Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

I have several cron jobs that run (in /etc/cron.daily, /etc/cron.hourly, /etc/cron.weekly, etc.) and email root@localhost with the results. I'd like to stop those emails if the jobs are succeeding, and only email on error (which I understand can be done by redirecting stdout to /dev/null). I understand how to do that for individual cron jobs, but the scripts in those special directories are run using run-parts. What is the best way to suppress success emails for those scripts?

share|improve this question

2 Answers 2

You should send successful email notifications to /dev/null so they disappear.

But you want to see unsuccessful email notifications.

This means you need to first direct stdout to /dev/null and then direct /dev/stderr to stdout

try changing the redirection part of your cronjobs to

>/dev/null 2>&1

See this link

share|improve this answer
    
Wouldn't that suppress the error emails too (because they wouldn't produce any output)? Also, I need to do this for my cron.XXX directories which use run-parts, so it's not as simple as redirecting for individual scripts. –  jrdioko May 29 '11 at 1:47
    
No the idea is that stderr is thrown away then stderr is redirected to stdout. I'm not sure what run-parts is, but however it works redirection of stdout and then stderr seems to be the way. –  pavium May 29 '11 at 2:01
    
Aha, I googled run-parts. That does complicate the issue, doesn't it. Maybe you should avoid run-parts and invoke each script separately. –  pavium May 29 '11 at 2:05
    
Ah ok, I understand. run-parts runs all scripts in directories like /etc/cron.daily, so the trick is passing along the redirection to the individual scripts it is running. –  jrdioko May 29 '11 at 2:06
1  
Are you sure about the >/dev/null 2>&1 bit? I tested it and that funnels everything to /dev/null, where if you drop the 2>&1 only stdout gets removed. –  jrdioko May 29 '11 at 4:42

You may want to use one of the wrappers for the programs, that output everything when something goes bad and swallow stdout otherwise.

One example might be cronic, just prepend 'cronic' to 'run-parts' e.g.:

# m h dom mon dow user  command
 17 *  *   *   *  root  cd / && /etc/cronic run-parts --report /etc/cron.hourly

where /etc/cronic is a place with executable cronic script, downloaded from the website mentioned.

share|improve this answer
    
+1 solved my problem –  Artur Czajka Jul 17 at 18:46

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.