We have a strange issue happening with elastic search in that it just stalls and stops accepting request after roughly 3 or so days. To fix the issue we need to restart ES and everything is fine for another 3 days.
At first, I thought it was an issue with memory so increased and reserved 3G for it but that didn’t help. When it stalls there is more than enough memory free in the system.
When checking the processors that are running it say ES is running, so it’s not the system killing the process.
ES Log files have no info related to the issue, it’s like nothing is wrong!
At this point our only option is to just restart the service every night.
Do you think increase the heap size will help?