Hi
We are missing some features or are not satisfied with existing when dealing with some of the operational aspects of running event store:
-
What are the size of my streams
-
What are my streams (accessible through the streams projection, but should be query-able…)
-
Scavenging time … e.g. if we put too much data into event store, will we even be able to finish a scavenge…
-
Long startup time of nodes… E.g. if we lose two nodes in a three nodes cluster, event store can be down for say +30 mins (e.g. for a 1 TB cluster running on our current on-premise vm’s)
-
We had an incident with a corrupt db which has scared us… (A lot of chunks were missing from the data folder on two nodes, we are not sure whether it’s even possible that event store could be responsible for this…)
-
The issue with the file cache not being released properly when running under windows. We know this can be fixed with a scheduled task that sets the max allowed value, but it’s just one of those things that makes the system look bad…
All comments are welcome and thanks for an otherwise great open source product! 