Your site is down and you haven't changed anything? Your monitoring shows an outage but your servers seem normal? The problem might come from your host or a third-party service.
Quickly identifying a hosting incident saves you from wasting time looking for a bug that doesn't exist, and lets you communicate appropriately with your users.
Signs of a Hosting Incident
- Multiple sites affected: All your sites on the same host are down simultaneously.
- Total inaccessibility: Even SSH/FTP don't work, not just web.
- No recent changes: No deployment, no modification that could explain the outage.
- Official status page: The host has posted an incident on their status page.
How to Verify
- Check status page: OVH, AWS, Gandi... all have status pages. Check there first.
- Twitter/X: Search for host name + "down". Major incidents are quickly reported.
- isitdownrightnow: Third-party services report major outages from big hosts.
Actions to Take
- Communicate: Inform your users via social media or an alternative channel.
- Document: Note times and symptoms for postmortem and potential SLA claims.
- Prepare plan B: If incident lasts, evaluate your failover or migration options.
Frequently Asked Questions
Can I request a refund for the outage?
According to your SLA, yes. Document the incident with evidence (screenshots, MoniTao logs) and contact support.
How to avoid dependence on a single host?
Multi-cloud, CDN with failover, backups on another provider. More complex but more resilient.
Does MoniTao differentiate local vs hosting outage?
MoniTao tests from multiple locations. If all fail, it's probably the host. If only one, it's network.
Should I change hosts after an incident?
An occasional incident happens to everyone. Evaluate frequency, communication, and resolution time before deciding.
Ready to Sleep Soundly?
Start free, no credit card required.