CentOS丢失的内存
今天nagios运维系统报警，说有一台pc server上的内存使用率飙到60%
ssh上去执行， free -g
total used free shared buffers cached Mem: 31 24 6 0 0 2 -/+ buffers/cache: 21 9 Swap: 0 0 0
used 21, free 9 为什么会这么高呢
TOP 然后M
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 11509 tomcat 20 0 19.7g 3.7g 12m S 2.3 11.9 8:56.05 jsvc 1562 root 20 0 244m 7532 1096 S 0.0 0.0 0:15.62 rsyslogd 1610 haldaemo 20 0 38176 6744 3448 S 0.0 0.0 0:49.03 hald 36266 root 20 0 98292 4076 3088 S 0.0 0.0 0:00.16 sshd 39074 root 20 0 98292 4068 3088 S 0.0 0.0 0:00.17 sshd 38019 root 20 0 98292 4052 3088 S 0.0 0.0 0:00.08 sshd 43550 root 20 0 98292 4048 3080 S 0.0 0.0 0:00.03 sshd 1764 postfix 20 0 81520 3428 2548 S 0.0 0.0 0:04.54 qmgr 1749 root 20 0 81272 3420 2516 S 0.0 0.0 0:22.90 master 42335 postfix 20 0 81352 3380 2508 S 0.0 0.0 0:00.00 pickup 1585 root 20 0 184m 3332 2448 S 0.0 0.0 0:00.01 cupsd 39076 root 20 0 105m 1996 1544 S 0.0 0.0 0:00.09 bash 36268 root 20 0 105m 1988 1540 S 0.0 0.0 0:00.08 bash 38021 root 20 0 105m 1940 1516 S 0.0 0.0 0:00.04 bash
发现根本没有多少地方用到内存， tomcat也就不到4个g
cat /proc/meminfo 看看
MemTotal: 32827220 kB MemFree: 7084524 kB Buffers: 606424 kB Cached: 2468112 kB SwapCached: 0 kB Active: 5700560 kB Inactive: 1282020 kB Active(anon): 3908224 kB Inactive(anon): 16 kB Active(file): 1792336 kB Inactive(file): 1282004 kB Unevictable: 0 kB Mlocked: 0 kB SwapTotal: 0 kB SwapFree: 0 kB Dirty: 16 kB Writeback: 0 kB AnonPages: 3916216 kB Mapped: 21864 kB Shmem: 212 kB Slab: 18510960 kB SReclaimable: 18469956 kB SUnreclaim: 41004 kB KernelStack: 4688 kB PageTables: 11668 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 16413608 kB Committed_AS: 9593256 kB VmallocTotal: 34359738367 kB VmallocUsed: 333876 kB VmallocChunk: 34342501180 kB HardwareCorrupted: 0 kB AnonHugePages: 3833856 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB DirectMap4k: 5056 kB DirectMap2M: 2045952 kB DirectMap1G: 31457280 kB
发现Slab高的离谱。。。。。。
到底是什么样呢， slabtop --once看看
slabtop --once Active / Total Objects (% used) : 225920064 / 226193412 (99.9%) Active / Total Slabs (% used) : 11556364 / 11556415 (100.0%) Active / Total Caches (% used) : 110 / 194 (56.7%) Active / Total Size (% used) : 43278793.73K / 43315465.42K (99.9%) Minimum / Average / Maximum Object : 0.02K / 0.19K / 4096.00K OBJS ACTIVE USE OBJ SIZE SLABS OBJ/SLAB CACHE SIZE NAME 221416340 221416039 3% 0.19K 11070817 20 44283268K dentry 1123443 1122739 99% 0.41K 124827 9 499308K fuse_request 1122320 1122180 99% 0.75K 224464 5 897856K fuse_inode 761539 754272 99% 0.20K 40081 19 160324K vm_area_struct 437858 223259 50% 0.10K 11834 37 47336K buffer_head 353353 347519 98% 0.05K 4589 77 18356K anon_vma_chain 325090 324190 99% 0.06K 5510 59 22040K size-64 146272 145422 99% 0.03K 1306 112 5224K size-32 137625 137614 99% 1.02K 45875 3 183500K nfs_inode_cache 128800 118407 91% 0.04K 1400 92 5600K anon_vma 59101 46853 79% 0.55K 8443 7 33772K radix_tree_node 52620 52009 98% 0.12K 1754 30 7016K size-128 19359 19253 99% 0.14K 717 27 2868K sysfs_dir_cache 10240 7746 75% 0.19K 512 20 2048K filp
尼玛逼，dentry高的跟狗屎一样。
我想是我tomcat大量操作本地文件导致的系统缓存过多
At the default value of vfs_cache_pressure = 100 the kernel will attempt to reclaim dentries and inodes at a “fair” rate with respect to pagecache and swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer to retain dentry and inode caches. Increasing vfs_cache_pressure beyond 100 causes the kernel to prefer to reclaim dentries and inodes.
参考文档：
https://major.io/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/
http://serverfault.com/questions/561350/unusually-high-dentry-cache-usage