在所有RHEL EC2实例上使用100%CPU利用率的Rsyslogd
自过去两天以来,RSyslogd在我的环境中的所有RHEL EC2实例上使用100%CPU利用率。我停下来开始RSYSLOG服务,但仍然存在发行。 这是我们第一次在多个服务器中面对这种行为。 所有服务器中都有足够的磁盘空间/内存。 我检查了内核日志(/var/log/log/bernel)和服务器消息(/var/log/log sagess),但找不到任何有用的信息。
以下是所有服务器的OS和内核版本。
操作系统:Red Hat Enterprise Linux Server 7.2(Maipo) 内核:Linux 3.10.0-1160.42.2.2.2.2.2.x86_64
可以有人建议。
Since past two days, rsyslogd is using 100% CPU Utilization on all RHEL EC2 Instances in my environment. I Stopped and started rsyslog service but still issue persists.
This is first time we are facing this kind of behaviour in multiple servers.
There is sufficient disk space/memory exists in all servers.
I checked kernel logs (/var/log/kernel) and server messages( /var/log/messages),but not find any useful info.
Following is the OS and Kernal versions of all servers.
Operating System: Red Hat Enterprise Linux Server 7.2 (Maipo)
Kernel: Linux 3.10.0-1160.42.2.el7.x86_64
Can someone please suggest on this.
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(2)
问题是由于RSYSLOG的证书到期。观察到了所有客户端的系统消息中的许多连接重试错误。 CPU在RSYSlog的续签证书下降。
Issue is due to cert expiry for rsyslog. Observed a lot of connection retry errors in system messages for all clients. Post renewing certs in rsyslog, cpu went down.
我也有类似的问题,在我的情况下,理由不是证书,而是一些日志写作问题。首先找到原因很重要。首先查看JournalCtl日志
,默认情况下可能会显示太多条目,因此按时间过滤可能会有所帮助。将sudo或登录作为根,因为系统日志等将不会显示。日期格式为yyyy-mm-dd HH:MM:SS
一些有用的CMDS
I had a similar problem and reason in my case was not the certs but some log writing issue. It is important to first find the reason. Start by looking at journalctl logs
By default it might show too many entries so filtering by time can be helpful. Use sudo or login as root as else system logs etc would not show up. The date format is YYYY-MM-DD HH:MM:SS
Some useful cmds