Getting around the limitations of Gatekeeper
Greetings Admins -
So, it goes without saying that Gatekeeper is extremely limited in its capability. Wht PTC didn't design the filter to look for certain check types, rather than a cumulative total amount is beyond me...perhaps some of you out there have opinions on the subject?
In spite of these limitations, my company is moving forward with using Gatekeeper to allow "0" errors.
Here are few of the items I hoping you folks can provide some guidance on -
Scenario: We have the “checks.mch” file configured where we have 5 checks identified as “Errors”. These 5 errors are the “critical” stuff like “FAILED_COMPONENTS” that must not be allowed into Windchill for models being “Released” under any circumstances. The other part of this scenario is we have Gatekeeper configured to allow “0” errors.
1st Question: If our methodology is that want to allow “In Work” models to be checked in under any circumstance, but truly have “0” errors when set to “Released”, how might that be done?
The idea is to allow looser standards at In Work, but more rigid ones at Released. Since the State change from In Work to Released is a Windchill function, not a Creo function, I am having some trouble getting my head around how we can accomplish this? One thought I had was to make it part of the Change process, where there is an associated Change Task to “fix” all errors before it goes to Released….thoughts, comments??
2nd Question: My company is insisting on of an “override” policy. The policy temporarily allow “Errors” to get into Windchill at the Released state. The process would be defined so that there would be tracking for these override scenarios, so they are fixed ASAP, rather than just forgotten about.
So, my idea on this would be to add a condition parameter into the Creo models, where the value of the parameter would determine whether Gatekeeper checks or ignores the model…something like a “CHECK/NOCHECK” value…thoughts, comments??
3rd Question: The setconf.mcc file allows us to have different “check configurations” based on how a model is classified (OOTB- Heavy, Medium, Light). Based on the scenario noted above, I am wondering how this might help us to get to where we want to be? I suppose we could have “In Work” models look at a set of configs where all checks are “warnings”, which would get us around Gatekeeper, but for “Released” models, we are back to same question that comes up with Question 2….need some guidance here.
4th Question: Legacy models: Lastly, I think it bears mentioning that out intent with legacy data would be to “NOCHECK” them unless they become part of the Change (revised back to In Work). Since this requires a condition.mcc to ignore “Released” models, how does that play into the above scenario?
My appreciate in advance for any feedback you are willing to provide...
Gary

