The issue turned out to be much more complicated than we first thought. To solve the root cause a fix in the Linux kernel needs to be implemented. Today we added a workaround that will fix the issue with faulty billing until the final fix is in place.
We are still working on a fix for this problem. It was not as simple as we hoped and at the moment we do not have an ETA.
This problem only exists on the regions "SE STO1 (new)" and "SE STO2 (new)". Other regions are not affected.
In some cases, we've seen containers using abnormal amount of RAM. We are in contact with Jelastic regarding this problem and they are working on a fix.
If you suspect that your container is affected, please contact us through the support portal or e-mail email@example.com and we'll investigate if you are affected by this issue.
If your environment is affected you'll get compensation for the over-commitment, once the fix is installed.