Build precise queries to find exactly what you need
Press ESC to close
@krutaw
Member Since: February 23, 2014
Favorites0
Views
Projects0
Amazing idea, absolutely love it Troy! Only problem is that when I unzip it to my dev Nagios server, it hangs when i click on "Configure" within XI. More than willing to provide whatever data you need to sort this out so others aren't impacted as well. Cheers mate! :)
Reviewed 11 years ago
Excellent idea, however, in order to make this actually effective, quite a bit of code needs to be modified in order to add custom variables so that the workflow in JIRA can be more intelligent (read: actually valuable.) The only reason why I'm not using this component is the fact that not every alert is an incident. In a large environment, you could be looking at 200-300 tickets being generated a day which will overwhelm the JIRA server unless you're deleting tickets on a regular basis (or have a Cray XC40 dedicated to running your JIRA instance.) A good way to fix this problem would be with the use of custom variables, however, that would require hand-editing every single check that you want to include. Given that XI won't properly import custom variables (current version), this is less than an ideal methodology. Again, if you've got a small shop, this might be a really interesting component. In my environment though, I had to go another route (and will share once I get corporate approval.) :-D
Okay seriously, this is awesome. Got it working with minimal effort (I too had to edit the highcharts.html to change the IP address). You know what would be an awesome followup to this? The ability to download the graphs via URL directly so that they can be included in email notifications.
The only thing I'd recommend is that you look at actually re-iterating the check of the stats so you can do a couple of things. First, it'll give you more than just a single point in time, but secondly, you can provide actual CPU %'s instead of just raw time output. Specifically, you would need to grab the data, wait a second and then grab the data again which you could then use math to calculate percentages. From there if you wanted to be really fancy you could all an average of x runs of the command to give a more "real" look into the system. Overall, good work.
The total count only goes up to 50 for the OS level in KMS. No real value above that count.
This plugin is missing perfdata in the output although the notes on the plugin clearly state that perfdata is included. Easy fix, but if you mention perfdata, please include it.