I am not sure if it is just me, but it often seems incredibly difficult to find where to add a rule to effect a check. In this specific case, I have a check that has a man page title of “LSI MegaRAID: State of Physical Disks (via MegaCLI)”. One would think somewhere in Setup->Service Monitoring Rules there would be something that would match with LSI MegaRAID, but there isn’t. And it seems like way too frequently, I want to tune a check, and I have to HUNT to find where to do it. It really feels like there should be a relatively direct way to go from a service check to where one would create a rule to effect that service check, but if there is I have yet to find it.
Very easy way. Go to the service you want to tune and there to the action menu and select “Parameters for this service”.
If this service has parameters you can change then you will see these.
That is easy, but I do not see how to change the thresholds for the check there.
For every check the box “Check origin and parameters” is relevant.
There inside you find normally a line with the name of parameters im my example “Disk IO levels” and the current value.
In this example the “Disk IO levels” have no default parameters.
And ok here it would be good to have a big button create some parameters. Instead you need to click on the name of the parameters.
This is a hard interface fail something for @TLI he asked for such problems.
Ok this check has no configurable parameters. Here is then more the question. Should this check have some parameters and it was forgotten to write the configuration extension for check or does the check by itself has no parameters.
There are values such as predictive disk failure that have a value.
If i take a look at the check itself then i see only “hard coded” parameters. The check also has in it’s header “_no_params”.
If you need to have some parameters in this check then you need to modify the existing one and also create some config template. This is more than only changing one or two lines.
At the moment it only checks for these two things.
- physical disk state → mapped with a defined dictionary
- if predictive failure exists the check goes warning
The only possible parameter would be a higher predictive failure count than 0.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact an admin if you think this should be re-opened.