Outage update – Nov 10, 2023

I wanted to provide an update on an outage we experienced yesterday. Method users were unable to use the platform starting on Thursday, November 9th, from 9:40 a.m. ET and lasting for approximately 140 minutes.

During this time, it was all hands on deck for our technical team to find the source of the issue and get the platform back to normal. We know how important Method is to you and your team, and maintaining uptime is very important.

Prior to this incident, Method held a 100% uptime over the previous 90 days, something we track at status.method.me. After this outage, that number has fallen to 99.89%. We know that anything less than 100% is not good enough.

What happened

Our post-mortem on yesterday’s events is not yet complete, but I wanted to share this update now. Initially, users were able to access their dashboard, but no apps were loading. The servers that are responsible for delivering the apps to the dashboard were starved of resources.

There are over 70 different types of server configurations in Method’s infrastructure, so I won’t go too deep into a technical explanation. However, I can say that in this specific situation, additional servers would have been redundant and unhelpful — as soon as we added more servers to handle the load, they too became overwhelmed.

The only solution was to find the downstream source and pacify it — once this was done, the Method platform became available again.

What we are doing

We are continuing to diagnose how the servers were starved of their memory. In the meantime, we have isolated the problem source from the rest of the infrastructure to prevent potential issues with the rest of the Method platform. Once we have finished our post-mortem, we’ll put a long-term fix in place to get us back to delivering 100% uptime — and ensure this sequence of events won’t happen again.

We appreciate your patience and understanding when working with our team members. When events like this happen, they share your stress. Rest assured that it has been everyone’s top priority to keep you up and running on Method.

Paul Jackson
Founder & CEO

P.S. What’s an outage vs. an incident?
On status.method.me, you’ll see two types of classifications. An “outage,” which is rare, is when Method is not usable. It’s widespread. An “incident,” on the other hand, is when one portion of the platform may not be fully functional for a group of users, whereas the rest of the platform is still functioning normally.