On Tuesday, November 7th, at 5:26 PM UTC we received internal alerts indicating increased latency in retrieving profile data. Users were experiencing longer load times when viewing tasks of some limited number of profiles.
Around 6:30 PM UTC, The alert stopped, since users stopped viewing the affected profiles. This was a temporary issue due on a subset of very large profiles, so no immediate action was taken. Actions to resolve this issue long term are being investigated.
Full service was restored around 6:50 PM UTC.