In case of "inversion of control" waiting, one could get the request processor into a starvation. This happened quite
often in datasystems tests, where modifications to disk happen in parael with scheduling and waiting for various
Created attachment 72816 [details]
Log & thread dump
How old is this? Reported against 7.0 but I presume a (potential) problem for much longer.
The code is there since 3.4, at least, maybe it is there even for longer amount of time.
Integrated into 'main-golden', will be available in build *200810300201* on http://bits.netbeans.org/dev/nightly/ (upload may still be in progress)
User: Jaroslav Tulach <email@example.com>
Log: #151682: Don't waitFinished() on tasks reported as isFinished()