Find the process ID of the master (parent process), then:

ls -ld /proc/3380

Output looks like:

dr-xr-xr-x 8 root root 0 Nov 12 23:05 /proc/3380

or

ps -eo user,pid,cmd,etime,ppid | grep http

ps -eo user,pid,cmd,etime,ppid | grep service

Show JUST the master process:

ps -eo user,pid,cmd,etime,ppid | grep -v grep | grep http | grep root

Output is simlar:

 3380 nginx: master process /usr/  1-02:50:02
 8543 nginx: worker process          02:50:10
 8544 nginx: worker process          02:50:10
 8545 nginx: worker process          02:50:10
 8546 nginx: worker process          02:50:10
 8547 nginx: worker process          02:50:10
 8548 nginx: worker process          02:50:10
 8549 nginx: worker process          02:50:10
 8550 nginx: worker process          02:50:10

The bit we are interested in is the master process:

 3380 nginx: master process /usr/  1-02:50:02

This represents 1 days, 2 hours, 50 mins and 2 seconds

---

w

uptime

---

CentOS

service httpd fullstatus

Ubuntu

apache2ctl fullstatus

Sometimes the correct packages are not installed to run these commands
Alternative:

grep -i resuming /var/log/httpd/error.log

grep -i resuming /var/log/apache2/error.log

---

You can use any of the following commands

mysqladmin  version | grep -i uptime

or

SHOW GLOBAL STATUS LIKE 'Uptime';

or

mysql -e 'SHOW GLOBAL STATUS LIKE "Uptime"' | grep ''[[:digit:]]'' | awk '{ print $2 / 60 / 60 / 24 " days"}'