On 29th September, Tink experienced a partial outage between 17:45 and 18:50 (CEST) affecting the availability of the majority of our services and products. During this time most end-users were unable to authenticate, aggregate accounts or initiate payments.
An auto-scaling issue in one of our services interrupted the on-demand refresh operation. This led to the saturation of the server resources and as a result credentials stopped being updated.
Affected services were restarted at 18:50 (CEST) after which authentication flow components returned to stability.
As a consequence of this incident we will focus on the following actions over the coming weeks to prevent a similar scenario from occurring again: