Hi, Josh

Appreciate for your advice.
I will try either both of those and will try to fix this issue.

Regards,
Kentaro

From: nginx [mailto:nginx-bounces@nginx.org] On Behalf Of Josh Cooley
Sent: Tuesday, August 28, 2018 2:02 PM
To: nginx@nginx.org
Cc: 小栗　雅士; 千葉　幸宏
Subject: Re: [nginx]access log and error log

Your postrotate script contains single ticks (') instead of backticks (`), but you should really use the $( command ) construct if your /bin/sh is bash.

INCORRECT (what you have now):
if [ -f /var/run/nginx.pid ]; then
kill -USR1 'cat /var/run/nginx.pid'
fi

CORRECT, BUT ARCHAIC:
if [ -f /var/run/nginx.pid ]; then
kill -USR1 `cat /var/run/nginx.pid`
fi

CORRECT AND MODERN:
if [ -f /var/run/nginx.pid ]; then
kill -USR1 $(cat /var/run/nginx.pid)
fi


On Mon, Aug 27, 2018 at 7:49 PM, 桐山　健太郎 <kiriyama.kentaro@tis.co.jp<mailto:kiriyama.kentaro@tis.co.jp>> wrote:
Hello,

I’m suffering a problem with access.log and error.log for nginx.service on RHEL 7.4.

The problem is either access.log and error.log are not recording the log.

To solve this issue, I have tried multiple time changing logrotation setting for nginx.

Below is the setting of logrotation for nginx.service.(/etc/logrotate.d/nginx)

/var/log/nginx/*.log {
daily
missingok
rotate 7
compress
create 644 nginx adm
sharedscripts
postrotate
if [ -f /var/run/nginx.pid ]; then
kill -USR1 'cat /var/run/nginx.pid'
fi
endscript
}

By the way, the log will start recording on both log files after restarting the nginx.service.

I would like to know the issue on here and how to solve it.

Regards,
Kentaro


_______________________________________________
nginx mailing list
nginx@nginx.org<mailto:nginx@nginx.org>
http://mailman.nginx.org/mailman/listinfo/nginx

_______________________________________________
nginx mailing list
nginx@nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx