memory slab é muito grande

1

Estou executando o servidor para testar o WAS no contêiner. O servidor tem 40 núcleos de CPU e 96 GB de memória. Eu corri o WAS em contêineres por vários dias e descobri tamanho de memória muito grande. Então parou tudo que era. Mas a laje não é reduzida. Abaixo está o estado da memória deste servidor.

root@G14-1:~# cat /proc/meminfo
MemTotal:       98767876 kB
MemFree:          480336 kB
MemAvailable:   24268044 kB
Buffers:            1784 kB
Cached:             8584 kB
SwapCached:         2256 kB
Active:             3852 kB
Inactive:          10264 kB
Active(anon):       1988 kB
Inactive(anon):     7336 kB
Active(file):       1864 kB
Inactive(file):     2928 kB
Unevictable:        5308 kB
Mlocked:            5308 kB
SwapTotal:      100438012 kB
SwapFree:       99383076 kB
Dirty:                 0 kB
Writeback:            32 kB
AnonPages:          9088 kB
Mapped:             7468 kB
Shmem:              1208 kB
Slab:           96499028 kB
SReclaimable:   24462728 kB
SUnreclaim:     72036300 kB
KernelStack:        9056 kB
PageTables:         9656 kB
NFS_Unstable:          0 kB
Bounce:                0 kB
WritebackTmp:          0 kB
CommitLimit:    149821948 kB
Committed_AS:    1609444 kB
VmallocTotal:   34359738367 kB
VmallocUsed:           0 kB
VmallocChunk:          0 kB
HardwareCorrupted:     0 kB
AnonHugePages:         0 kB
ShmemHugePages:        0 kB
ShmemPmdMapped:        0 kB
CmaTotal:              0 kB
CmaFree:               0 kB
HugePages_Total:       0
HugePages_Free:        0
HugePages_Rsvd:        0
HugePages_Surp:        0
Hugepagesize:       2048 kB
DirectMap4k:    33518292 kB
DirectMap2M:    66918400 kB
DirectMap1G:     2097152 kB

slabtop

Active / Total Objects (% used)    : 314177865 / 316372600 (99.3%)
 Active / Total Slabs (% used)      : 6945664 / 6945664 (100.0%)
 Active / Total Caches (% used)     : 97 / 138 (70.3%)
 Active / Total Size (% used)       : 94206435.41K / 94565538.08K (99.6%)
 Minimum / Average / Maximum Object : 0.01K / 0.30K / 18.69K

  OBJS ACTIVE  USE OBJ SIZE  SLABS OBJ/SLAB CACHE SIZE NAME
59191120 59188777   0%    0.20K 1479869       40  11838952K vm_area_struct
1237149 1235053   0%    5.81K 252301        5   8073632K task_struct
11548224 11541173   0%    0.59K 218000       54   6976000K inode_cache
3071838 3071813   0%    2.06K 207984       15   6655488K mm_struct
9656484 9654811   0%    0.66K 206299       48   6601568K proc_inode_cache
6502404 6502396   0%    1.00K 205455       32   6574560K signal_cache
3045933 3045926   0%    2.06K 205165       15   6565280K sighand_cache
9319002 9317989   0%    0.69K 204555       46   6545760K sock_inode_cache
9330629 9330629 100%    0.69K 204199       46   6534368K files_cache
6354064 6301629   0%    1.00K 200073       32   6402336K kmalloc-1024
66215808 66213574   0%    0.06K 1034622       64   4138488K anon_vma_chain
5959856 5914764   0%    0.50K 186475       32   2983600K kmalloc-512
31272282 31270906   0%    0.08K 613182       51   2452728K anon_vma
12527025 12506741   0%    0.19K 298977       42   2391816K dentry
12046041 12045609   0%    0.19K 287004       42   2296032K cred_jar
1063926 1052915   0%    2.00K  67884       16   2172288K kmalloc-2048
8383326 8297280   0%    0.19K 199625       42   1597000K kmalloc-192
5566224 5521817   0%    0.25K 174044       32   1392352K kmalloc-256
1052868 1049607   0%    0.94K  31662       34   1013184K xfs_inode
6536288 6536288 100%    0.12K 204259       32    817036K pid
1251180 1244965   0%    0.57K  44685       28    714960K radix_tree_node
5491408 5058053   0%    0.12K 161512       34    646048K kernfs_node_cache
541320 539776   0%    1.05K  19493       30    623776K ext4_inode_cache
31706624 31685396   0%    0.02K 123854      256    495416K kmalloc-16
479202 475497   0%    0.69K  10544       46    337408K shmem_inode_cache
2166016 1678856   0%    0.06K  33844       64    135376K kmalloc-64
345324 143718   0%    0.38K   8222       42    131552K kmem_cache
2381696 2060487   0%    0.03K  18607      128     74428K kmalloc-32
682836 674307   0%    0.09K  16258       42     65032K kmalloc-96
658944 285355   0%    0.06K  10296       64     41184K kmem_cache_node
  2874   2698   0%    8.00K    723        4     23136K kmalloc-8192
  4491   4307   0%    4.00K    571        8     18272K kmalloc-4096
131136  78644   0%    0.12K   4098       32     16392K kmalloc-128
160384 160384 100%    0.07K   2864       56     11456K Acpi-Operand
 10793  10537   0%    0.75K    316       42     10112K au_icntnr
  6722   6484   0%    0.94K    203       34      6496K RAW
 23320  23188   0%    0.18K    588       44      4704K ip4-frags
  9954   9834   0%    0.31K    290       51      4640K bio-4
  3640   3360   0%    1.12K    130       28      4160K RAWv6
 15551  15477   0%    0.21K    504       37      4032K bsg_cmd
 18144  17790   0%    0.16K    504       48      4032K nfsd4_stateids
  4719   4719 100%    0.81K    121       39      3872K fuse_inode
  6468   6468 100%    0.56K    231       28      3696K task_group
  8842   8721   0%    0.38K    225       42      3600K mnt_cache
  1472   1440   0%    2.00K     92       16      2944K TCP
 46976  46976 100%    0.06K    734       64      2936K ext4_io_end
 14742  14742 100%    0.19K    351       42      2808K au_finfo
  6664   6664 100%    0.32K    136       49      2176K taskstats
 37303  36706   0%    0.05K    511       73      2044K Acpi-Parse
 14016  13762   0%    0.12K    438       32      1752K au_dinfo
    53     53 100%   18.69K     53        1      1696K kvm_vcpu
 16029  15953   0%    0.10K    411       39      1644K buffer_head
  5646   5646 100%    0.24K    202       33      1616K xfs_buf_item
  3510   3510 100%    0.41K     90       39      1440K btrfs_ordered_extent
  1394   1394 100%    0.94K     41       34      1312K mqueue_inode_cache
   200    200 100%    5.94K     40        5      1280K net_namespace
   560    560 100%    2.18K     40       14      1280K pid_namespace
   560    560 100%    2.19K     40       14      1280K TCPv6

free -m -h

root@G14-1:~# free -m -h
              total        used        free      shared  buff/cache   available
Mem:            94G         70G        456M         32K         23G         23G
Swap:           95G        1.0G         94G

htop

   PID USER      PRI  NI  VIRT   RES   SHR S CPU% MEM%   TIME+  Command
   2045 root       10 -10 25784  5164  3936 S  0.0  0.0  0:00.00 /sbin/iscsid
2010867 root       20   0 26488  3672  2120 R  1.6  0.0  0:02.67 htop
2009930 root       20   0  105M  2300  2260 S  0.0  0.0  0:00.54 sshd: root@pts/2
      1 root       20   0  240M  2844  2516 S  0.0  0.0  9:16.57 /sbin/init
2008253 root       20   0  105M  2024  2024 S  0.0  0.0  0:02.44 sshd: root
2005975 root       20   0  105M  1996  1996 S  0.0  0.0  0:00.82 sshd: root
2006164 root       20   0  105M  1972  1972 S  0.0  0.0  0:05.95 sshd: root
   1783 root       20   0 29952  2108  2108 S  0.0  0.0  0:04.36 /usr/sbin/cron -f
2006718 root       20   0  105M  2180  2180 S  0.0  0.0  0:19.69 sshd: root@pts/4
2009477 root       20   0  105M  2176  2172 S  0.0  0.0  0:11.01 sshd: root@pts/5
2005562 root       20   0  105M  2336  2336 S  0.0  0.0  0:04.28 sshd: root
   1420 root       20   0 52172  2360  2300 S  0.0  0.0  0:03.76 /sbin/rpcbind -f -w
2008894 root       20   0  105M  2004  2004 S  0.0  0.0  0:00.74 sshd: root
2005805 root       20   0  105M  2276  2276 S  0.0  0.0  0:00.78 sshd: root
2009025 root       20   0  105M  2116  2116 S  0.0  0.0  0:12.04 sshd: root@pts/1
2009791 root       20   0  105M  2116  2116 S  0.0  0.0  0:00.14 sshd: root@pts/0
  32018 root       20   0  195M  2108  2108 S  0.0  

Como resultado do slabtop, vm_area_struct e task_struct são muito grandes. Mas não há muitos processos neste servidor agora. Max PID é 3303593, mas o número de processos é de cerca de 481. Eu não posso adivinhar isso Por que isso aconteceu?

    
por Supsupi 30.08.2018 / 12:24

0 respostas