute legal advice.

  • Can Metadrop take over a Drupal site the previous agency left undocumented?

    Metadrop takes over undocumented Drupal sites by securing access first (hosting, repository, domain and DNS, admin accounts) and then stabilising the platform before any rebuild. During an incident, urgent security and stability fixes come first, and structural changes wait for a planned phase.

  • Does Metadrop work on Drupal 7 emergencies?

    Metadrop works on Drupal 7 platforms past end of life, where an incident can be linked to unpatched core, modules or PHP. The incident is stabilised first, and the migration path is a separate decision, covered by Drupal 7 support and migration.

Report your Drupal incident

Describe the incident and its impact. An engineer reviews the report and contacts you to confirm the priority and arrange access. Reporting an incident does not commit you to any work.

For a critical production incident during office hours (Monday to Friday, 08:00 to 16:00, Madrid time), calling +34 910 053 180 is the most direct route.

Escribe tu mensaje aquí...
I have read and accepted the privacy policy regarding data processing.
Answers are generated automatically by AI and may not be accurate.