Peter, I don't think its going to work that way. The base element auto-identify rule is going to grant recon ids to all instances (except computer system and its subclasses' instances). Then, when you merge into production, its going to merge computer systems with pre-existing matches in production, it will merge the weak children of those computer systems, but it will also merge a hoard of orphaned weak children whose parent computer systems did not have a recon id and were thus left out of the merge activity.
As an alternative, I let everything get a recon id in the discovery tool dataset, then I created a custom base element integer attribute called "mergeid", I set that to a value of 1 for a couple of computer system instances. THEN, I created a merge activity for the discovery tool dataset -> BMC ASSET, applied a qualification group requiring that 'mergeid' = 1, but also set the option to commit child CIs. Unfortunately, this did not work, the qualification group barred the child CIs from entering, so the only merged CIs were the handful of computer systems where I had set mergeid to 1. SO, I am pretty sure there is no way to do this. -----Original Message----- From: Action Request System discussion list(ARSList) [mailto:[EMAIL PROTECTED] On Behalf Of Peter Romain Sent: Friday, October 31, 2008 11:09 AM To: [email protected] Subject: Re: Merging of non-reconciled child CIs Tim, I think I might have misled you in my earlier reply. You need a reconciliation job with an identification activity with an entry in the table on the identification tab for computer systems with auto-identify off and another for base element with auto-identify on. When the activity runs, an attempt will be made to identify any CI that has a recon id of '0'. The rule in the above table that most closely matches the class of the CI will be used. Therefore all computer systems (and sub classes of computer system) will use the computer system identification rule and all other CIs will use the base element rule. Weak children of identified computer systems will be identified. Weak children of unidentified computer systems will remain unidentified. Cheers Peter (Romain is my surname!) > Tim > > Yes you could qualify a merge down to even computersystems with a specific > categorisation for example. > > The identify job doesn't actually merge so the fact they have been auto > identified is neither here nor there. > > -----Original Message----- > From: Tim O'Connor <[EMAIL PROTECTED]> > Sent: Friday, October 31, 2008 4:49 PM > To: [email protected] > Subject: Re: Merging of non-reconciled child CIs > > Romain, > > So lets say I auto-identify base element, which should give a RE id to > all populated instances. That means its also going to auto identify all > comp sys instances, basically its just going to dump everything into BMC > ASSET; that's what I want to avoid. > > Now heres a different idea; lets say I auto identify the super-classes > of all component child objects. I don't auto-identify > BMC_ComputerSystem. Lets say reconciliation finds 5 matching Comp Sys > instances in BMC ASSET. Then, when I start the merge could I set it to > include Child CIs, but also apply a qualification group to that merge > that limits it to class BMC_ComputerSystem? Would that essentially allow > only the 5 reconciled BMC_ComputerSystem instances to be merged, along > with their child objects? > > The documentation is so poor. I don't know whether the child CIs would > be filtered out of the merge by the qualification group or not. > > -----Original Message----- > From: Action Request System discussion list(ARSList) > [mailto:[EMAIL PROTECTED] On Behalf Of P Romain ARSlist > Sent: Friday, October 31, 2008 3:23 AM > To: [email protected] > Subject: Re: Merging of non-reconciled child CIs > > > > [The entire original message is not included] > ________________________________________________________________________ _______ > UNSUBSCRIBE or access ARSlist Archives at www.arslist.org > Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are" > ________________________________________________________________________ _______ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are" _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are"
