>On 23/07/2025 15:36, Daniel Sheridan wrote: > ><snip/> > >> Hi Rainer, >> >>> it might be a search for non-existing classes or resources >> The JARs are being scanned for specific classes during the delay, but they >> are classes that should already be loaded in, which is the confusing bit. >> >> I've got the highest log level set on org.apache.catalina.loader and >> org.apache.catalina.webresources already, but no signs of the classes being >> unloaded at any point before the delay occurs unfortunately. > >Can we see the full stack trace for the first time one of these classes is >loaded and any subsequent loads of the same class. > >Possible causes include: >- same class being loaded by different class loaders >- some transformation is taking place so each load is actually creating > a different class >- something else I haven't thought of yet > >Mark > > >--------------------------------------------------------------------- >To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org >For additional commands, e-mail: users-h...@tomcat.apache.org
Hi Mark, After some thought, I'm just going to go with the indexing fix for now, and not spend any more time looking into the class loading questions. The indexing change solves the issue for us, and I'm not seeing any other issues linked to the class loading, so I'll conclude this here assuming we don't find any related issues in future. Really appreciate you and Chris's help with this and anyone else who chimed in, I was really mystified by this issue so very happy we could get it sorted. Thanks for the support! - Dan
