Monit and pt-heartbeat

[Callum Macdonald]

I'd like to monitor mysql replication delay with monit using the
pt-heartbeat tool. The tool writes the replication delay in seconds to a
file. The file is a single line like:
12.00

The file is opened, truncated, and closed every interval, so it will
only contain the most recent statistics.

My first thought was to use monit's file timestamp check like this[1],
but I realise the file will be updated every check, no matter whether
the replication is up to date or not.

Second thought, use the file content check. However, I realise that is
unlikely to work because, from the monit man page:
On startup the read position is set to the end of the file and Monit
continue to scan to the end of file on each cycle. But if the file size
should decrease or inode change the read position is set to the start of
the file.

As per above, the file inode or size are unlikely to change. So monit
will most likely never read the changes.

That leaves 2 options I can think of.

Option 1 is to write a script, if the logged replication delay is <
threshold, exit 0, if not, exit 1. Then use monit's exit status check.
However, this moves the check logic out of monit into another script and
reduces the available information in the monit dashboard.

Option 2 would be to use the approach outlined here[1]. So if the
replication delay is < some threshold, touch a file. Then use monit's
timestamp checking to alert if that file gets too old.

Any advice on which of these options to pursue? Anyone else doing this
or any tutorials / code samples online somewhere?

I'd be grateful for any input / ideas on this one.

Love & joy - Callum.

[1]
http://www.elevatedcode.com/articles/2007/12/05/replication-monitoring-with-monit/