Mastodon Instance Going Offline: How to Detect and React

When your Mastodon instance goes offline, you lose access to your account, your followers cannot see your posts, and the entire community you built becomes unreachable. This can happen due to server hardware failure, a denied hosting renewal, or the admin simply walking away from the project. This article explains the warning signs that precede an instance shutdown, the tools you can use to detect an outage, and the exact steps to take to preserve your data and move your account to a new instance safely.

Why Mastodon Instances Go Offline and How to Spot the Warning Signs

Mastodon instances are independently operated servers. Unlike a centralized platform, there is no single company guaranteeing uptime. An instance can go offline for several reasons: the admin runs out of money to pay the hosting bill, the server hardware fails, the admin loses interest and stops maintaining the software, or the domain name expires. In rare cases, a hosting provider may suspend the server due to terms-of-service violations or spam complaints.

Before a total shutdown, most instances show warning signs. The most common signal is a gradual decline in server responsiveness. Posts may take longer to send, the timeline may show a “Failed to fetch” error more frequently, and the instance’s public health page may report a high error rate. Another warning sign is the admin becoming inactive on their own instance — they stop posting, stop replying to moderation reports, and stop communicating on the instance’s announcement feed. If the instance has a status blog or a Patreon page, you may see that funding goals are no longer being met.

You can detect an outage using three methods. First, try loading your instance URL in a browser. If you see a “This site can’t be reached” error, the server is down. Second, use a third-party uptime checker like UptimeRobot or Down For Everyone Or Just Me to confirm the outage is not just your local network. Third, check the instance’s health endpoint, which is typically available at https://yourinstance.com/health and returns a JSON object with server status. If the endpoint returns a 5xx error or no response, the instance is offline.

Steps to React When Your Mastodon Instance Goes Offline

The moment you suspect your instance is offline, act quickly. The window to export your data and migrate your account may be short, especially if the admin has abandoned the server permanently. Follow these steps in order.

1. Confirm the outage with multiple sources
   Open your instance URL in a browser. If it fails, use Down For Everyone Or Just Me to check if the site is down globally. Then check the instance’s Mastodon status page — many instances have a subdomain like status.yourinstance.com that reports server health. Finally, search for your instance name on Mastodon’s official instance picker or on fediverse observer sites like instances.social. If the instance is listed as “down” or “unreachable,” proceed to the next step.
2. Check the admin’s contact information
   If the instance is still partially accessible, visit the About page (usually /about or /about/more) and look for the admin’s email address or a link to their personal website. Send a polite message asking about the outage and expected restoration time. If the instance is completely unreachable, search for the admin’s handle on other instances — many admins have secondary accounts. Knowing the admin’s status helps you decide whether to wait or migrate.
3. Export your data archive
   If the instance comes back online even briefly, immediately export your data. Log in and go to Preferences > Import and Export > Export. Click the “Request your archive” button. Mastodon will generate a zip file containing your posts, media, lists, and bookmarks. Download this file to your local computer. Without an archive, you lose all your content permanently.
4. Create a new account on a different instance
   While waiting for the old instance to recover, sign up for a new account on a stable, well-funded instance. Choose an instance that has been operating for at least one year, has a clear code of conduct, and lists its admin contact publicly. Instances run by universities, nonprofits, or established communities are generally more reliable than single-admin hobby servers.
5. Initiate account migration
   Once your new account is set up, go to Preferences > Account > Move from a different account. Enter your old account’s username and the old instance domain. Mastodon will send a verification token to the email address associated with your old account. If the old instance is offline, you cannot complete this step — you must wait until the instance comes back or you lose the ability to migrate followers. If the old instance is accessible, follow the on-screen instructions to transfer your followers and follows to the new account.
6. Redirect your old profile
   On the old instance, go to Preferences > Account > Account settings and set your new account’s profile URL in the “Move to a different account” field. This tells Mastodon servers that your old profile now points to your new one, so anyone who still follows your old account will be redirected automatically. If the old instance is offline, this step cannot be performed.

Common Mistakes When Reacting to an Instance Outage

Waiting too long assuming the admin will fix it

Many users assume the admin is aware of the outage and will restore the server within a few hours. In reality, many instances that go offline never come back. If the instance has been down for more than 24 hours and you have no communication from the admin, treat the instance as permanently lost. Start the migration process immediately.

Not exporting data before the instance disappears

The most common regret is failing to export the account archive. Mastodon does not back up user data automatically. If the server’s hard drive fails or the hosting account is deleted, your posts, images, and lists are gone forever. Export your archive at least once a month as a routine practice, not just during an outage.

Choosing a new instance that is also unstable

After an outage, users often rush to the first alternative instance they find. This can lead to a repeat of the same problem. Before joining a new instance, check its uptime history on instances.social or fediverse.observer. Look for instances that have been running for more than two years, have multiple admins, and show consistent posting activity from the admin account.

Attempting to log in during a partial outage

When an instance is partially online, parts of the interface may load but login may fail. Repeated login attempts can lock your account temporarily due to rate limiting. If you cannot log in after two tries, stop and use the password reset feature instead. If the reset email never arrives, the server’s mail system is likely also down.

Mastodon Instance Health Check Methods

Method	What It Checks	Best For
Browser URL load	Basic HTTP response	Quick first check
Down For Everyone Or Just Me	Global vs local outage	Confirming the problem is not your network
Instance /health endpoint	Server-side JSON status	Detailed server health info
UptimeRobot	Automated 5-minute checks	Ongoing monitoring
instances.social	Community-reported uptime	Historical reliability data

You now have a clear action plan for detecting and reacting to a Mastodon instance going offline. Start by setting up an UptimeRobot monitor for your instance’s URL so you receive alerts before the outage becomes critical. Export your account archive monthly using Preferences > Import and Export > Export to safeguard your content. If you ever need to migrate, use the Move to a different account feature to transfer followers without losing your community. The most resilient Mastodon users are those who back up early and move fast.