Re: Using monit to monitor hundreds of thousands of file system resources?

[Jan-Henrik Haukeland]

On 1. sep. 2007, at 12.07, Sergio Trejo wrote:

> why couldn't monit be evolved to the next level of evolution such  
> that it could be just as good at scaling to hundreds of thousands  
> of file system resources for monitoring purposes? Personally, I  
> really like monit a lot and prefer to stay using it for these  
> purposes rather than have to spend time getting to know about  
> tripwire or something similar. I'd love to see an evolution of  
> monit for scaling to large resources. Is there a good reason why  
> monit should not evolve to that next level of maturity?

To do this scalable we will need to hook up monit to a database. This  
is certainly possible, we even have our own GPL database library we  
could use, with support for the 3 most popular open source databases[1]

However, I guess, 99.99% of those using monit will not use monit to  
handle over 100K of files. With the very limited development  
resources we have, I'm not sure its in our best interest to do this  
now. We still have a long list of TODO items [2] that are of more  
general interest. Besides doing this brings monit into the more  
"enterprise" segment of complex software. Not sure we want to go  
there. What do others think?

Just my $0.02

[1] http://tildeslash.com/libzdb/
[2] http://www.tildeslash.com/monit/doc/next.php