We have monitoring the MDS service and have not seen any additional issues during the day.
The incident that started last night was triggered by a Java out of memory problem, an error that we have seen rarely in the past (1-2x times a year). We are working on deploying a redundant MDS system, which should prevent these kinds of problems in the future.
Posted over 1 year ago. Jun 29, 2017 - 15:47 UTC
We have now re-established access to the MDS for all our users and we are monitoring the incident.