On 28th April, 2019 at 8.11am UTC our engineers identified an issue with Recording searches, deletes, and updates not working properly.
Due to connection issues a number of queries to the datastore for Recordings failed and would not have executed. This means that a proportion of all searches in Recordings, as well as actions such as favoriting, would fail to complete.
The incident was resolved at 8.24am UTC.
Why did this issue occur?
Hotjar engineers made an infrastructure update to upgrade the security protocol of how we connect to the datastore. This caused a temporary issue with current connections, which had to be terminated and reconnected.
What will we do to prevent this from happening in the future?
We will hold an internal postmortem and update this report later.