Hello Dan,
Thanks for your response.
Can you confirm this is indeed the same DC issue? The symptoms are subtly different as the second document is already broken. If not, do I need to log an incident report for this issue?
We are indeed proposing both options to our customers. Recycling Word instances is fortunately a configuration option in our product, but one that was intended for recovery (leaky PCL printer drivers in Windows NT), not for standard use.
The performance impact of restarting Word for _each_ print and the unavoidable overhead of printer setup which now has to be done for each print does have a tremendous impact. If I recall the logs correctly for the customer that reported the problem to us
I estimate that each print will go from 100ms to 1200ms or more - claiming a full CPU core during this time.
I was starting to write a response on the 'whether or not practical' for our circumstances -- which are very 'enterprise' focused. However, that response got way too long and way too detailed for this public forum. Contact me directly if you want to discuss
this.
We are urgently waiting for a formal statement on this issue that we can refer to customers. MS16-098 is a security fix so it is very hard to justify not deploying it; this tends to violate operational procedures and compliance requirements and that makes
it hard to argue against deployment. There is a definite conflict of interests between business and IT here, and it helps to make an informed decision if it is known whether or not the effects of that choice will be temporary (this applies to delaying deployment
as well as acceptance of a workaround with severe impact).
You are aware there are already quotes from an incident report with an explicit ETA date floating around?
Thanks,
Stefan
E: stefan(dot)ten[no-dot]hoedt(at)outlook(dot)com