How Crisp Mitigated the November 18th & December 5th Cloudflare Incidents
While such disruptions of this magnitude remain rare (the last one in 2019), they can temporarily affect platforms that rely on Cloudflare’s global network performance and routing capabilities, including Crisp.
Over the past few weeks, Cloudflare — a critical infrastructure provider for much of the internet — experienced two major incidents, on November 18th, 2025 and December 5th, 2025. While such disruptions of this magnitude remain rare (the last one in 2019), they can temporarily affect platforms that rely on Cloudflare’s global network performance and routing capabilities, including Crisp.
Crisp is prepared for such incidents, and in this article, we will explain how we mitigated such impacts.
Why use Cloudflare?
Cloudflare plays a central role in today’s internet. As one of the largest global networks, it provides security, performance optimization, and traffic routing for millions of websites and SaaS platforms. More than 60% of online businesses rely on Cloudflare to protect their applications from cyberattacks, accelerate content delivery, and ensure stable connectivity across continents.
Nowadays, given the scale of DDoS attacks, operating a SaaS with CloudFlare is mandatory.
However, because Cloudflare sits at a critical junction between users and online services, any disruption in its global network can have ripple effects across the web.
November 18th, 2025 — No Impact on Crisp
On November 18, 2025, Cloudflare experienced a major global outage caused by a faulty update in its new Bot Management system. A misconfigured internal database produced an oversized feature file, which was pushed to Cloudflare’s edge network. This file exceeded expected limits and caused the updated bot-management module to crash on thousands of servers, leading many websites and apps to return 5xx errors for several hours.
Crisp was not impacted at all. As a Cloudflare Enterprise customer, Crisp was still operating on Cloudflare’s previous anti-bot system, which did not use the new feature-file mechanism that failed. Because our traffic path did not depend on the affected module, the faulty update never touched our configuration.
However, the incident had significant consequences across the internet. Many high-traffic platforms — including OpenAI, Twitter/X, Canva, Medium, Discord, DoorDash, and several large SaaS products — experienced partial or complete outages during the event.
December 5th, 2025 — Limited Impact on Crisp
The second Cloudflare outage, on December 5th, 2025, was shorter and less severe than the November event but still affected parts of the internet. For Crisp, the impact was limited:
app.crisp.chat was temporarily unreachable for about 30 minutes, due to Cloudflare 500 errors.
However, the Crisp platform itself remained healthy, and all backend systems continued operating normally.
Throughout the incident, customers could still access the service via https://beta.app.crisp.chat, which bypasses Cloudflare entirely and serves as our alternative entry point during network-level issues.
How Crisp Is Built to Handle Incidents Like This
Redundant Systems Designed to Operate Without Cloudflare
Crisp relies on Cloudflare for performance and security — but not for core availability.
Many of our systems can operate independently of Cloudflare, and we continuously improve our routing and failover capabilities to reduce dependency on any single provider.
An Extremely Active Monitoring System
We monitor the entire Crisp platform 24/7 at multiple layers: application health, networking, infrastructure, regional availability, and dependency resilience.
When something goes wrong, we know it. Immediately.
Because of this, there is no need to contact our support team when an incident occurs.
The best source of information is always our live status page:
This page is updated automatically and manually by our engineering team as events unfold.
What You Should Do During an Outage
Avoid Refreshing the Crisp App
When an error appears, your first instinct might be to refresh the page — but that often has the opposite effect.
Crisp includes automatic fallback mechanisms that detect upstream issues and transparently switch to alternate routes or degraded-mode services.
Refreshing the app interrupts these mechanisms and can delay recovery on your side.
It’s better to leave the app open and let Crisp handle the failover process for you.
Use beta.app.crisp.chat as a Backup Entry Point
In rare cases where Cloudflare is the source of the problem, you can use a dedicated Cloudflare-free entry point:
This environment bypasses Cloudflare entirely and gives you direct access to the Crisp platform.
We recommend adding it to your bookmarks for emergencies.
Closing Thoughts
Cloudflare incidents of this scale are uncommon, but they remind us how interconnected the modern internet is.
Our team takes these events seriously and continually enhances Crisp’s resilience to external disruptions.
We remain fully committed to providing a reliable, fault-tolerant platform — even when the wider web experiences turbulence.
As always, for real-time updates:
👉 https://status.crisp.chat
Why Self-Service is the Future of Support (Without Losing the Human Touch)
True self-service isn’t about AI chatbots. It’s about giving customers the power to complete tasks on their own: fast, intuitively, and without contacting support. When your product removes friction, retention rises, satisfaction grows, and support scales effortlessly.
Automated ticket resolution: What It Is and how to measure success
Want to kick 2026 the right way for your support experience? Get familiar with automated ticket resolution and experience what it means for your business.
From FAQs to AI: How Chatbots Deliver AI Self-Service Support at Scale
want to improve your self-service support strategy? Here is how you can leverage AI to scale your customer support strategy
