messages日志看只出现6次，alert无对应报错，应该不是报错信息。

提供相应时间段的ohasd.trc文件看看。

Starting of Grid Infrastructure (GI) 12c with the text "Starting execution of Oracle Clusterware init.ohasd" logs to syslog as an error and not as an information

Following logs should be lowered to info instead of err priority:

[Thu Aug 31 18:12:16][13844][root@<nodename1>:~][0]# grep init.ohasd /var/log/messages
May 20 14:15:18 rachost1 logger: Starting execution of Oracle Clusterware init.ohasd
May 21 02:03:28 rachost1 logger: Starting execution of Oracle Clusterware init.ohasd
May 21 05:38:44 rachost1 logger: Starting execution of Oracle Clusterware init.ohasd
May 21 06:22:17 rachost1 logger: Starting execution of Oracle Clusterware init.ohasd

The impact is, monitoring tools scanning for errors in syslog, give false alerts and tickets when a normal GI startup happens.

CHANGES

CAUSE

The monitoring tool checks the content of /var/log/messages_err. This file results of this line in the rsyslogd configuration:

*.error /var/log/messages_err

ohasd and init.crs utilizes p.err for informational messages. This is reported through bug 26738408, which is closed as duplicate of 26733329

SOLUTION

Bug 26733329 fixed in future release. Request/download and apply one-off patch if the system is impacted