Using crontab for the first time

[c00kie]

Hi,

I'm struggling to get crontab to run shell scripts. I'm running the csh shell. I've tried both shebangs, csh and sh in my script. # cd

In /root/restart_mysqld.sh:

#!/bin/sh
/usr/local/etc/rc.d/mysql-server restart

And my crontab -e:

5 * * * * sh /root/restart_mysqld.sh

So I'm expecting cron to re-start mysqld every five minutes. It doesn't, it fails to execute. I've checked:

# cat /etc/cron.allow

cat: /etc/cron.allow: No such file or directory

and
# cat /etc/cron.deny

cat: /etc/cron.deny: No such file or directory

Although I'm running cron as root so it shouldn't be a problem. If I run the script directly from the shell:

# sh restart_mysqld.sh

mysql not running? (check /var/db/mysql/my.domain.net.pid).
Starting mysql.

...the script works fine, everything is back to normal, mysqld restarted, websites up and running. What am I doing wrong?

 

[J65nko]

Instead of giving you a fish, I will teach you how to fish

• Run # /usr/bin/env and save the output somewhere.
• Add the following three crontab tasks:
  
  

  MAILTO = root 
  # 
  */1     *       *       *       *       (date; /usr/bin/env; echo) >>/tmp/rootcron.txt
  */1     *       *       *       *       date; /usr/bin/env; echo Mailed
  */1     *       *       *       *       bla

• In one terminal run # tail -f /var/mail/root
• In a second one run # tail -f /tmp/rootcron.txt

 

[ShelLuser]

In /root/restart_mysqld.sh

#!/bin/sh
/usr/local/etc/rc.d/mysql-server restart

And my crontab -e

5 * * * * sh /root/restart_mysqld.sh

So I'm expecting cron to re-start mysqld every five minutes. It doesn't, it fails to execute.

Two things come to mind here. First; why specify sh when you already have a shebang present in your shell script? That's a bit overkill, and it will most likely get you different results.

Second; I'd really recommend going over crontab(5) again. Because you didn't set a range up there, you simply set a static value which will result in a static time. So in this case you don't get a job which will run every five minutes. Instead you'll get a job which will run 5 minutes within the hour. And since you used asterisks in the rest of the fields this ends up with "5 minutes within the hour, every hour.".

So in this case you should either fill out the rest of the minutes when this should run (0,5,10,15,20,25,30,35,40,45,50,55) or simply use a step value in the likes of */5.

(Edit)

PS: Resetting MySQL every 5 minutes doesn't sound like a healthy thing to do here

 

[c00kie]

My server is in America although I'm in the UK:

# history

 106	12:53	/usr/local/etc/rc.d/mysql-server stop // 17:53 English time
 107	12:53	history

and my mysql server was stopped @ at 17:53 manually, by myself at the command line. I checked back @ at 17:59 and it still hasn't started. Although @ at 18:05 it HAS restarted successfully.

But that's a 12 minute period, not 5? WTF!

 

[c00kie]

PS: Resetting MySQL every 5 minutes doesn't sound like a healthy thing to do here

Indeed, it's mainly a test to see some quick results in the short term. I'm going to try (in crontab);

@daily /root/restart_mysqld.sh

for now, and see if it starts automatically given 24 hours or so. Thanks again.