[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [monit] File contents check

From: Martin Pala
Subject: Re: [monit] File contents check
Date: Sun, 09 Nov 2008 23:34:59 +0100
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv: Gecko/20080829 Iceape/1.1.12 (Debian-1.1.12-1)


the content matching test saves the read position from last cycle. The next cycle begins the match test on the last read position if the size of the file is bigger then saved read position.

The read position can be reset and thus monit will start to check match from the file beginning in following cases:
1.) either the inode changed (i.e. file was replaced)
2.) or the file size shrinked (i.e. the file was rewritten for example)

It was designed this way to allow to parse for example log files which are being appended - in such case it's not possible to check everytime from beginning because it will emit alarms for content which was handled in previous cycles already and also the parsing will be slower as the file will grow if restarted from the beginning each time.

Workaround for your situation:

Instead of rewriting the file for example by:
 echo master > /var/run/cyso-ha/inetd

just apped the new state this way:
 echo master >> /var/run/cyso-ha/inetd

Monit will in such case parse the file from last read position and grab just the new state. The /var/run/cyso-ha/inetd will also contain history of state changes this way as by-product.


Wijnand Wiersma wrote:
I think I don't really understand that feature, but I do need it for
what I have in mind.

I am trying to build a role based cluster and in my test setup I just
use inetd to test it.

check file inetd-group with path /var/run/cyso-ha/inetd
    start program = "/opt/scripts/cyso-ha/start_role inetd"
    stop  program = "/opt/scripts/cyso-ha/stop_role inetd"
    if match "master" then start
    if match "slave" then start
    if match "stopped" then stop

When I echo something into /var/run/cyso-ha/inetd I get following
message in my log:
Oct 28 20:16:02 cnode2 monit[2699]: 'inetd-group' file exist
and the start script really runs.

But when I echo "stopped" into that file nothing happens. I would expect
after reading the documentation my stop script should have run.

Is there something lacking in the documentation or is this a lack of brains?


To unsubscribe:

reply via email to

[Prev in Thread] Current Thread [Next in Thread]