Re: daemon not really doing anything on Fedora 13

From: Mike Young <myoung_at_wildernessvoice.com>
Date: Wed, 28 Jul 2010 18:32:52 -0600

You're right. Reboot works fine. But stop the service and restart looks like a debug session with everything going to stdout.

Mike

-- Sent from my Palm Pre
On Jul 28, 2010 5:49 PM, David Ward &lt;david_at_dward.us&gt; wrote:

That is right Mike.

A normal system startup shouldn't see the daemon write to all ttys.

I am surprised that

/etc/init.d/incrond start

is doing this, actually.

-- 
Regards
David Ward
On Wed, 28 Jul 2010 15:13:07 -0600, Mike Young
&lt;myoung_at_wildernessvoice.com&gt;
wrote:
&gt; So it seems to be working pretty well now. I created a script that
&gt; actually contains my regex and functions. Now incron gets triggered and
&gt; then looks at that file, which probably what I should have done in the
&gt; first place.
&gt; 
&gt; One thing I've noticed, though, is that the terminal I start my incrond
&gt; service in is going nuts with all sorts of activity. I am running the
&gt; daemon via /etc/init.d/incrond. 
&gt; 
&gt; Is there a way to quiet the output? Or does it just go away if it's run
at
&gt; boot up via runlevels?
&gt; 
&gt; Thanks,
&gt; 
&gt; Mike
&gt; On Jul 28, 2010, at 11:06 AM, Matthew Miller wrote:
&gt; 
&gt;&gt; On Wed, Jul 28, 2010 at 10:23:01AM -0600, Mike Young wrote:
&gt;&gt;&gt;&gt; My guess is that what's happening is that the log file is being
created
&gt;&gt;&gt;&gt; with permissions determined by the umask, and then chmodded. So, when
&gt;&gt;&gt;&gt; you do the chmod on file creation with incron, that works, but then
&gt;&gt;&gt;&gt; immediately after the application sets it back.
&gt;&gt;&gt; The problem is that log files get rolled with a time stamp and then a
&gt;&gt;&gt; new,
&gt;&gt;&gt; blank one is created. When the new one is created with 644, the app no
&gt;&gt;&gt; longer works. Changing the log via a cron takes care of the problem.
But
&gt;&gt;&gt; then I have the down time in between. This deals with it right from
the
&gt;&gt;&gt; start. Plus, I can use this for other items not related to this log
&gt;&gt;&gt; issue.
&gt;&gt; 
&gt;&gt; Yeah. So what I'm suggesting is that changing permissions on IN_CREATE
&gt;&gt; may
&gt;&gt; be *too soon*.
&gt;&gt; 
&gt;&gt; 
&gt;&gt; -- 
&gt;&gt; Matthew Miller           mattdm_at_mattdm.org         
&lt;http://mattdm.org/&gt;
Received on Tue Jun 05 2012 - 22:14:21 CEST

This archive was generated by hypermail 2.2.0 : Tue Jun 05 2012 - 22:14:21 CEST