First of all, we are very sorry for the inconvenience that these disturbances may have caused you. Here is a brief description of what happened, what caused it and what we are doing to prevent this from happening again.
The outages that we experienced this week caused some panels and communities to go offline. This happened on the 5th of February and again on the 6th.
After analysing the logs, we found that the caching put in place did not work as expected which caused several applications to stop and not being able to restart themselves.
As as result of this, we have two fixes that we will start rolling out next week.
The first fix deals with the way our cache is setup and will ensure that when items are removed from the cache will not cause the applications to recycle.
The second fix address caching on a back-end level and involves several code changes. This fix is currently going through testing and we expect this to be ready for deployment next week.