程序被 kill 掉，但是并不知道为什么被 kill，
是因为内存原因还是其他原因，通过 dmesg 命令可以确认这部分信息。
dmesg 命令的三种用法
三种方法都能得到 kill 信息，推荐使用第一种。
1
2
3
4
5
6
7
| # 有详细的时间
grep -i 'killed process' /var/log/messages
# 查询结果
[root@cdh12 ~]# grep -i 'killed process' /var/log/messages
Jun 17 04:12:20 cdh12 kernel: Killed process 4513 (palo_be), UID 0, total-vm:52091820kB, anon-rss:40273088kB, file-rss:0kB, shmem-rss:0kB
Jun 19 04:30:45 cdh12 kernel: Killed process 20561 (palo_be), UID 0, total-vm:51151564kB, anon-rss:41944112kB, file-rss:0kB, shmem-rss:0kB
|
1
2
3
4
5
6
7
| # 没有详细的时间
dmesg | grep -i 'killed process'
# 查询结果
[root@cdh12 ~]# dmesg | grep -i 'killed process'
[10257349.805906] Killed process 4513 (palo_be), UID 0, total-vm:52091820kB, anon-rss:40273088kB, file-rss:0kB, shmem-rss:0kB
[10431255.475118] Killed process 20561 (palo_be), UID 0, total-vm:51151564kB, anon-rss:41944112kB, file-rss:0kB, shmem-rss:0kB
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
| # 有详细的时间和更多的信息
journalctl -xb | egrep -i 'killed process' -C 5
# 得到的信息比较多，但是感觉没什么用
--
Jun 19 04:30:45 cdh12 kernel: [ 5108] 0 5108 28321 134 14 0 0 sqoop
Jun 19 04:30:45 cdh12 kernel: [ 5154] 0 5154 28321 91 11 0 0 sqoop
Jun 19 04:30:45 cdh12 kernel: [ 5155] 0 5155 28356 209 13 0 0 bash
Jun 19 04:30:45 cdh12 kernel: [ 5179] 0 5179 28359 131 13 0 0 bash
Jun 19 04:30:45 cdh12 kernel: Out of memory: Kill process 20561 (palo_be) score 629 or sacrifice child
Jun 19 04:30:45 cdh12 kernel: Killed process 20561 (palo_be), UID 0, total-vm:51151564kB, anon-rss:41944112kB, file-rss:0kB, shmem-rss:0kB
Jun 19 04:30:49 cdh12 systemd-logind[778]: Removed session 20698.
-- Subject: Session 20698 has been terminated
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- Documentation: http://www.freedesktop.org/wiki/Software/systemd/multiseat
|
个人感觉 anon-rss 比较重要（实际占用的物理内存）
1
2
3
| total-vm：进程总共使用的虚拟内存
anon-rss：虚拟内存实际占用的物理内存
file-rss：虚拟内存实际占用的磁盘空间
|