All sites are back online. We're continuing to monitor the situation.
Posted Dec 04, 2018 - 10:55 UTC
Update
We're 75% done with the downgrade of all compute nodes. We expect the operation to take about 5-10 minutes more.
Posted Dec 04, 2018 - 10:48 UTC
Identified
We identified the issue. We'll roll back a part of the upgraded packages which have been rolled out during maintenance. First tests show that this solves the issue.
Posted Dec 04, 2018 - 10:26 UTC
Update
A reboot of the compute node didn't solve the issue. We'll be rolling back the update that has been done during the regular maintenance. We'll update this incident as we progress.
Posted Dec 04, 2018 - 10:20 UTC
Update
We are continuing to investigate this issue.
Posted Dec 04, 2018 - 09:55 UTC
Investigating
During regular maintenance we saw that a portion of the containers don't start correctly. We're looking into this issue.