Grouper Users - Open Discussion List

[Sean Mason <>]

Hi There,

 

I’ve been experimenting with Grouper for some time with some success.  Recently a small production use case to fill a need that arose suddenly that I’m looking to fill with Grouper, and I’m having a very difficult time.  Time is now tight, and I will have to move to another strategy to make deadlines, but I would like to know where I’ve gone wrong.   I will break the two issues I’m having into two email to hopefully keep any offered help focused.

 

I’m running Grouper 2.3.0, on RHEL 7, Oracle JDK 1.8.0_91.  The repository is Postgres 9.2.14.

 

I have a single SQL_SIMPLE load job. 

Parameters:

grouperLoaderDbName: CSPR

grouperLoaderQuartzCron: 0 0 * * * ?

grouperLoaderQuery:  select uid as SUBJECT_ID from sysadm.firsttimeundergrad_vw

grouperLoaderScheduleType: CRON

grouperLoaderType: SQL_SIMPLE

 

When I run loaderRunOneJob(group) from gsh, I see:

loader ran successfully, inserted 8307 memberships, deleted 8307 memberships, total membership count: 8307, unresolvable subjects: 0

If I run the loaderRunOneJob(group) from gsh once again, I see:

loader ran successfully, inserted 8307 memberships, deleted 8307 memberships, total membership count: 8307, unresolvable subjects: 0

I know the contents of the view have not changed between runs.

 

If I set loader.failsafe.use = true, loader.failsafe.maxPercentRemove = 30, the load fails as it’s trying to remove the entire population before re-adding.

 

Surely this isn’t the expected behavior.  I’m at a loss where else to turn.  Is there another configuration step that is required?  Is something amiss?

 

Please advise,

Sean.