On 5/28/2010 at 10:55 PM, Gianluca Cecchi <gianluca.cec...@gmail.com> wrote: > Hello, > I tried this approach and it seems to work ootb. > But I would like to know if there could be silent drawbacks or potential > problems from a technical point of view. > > Under /usr/lib/ocf/resource.d/ I create a directory named myocfdir and > inside it I put a copy of an existing RA (eg apache) > /usr/lib/ocf/resource.d/heartbeat/apache renaming it apache2 in this new > directory > > Then I can customize its parameters or workflow or whatever else, respecting > RA principles, and refer to it in my resources in this way: > > primitive myapache ocf:myocfdir:apache2 \ > params configfile="/usr/local/apache2/conf/httpd.conf" statusurl=" > http://localhost:80/server-status"; \ > op monitor interval="1min" \ > op start interval="0" timeout="40" \ > op stop interval="0" timeout="60" > > Keeping save gpl obligations for the ra, the sharing > of personalizations that could be useful for the community, ecc... > (it is only an example to give the idea... no intention to modify the apache > RA ;-) > > Could this be a runnable approach? > Also to put for example totally new personal RAs in new dirs?
Nothing wrong with that. That's actually exactly what you *should* do if you're writing your own RAs that aren't going to be pushed upstream into either of the resource-agents or pacemaker packages. Regards, Tim -- Tim Serong <tser...@novell.com> Senior Clustering Engineer, OPS Engineering, Novell Inc. _______________________________________________ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
