Re: WDSC lockup (was: Rational Updater)

>Dean, Robert wrote:
>> With all due respect, what you've written shows exactly what the
>> disconnect was.  No knowledgeable person would think that an app can
>> block the system UI thread (we're not talking OS/2's single kernel
>> msgq after all), so I didn't see how what I wrote would be
>> misinterpretted.
>
>Um, how many people understand OS/2's single kernel msgq?  I certainly
>don't.  One wouldn't think that a single app could cause an entire
>system to crash, but I've seen it happen.

I've seen it happen, too.  In this case, and to XP's credit, XP was 
able to become responsive again after RSE completed its work.

>I would request that everyone be civil & polite in all aspects of list
>interaction.  We're all professionals.

Indeed, but at the same time, don't mistake directness for rudeness or
lack of civility.

>> The bottom line is that the RSE blocks the WDSC UI thread without the
>> courtesy of at least a progress bar with a Cancel button.  It should
>> provide that at the very least.  As an Eclipse 3.x plugin, it should
[snip]
>I don't think it's a the blocking of the UI thread is something that RSE
>is doing deliberately.  I've talked to some of the the WDSC developers
>and they indicated this was something that is NOT normal.

For me, it happens *every* time I open a filter that returns more than a
trivial number of results.  It also affects the iSeries Table View.  
During my testing of source configurations, <= 5,000 results took up to
30 seconds to come back.  5,000 through ~13,000 took 3-4 minutes, and
the result that returned 23,000 hits took 15 minutes.  During compute
time, WDSC was locked up.  I haven't retested on the new PC because the
test files have been deleted.