I may open an IBM support case on this, but as always, the fastest and best
answers are usually here (when we're not debating if the Parrot is dead or
not, and what tags we need/don't need <g>)

UniData Replication offers a flag called "ReEvaluate Virtual Fields" and the
documentation says:

By default, the replication writer processes re-evaluate any virtual
fields on the subscribing system. If you want to use the values from
the publisher, set this value to 0.
Note: Great care should be given to changing this parameter from
its default, as it may lead to some unwanted results, especially in
virtual indexes.

I'm trying to wrap my brain around this -- how would the Virtual Field
evaluation be any different on the Subscriber?  Why would I *want* the
subscriber to re-evaluate a virtual field?  And how could this lead to an
indexing issue?  I'm trying to think of examples - and cannot! In my case,
virtual fields are usually performing some sort of math, string build, or
even a remote file update. But the 'product' of the virtual field is 'right'
- so why would the subscriber need to re-do the calculation?  And how could
anything be wrong?

Lastly - I'm curious if others have found many issues starting at 7.1.9 - it
appears to me that Replication is VERY stable on Windows at this point - and
when it does go 'offline', we're able to 'sync' and get right back in line
again -- I'm wondering if everyone else is having the same experience as I
prepare to start telling my client base 'it's really ready'.  Anyone have
experience to the contrary or 'caution when' stories?

David W.
-------
u2-users mailing list
[email protected]
To unsubscribe please visit http://listserver.u2ug.org/

Reply via email to