RemoveIPC=yes 的RHEL7.2 會crash掉Oracle asm 實例和Oracle database實例

一個在centos7.2上的11204庫的實例crash了，檢查alert發現如下：

Wed Sep 21 00:00:11 2016
Errors in file /apps/oracle/diag/rdbms/xberp/xberp/trace/xberp_ckpt_17537.trc:
ORA-27157: OS post/wait facility removed
ORA-27300: OS system dependent operation:semop failed with status: 43
ORA-27301: OS failure message: Identifier removed
ORA-27302: failure occurred at: sskgpwwait1
Wed Sep 21 00:00:11 2016
Errors in file /apps/oracle/diag/rdbms/xberp/xberp/trace/xberp_dbrm_17523.trc:
ORA-27157: OS post/wait facility removed
ORA-27300: OS system dependent operation:semop failed with status: 43
ORA-27301: OS failure message: Identifier removed
ORA-27302: failure occurred at: sskgpwwait1
Wed Sep 21 00:00:11 2016
Errors in file /apps/oracle/diag/rdbms/xberp/xberp/trace/xberp_mman_17527.trc:
ORA-27157: OS post/wait facility removed
ORA-27300: OS system dependent operation:semop failed with status: 43
ORA-27301: OS failure message: Identifier removed
ORA-27302: failure occurred at: sskgpwwait1
Wed Sep 21 00:00:11 2016
Errors in file /apps/oracle/diag/rdbms/xberp/xberp/trace/xberp_j002_14570.trc:
ORA-27157: OS post/wait facility removed
ORA-27300: OS system dependent operation:semop failed with status: 43
ORA-27301: OS failure message: Identifier removed
ORA-27302: failure occurred at: sskgpwwait1
CKPT (ospid: 17537): terminating the instance due to error 27157
Wed Sep 21 00:00:11 2016
Errors in file /apps/oracle/diag/rdbms/xberp/xberp/trace/xberp_j000_14566.trc:
ORA-27157: OS post/wait facility removed
ORA-27300: OS system dependent operation:semop failed with status: 43
ORA-27301: OS failure message: Identifier removed
ORA-27302: failure occurred at: sskgpwwait1
Instance terminated by CKPT, pid = 17537
Errors in file /apps/oracle/diag/rdbms/xberp/xberp/trace/xberp_ckpt_17537.trc:
ORA-27300: OS system dependent operation:semctl failed with status: 22
ORA-27301: OS failure message: Invalid argument
ORA-27302: failure occurred at: sskgpwrm1
ORA-27157: OS post/wait facility removed
ORA-27300: OS system dependent operation:semop failed with status: 43
ORA-27301: OS failure message: Identifier removed
ORA-27302: failure occurred at: sskgpwwait1
Wed Sep 21 10:20:34 2016

描述：

在centos7.2中，systemd-logind 服務引入了一個新特性，該新特性是：當一個user 完全退出os之后，remove掉所有的IPC objects。

該特性由/etc/systemd/logind.conf參數文件中RemoveIPC選項來控制。

在centos7.2中，RemoveIPC的默認值為yes

因此，當最后一個oracle 或者Grid用戶退出時，操作系統會remove 掉這個user的shared memory segments and semaphores

由于Oracle ASM 和database 使用 shared memory segments ，remove shared memory segments將會crash掉Oracle ASM and database  instances.

Oracle Linux 7.2 通過在/etc/systemd/logind.conf配置文件中明確設置RemoveIPC為no，Oracle Linux7.2 避免了該問題，

變通的解決方法：

1) Set RemoveIPC=no in /etc/systemd/logind.conf

2) Reboot the server or restart systemd-logind as follows:

    # systemctl daemon-reload

    # systemctl restart systemd-logind

這些錯誤跟os有關，在mos上有文檔可能參考：

ALERT: Setting RemoveIPC=yes on Redhat 7.2 Crashes ASM and Database Instances as Well as Any Application That Uses a Shared Memory Segment (SHM) or Semaphores (SEM) (文檔 ID 2081410.1)