I have about 25000 failed writes within each 30 minutes. Previously I had an issue with timestamps ahead of current server clock, but this was addressed. When looking at collected data, it is OK - no gaps all values are good.
Is there any way I can tell why those writes fail?