No notices reported this month
Relevant metrics have remained normal for three full days and we consider the incident resolved.
The root cause was a recent minor version library upgrade that included performance enhancements, but also an unexpected side effect on how we handled reconnections, particularly when a computer wakes up from sleep. Our team was able to adjust for the upgrade change, so reconnections would again be immediate, resulting in faster post saving like before.
We have observed the expected subsequent drop in the slow save warning events and will continue monitoring this and related metrics.
The fix has been deployed and we are currently monitoring the result.
The potential fix has been tested on our staging environment, and the fix is now awaiting deployment.
A potential fix has been implemented.
A potential root cause has been identified.
The post should eventually save and there is no system degradation or increased failure to save. We are investigating why our application client is not attempting to save immediately or sooner when it could and should.
Dec 2023 to Feb 2024