Am 09.09.19 um 14:49 schrieb Michal Hocko: > On Mon 09-09-19 14:37:52, Stefan Priebe - Profihost AG wrote: >> >> Am 09.09.19 um 14:28 schrieb Michal Hocko: >>> On Mon 09-09-19 14:10:02, Stefan Priebe - Profihost AG wrote: >>>> >>>> Am 09.09.19 um 14:08 schrieb Michal Hocko: >>>>> On Mon 09-09-19 13:01:36, Michal Hocko wrote: >>>>>> and that matches moments when we reclaimed memory. There seems to be a >>>>>> steady THP allocations flow so maybe this is a source of the direct >>>>>> reclaim? >>>>> >>>>> I was thinking about this some more and THP being a source of reclaim >>>>> sounds quite unlikely. At least in a default configuration because we >>>>> shouldn't do anything expensinve in the #PF path. But there might be a >>>>> difference source of high order (!costly) allocations. Could you check >>>>> how many allocation requests like that you have on your system? I've another system which might be interesting. Not sure which stuff to gather. It never builds up any read cache cause memory is constantly under pressure. But memfree is 28G. What would be interesting to collect here? Pressure is not very high just 1-3% but it seems it prevents the system from building up file cache. Mostly at the night where no pressure is it starts building up a read cache until pressure happens again. But all this happens with MemFree at nearly 30GB of memory. Greets, Stefan
