Am 09.09.19 um 13:49 schrieb Vlastimil Babka:
> On 9/9/19 10:54 AM, Stefan Priebe - Profihost AG wrote:
>>> Do you have more snapshots of /proc/vmstat as suggested by Vlastimil and
>>> me earlier in this thread? Seeing the overall progress would tell us
>>> much more than before and after. Or have I missed this data?
>>
>> I needed to wait until today to grab again such a situation but from
>> what i know it is very clear that MemFree is low and than the kernel
>> starts to drop the chaches.
>>
>> Attached you'll find two log files.
>
> Thanks, what about my other requests/suggestions from earlier?
Sorry i missed your email.
> 1. How does /proc/pagetypeinfo look like?
# cat /proc/pagetypeinfo
Page block order: 9
Pages per block: 512
Free pages count per migrate type at order 0 1 2 3
4 5 6 7 8 9 10
Node 0, zone DMA, type Unmovable 1 0 0 1
2 1 1 0 1 0 0
Node 0, zone DMA, type Movable 0 0 0 0
0 0 0 0 0 1 3
Node 0, zone DMA, type Reclaimable 0 0 0 0
0 0 0 0 0 0 0
Node 0, zone DMA, type HighAtomic 0 0 0 0
0 0 0 0 0 0 0
Node 0, zone DMA, type Isolate 0 0 0 0
0 0 0 0 0 0 0
Node 0, zone DMA32, type Unmovable 1141 970 903 628
302 106 27 4 0 0 0
Node 0, zone DMA32, type Movable 274 269 368 396
342 265 214 178 113 12 13
Node 0, zone DMA32, type Reclaimable 81 57 134 114
60 50 25 4 2 0 0
Node 0, zone DMA32, type HighAtomic 0 0 0 0
0 0 0 0 0 0 0
Node 0, zone DMA32, type Isolate 0 0 0 0
0 0 0 0 0 0 0
Node 0, zone Normal, type Unmovable 39 36 13257 3474
1333 317 42 0 0 0 0
Node 0, zone Normal, type Movable 1087 9678 1104 4250
2391 1946 1768 691 141 0 0
Node 0, zone Normal, type Reclaimable 1 1782 1153 2455
1927 986 330 7 2 0 0
Node 0, zone Normal, type HighAtomic 1 1 2 2
2 0 1 1 1 0 0
Node 0, zone Normal, type Isolate 0 0 0 0
0 0 0 0 0 0 0
Number of blocks type Unmovable Movable Reclaimable
HighAtomic Isolate
Node 0, zone DMA 1 7 0
0 0
Node 0, zone DMA32 52 1461 15
0 0
Node 0, zone Normal 824 5448 383
1 0
> 2. Could you also try if the bad trend stops after you execute:
> echo never > /sys/kernel/mm/transparent_hugepage/defrag
> and report the result?
it's pretty difficult to catch those moments. Is it OK so set the value
now and monitor if it happens again?
Just to let you know:
I've now also some more servers where memfree show 10-20Gb but cache
drops suddently and memory PSI raises.
Greets,
Stefan
