StuckThreadDetectionValve with threads on IO

[1] uses a time based threshold to mark a request thread as "stuck" above the configured threshold. This is definitely useful, but maybe a thread is busy transferring a large request (IO). This could become a wrong conclusion for the thread to be marked as stuck. A relevant contextual example is

Re: Tomcat 8.0.38 to 9.0.6 -- Default Class Loader and Naming InitialContext

Managed to narrow down to a regression: https://bz.apache.org/bugzilla/show_bug.cgi?id=62224 On Tue, Mar 27, 2018 at 2:09 PM, Behrooz Nobakht wrote: > Hi Chris, > > Thanks for the reply. Comments inline: > > On Tue, Mar 27, 2018 at 4:31 AM, Christopher Schultz < > ch...@c

Re: Tomcat 8.0.38 to 9.0.6 -- Default Class Loader and Naming InitialContext

Hi Chris, Thanks for the reply. Comments inline: On Tue, Mar 27, 2018 at 4:31 AM, Christopher Schultz < ch...@christopherschultz.net> wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA256 > > Behrooz, > > On 3/26/18 12:26 AM, Behrooz Nobakht wrote: > > Hi,

Tomcat 8.0.38 to 9.0.6 -- Default Class Loader and Naming InitialContext

Hi, I am migrating a code base from Tomcat 8.0.38 to 9.0.6. There are a few things I'd like to confirm based on observations so far and ask for advice: 1. Default Class Loader In 9, it seems that the default is ParallelWebAppClassLoader compared with WebAppClassLoader in 8.0. What behavior chang