https://bz.apache.org/bugzilla/show_bug.cgi?id=60623
--- Comment #10 from romain.manni-bucau <rmannibu...@gmail.com> ---
> Code that relies on the deployment mechanism for correct clean-up is broken
> and needs to be fixed. Deployment threads may be re-used before they are
> stopped and if any component fails to fully clean-up, the data could leak to
> the next deployment and potentially cause a security issue.
No cause it is only in these threads so this pattern works well without
leaking:
deploymentStart() { context.set(new DeploymentEn(...)); }
deploymentStop() {}
Don't get me wrong, I don't like that code but saw it often and with tomcat
behavior it was more than valid.
> Re-defining a value of 1 for startStopThreads to mean 'serial deployment on
> the main processing thread' rather than 'serial deployment via a thread pool
> with a single thread' looks perfectly reasonable to me at this point.
I think I didn't get what was the issue using 0 for that? It keeps
compatibility and allow this new feature as well.
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
