TEC and Netcool "synchronization" demystified

Amongst all of the Netcool/TEC talk flurrying around, for those of you who are wondering how TEC and Netcool currently "integrate", read on...

Speaking from personal experience, IMHO the *current* TEC to Netcool synchronization, which was developed my MicroMuse prior to IBM stepping in, leaves a lot to be desired...

The "synchronization" is simply Netcool Omnibus doing a select based on timestamp from the TEC database via a JDBC connection. Omnibus uses proprietary "probes" for just about any external source one would want to pull data from, one of which is a "tec_probe" which reads this incoming result set from the select and parses the data for Omnibus to store in the event bus.

Once the data is available, it can be displayed on the Omnibus console.

Where's the "two-way synch" in this? You have to write a rule in Netcool Impact (separate product that Omnibus, BTW) using a proprietary "perl-ish" rule language and perform a postemsg back to TEC to update the original event if the status changes in Netcool.

Again, this is how it exists currently based on my experience. This "integration" also makes a big assumption that you want to flow events from TEC *up to* Netcool. Hopefully IBM has a much cleaner method of integration they are *feverishly* working on (or is it "migration" now? ;) ). I think that will remain to be seen for probably quite a while...