Hi Benson, fair enough, still thanks for clarifying that the given code seems of no issue for your case. Regarding the tccl and threads, could you open an issue for that, so we don't miss it for a future 6.0?
Thanks, Achim 2016-11-09 13:45 GMT+01:00 Benson Margulies <[email protected]>: > 'Achim, > > In the situation I could repro, the bad class loader was always the > original plain old application boot class loader, not some class > loader from inside pax-web. > > In the backtrace, and the debugger, it was clear that pax-web just > wasn't setting the TCCL in this thread. I would offer the opinion that > when pax-web creates a thread pool, it should arrange to set the tccl > to its own bundle class loader for each thread. > > I'm changing jobs in two weeks, and I have to focus on things really > needed until then, so I'm afraid that I won't have time to try the > experiment of adding a tccl setting to the threads and seeing if the > problem goes away, or building a test case in the open. > > --benson > > > On Wed, Nov 9, 2016 at 6:09 AM, 'Achim Nierbeck' via OPS4J > <[email protected]> wrote: > > Hi Benson, > > > > it might be interesting, if this could be the cause of some of your > issues. > > Marc just found an interesting effect while trying to fix 906: > > > > https://ops4j1.jira.com/browse/PAXWEB-906?focusedCommentId=37803&page= > com.atlassian.jira.plugin.system.issuetabpanels% > 3Acomment-tabpanel#comment-37803 > > > > Could you verify if this maybe the cause for your issue? > > > > regards, Achim > > > > > > 2016-11-07 12:38 GMT+01:00 Benson Margulies <[email protected]>: > >> > >> On Mon, Nov 7, 2016 at 3:59 AM, 'Christoph Läubrich' via OPS4J > >> <[email protected]> wrote: > >> > Am 04.11.2016 13:23, schrieb Benson Margulies: > >> >> > >> >> > >> >> It's an important goal of this project to _not_ require an extra JVM, > >> >> so your suggestion > >> >> of using a process builder, while it would allow me to make this test > >> >> work, would not > >> >> be something I can use in production. > >> >> > >> > > >> > The you should not start something from within the (boot)class-path > but > >> > embedd everything in OSGi-Bundles and carefully craft import/exports > >> > instead > >> > of rely on automatic guessing from BND and such tools. This is the > only > >> > way > >> > to have true classloader-isolation. > >> > >> Christoph, the fundamental goal here requires a tiny, very controlled, > >> amount of system bundle/bootclasspath communication. I don't think > >> it's the source of my problems. > >> > >> > > >> > Pax Web (with embedded jetty) as well as Pax Exam rely on some parts > on > >> > Dynamic-Imports and thus can easily lead to have classes loaded from > the > >> > boot-class-path as well als some libs are not behave very frndly by > >> > using > >> > Class.forName or static Methods to obtain Connections and such. > >> > Context-Classloading is some kind of work-around to adress these > >> > problem, > >> > but not a real solution. > >> > >> OK. I had hoped that by trimming down the list of packages on the > >> system bundle, I could cut off this, but I take your point. > >> > >> > > >> > So I would recommend (if you really do not want to start up an extra > vm > >> > for > >> > testing, what would be in the sense of test-isolation preferable) that > >> > you > >> > pack your "launch the test service using the CXF API; no OSGi > involved." > >> > inside a bundle (embedding all dependecies) make sure the bundle does > >> > not > >> > export or import any packages and use an Activator to fire up the > >> > test-service. You can then use this bundle in your pax-exam test > without > >> > polluting your test-classpath with other libs (CXF itself requires a > lot > >> > as > >> > of the last time I have embedded this in a container). You also should > >> > use > >> > the remote-test-container to make sure nothing trickles in from other > >> > parts. > >> > > >> > > >> > -- > >> > -- > >> > ------------------ > >> > OPS4J - http://www.ops4j.org - [email protected] > >> > > >> > --- You received this message because you are subscribed to the Google > >> > Groups "OPS4J" group. > >> > To unsubscribe from this group and stop receiving emails from it, send > >> > an > >> > email to [email protected]. > >> > For more options, visit https://groups.google.com/d/optout. > >> > >> -- > >> -- > >> ------------------ > >> OPS4J - http://www.ops4j.org - [email protected] > >> > >> --- > >> You received this message because you are subscribed to the Google > Groups > >> "OPS4J" group. > >> To unsubscribe from this group and stop receiving emails from it, send > an > >> email to [email protected]. > >> For more options, visit https://groups.google.com/d/optout. > > > > > > > > > > -- > > > > Apache Member > > Apache Karaf <http://karaf.apache.org/> Committer & PMC > > OPS4J Pax Web <http://wiki.ops4j.org/display/paxweb/Pax+Web/> Committer > & > > Project Lead > > blog <http://notizblog.nierbeck.de/> > > Co-Author of Apache Karaf Cookbook <http://bit.ly/1ps9rkS> > > > > Software Architect / Project Manager / Scrum Master > > > > -- > > -- > > ------------------ > > OPS4J - http://www.ops4j.org - [email protected] > > > > --- > > You received this message because you are subscribed to the Google Groups > > "OPS4J" group. > > To unsubscribe from this group and stop receiving emails from it, send an > > email to [email protected]. > > For more options, visit https://groups.google.com/d/optout. > > -- > -- > ------------------ > OPS4J - http://www.ops4j.org - [email protected] > > --- > You received this message because you are subscribed to the Google Groups > "OPS4J" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/d/optout. > -- Apache Member Apache Karaf <http://karaf.apache.org/> Committer & PMC OPS4J Pax Web <http://wiki.ops4j.org/display/paxweb/Pax+Web/> Committer & Project Lead blog <http://notizblog.nierbeck.de/> Co-Author of Apache Karaf Cookbook <http://bit.ly/1ps9rkS> Software Architect / Project Manager / Scrum Master -- -- ------------------ OPS4J - http://www.ops4j.org - [email protected] --- You received this message because you are subscribed to the Google Groups "OPS4J" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
