各路大神好
是这样的
公司有一台只做版控的server
因zabbix侦测该server process太多
一天到晚告警让主管觉得很烦
进来看了之后发现有这个东西
而且他的进程有慢慢变多的迹象
ps axjf看起来可疑的部分如下
\_ runsv postgresql
| \_ svlogd -tt /var/log/gitlab/postgresql
| \_ /opt/gitlab/embedded/bin/postgres -D /var/opt/gitlab/postgresql/data
| \_ postgres: checkpointer process
| \_ postgres: writer process
| \_ postgres: wal writer process
| \_ postgres: autovacuum launcher process
| \_ postgres: stats collector process
| \_ postgres: gitlab-psql postgres [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
| \_ postgres: gitlab gitlabhq_production [local] idle
查了官方手册只提到
postgres
-D datadir
Specifies the file system location of the database configuration files. See Se
ction 19.2 for details.
请问这些闲置进程可以直接kill掉吗
或是gitlab server本身就有常常重开的需要吗？

那是DB(postgresql) 你怎么会想要砍掉他 = =觉得zabbix很烦，就通通mute 不要发通知啊(误

你的 kernel 也常常 idle 要不要关机算了 (x

那你要去看gitlab的source code了... 还是用我说的那招让zabbix 变哑巴就好了....

一楼别这样

如果 Postgersql 是独立的一台，把 TCP keepalive设短一点

你有没有想过也许是zabbix告警值设太低呢?

具体来说，他阀值设了多少呢？然后你目前实际有多少process?