Grouper Users - Open Discussion List

[Scott Koranda <>]

> Hi,
> 
> On this wiki page for Grouper diagnostics
> 
> https://spaces.internet2.edu/display/Grouper/Grouper+diagnostics
> 
> it shows that one configuration that can be adjusted is
> whether or not an issue with a JDBC source is diagnosed and
> reported on:
> 
> ws.diagnostic.ignore.source_jdbc = false
> 
> Does that option only allow one to ignore issues with JDBC
> sources?
> 
> Is there a similar option for JNDI (LDAP) subject sources?
> 
> The issue is an LDAP server going away temporarily during
> regular maintenance and that causing a relatively long-lived
> "error" reported by the diagnostics (via WS) due to the
> caching. We would like to avoid having an LDAP short outage register
> as an error (500 HTTP code), since Grouper itself and the
> downstream tools we have relying on it handle short outages
> without any real trouble. 
> 
> An option is, of course, to skip the "Trivial option" as it is
> labeled and instead do specific monitoring using the
> 'diagnosticType' query parameter for the WS url. We will
> consider that but were also wondering if we can just turn off
> the LDAP source check in some way.

I should have been more clear. Right now we use

diagnosticType=all

We are considering using 'db' and 'sources'. The wiki page
says 

"Note that the same sources.xml settings that configure the
Grouper startup settings will apply here as well.  i.e. you
can skip a source, or set the ID to search for."

What is the syntax to "skip a source"?

It would also be convienent if there was a diagnosticType
option for just the loader jobs, or even more useful to check
for a successful run of a particular loader job.

Can that enhancement be added to the roadmap?

Thanks,

Scott K