Mandrill

We’re also committed to a full post-event review. We plan to learn as much as we can from this so that we can make improvements and better serve our customers. We sincerely apologize for the pain this has caused.

We’re committed to compensating our customers for the disruption this outage has caused. Once we’ve fully resolved all issues, we’ll let you know more about how and when we plan to issue refunds or credits.

Users may see lingering effects. Stats and metrics that should have been collected during the outage may be incorrect or missing entirely. Most customers should now start to see features working as expected.

We determined that some data was preventing necessary automatic processes from running. We deleted that data, which allowed those processes to successfully run and return the database to a usable state.

The outage began around 10:30pm EST on Sunday, February 3. We tried to restore a damaged database while preserving as much data as possible, but we changed direction to get back online faster.

We’ve emailed our customers, and have more details about the outage and our efforts to resolve it. Thread below.

Our Engineering team has resolved the outage. We’re now working to get cleanup efforts underway. We’re so sorry for the disruption.

Our Engineers are still working hard to resolve outages. Updates will be posted here.

We know this is painful—we're working hard to resolve this outage. We’ll keep providing updates here.

We’re very sorry for the disruption to your business. We plan to offer refunds to all affected customers. We hope to have more information and a resolution soon.

The database is large and this process is time consuming. Mandrill users could be impacted in a variety of ways including tracking, bounces, email sends, inbound email, and webhook events. We don’t have an estimated time frame but will post regular updates on our progress.

A significant spike in writes to one of our five main Postgres datastore instances triggered a Transaction ID Wraparound issue. When this occurs, the database sets itself in read-only mode until offline maintenance (vacuuming) can occur and activity is completely halted.

We just emailed Mandrill users about an ongoing outage. The thread below includes more details and next steps.

We're still working to resolve application issues, and active users will receive a detailed update via email today.

Update: Most outbound emails are sending, but inbound emails are still not coming through. We're also seeing ongoing errors with scheduled mail and webhooks. We're sorry for the disruption and are doing our best to resolve these issues quickly.

Our team is still working to resolve sending and related issues, now expected to affect most users. We’ll share further updates with customers by email. We apologize for the inconvenience this has caused.

We’re actively working to resolve sending issues affecting ~3,000 users. At this time, we do not have an estimated time of resolution. We know this is a disruption and impacts your business. We sincerely apologize and we’ll share updates here.

Our Engineering team is working as quickly as possible to resolve issues impacting both inbound and outbound emails. We apologize for the disruption and we’ll continue to share updates here.

Our Engineering team is focusing on the issues impacting inbound emails. Viewing the content of sent emails may result in errors. We'll post further updates here.

We're still investigating errors with email sending. We'll post news here - thanks for your patience.