matlux / jvm-breakglass Goto Github PK

I think it might make sense not to catch the exception when the Server start or stops of or when the MBean registration fails?

What do you think?

Should it be optional?

Should it the exception not probagate by default in order not to mess up the hosting application when this tool is not working properly?

https://travis-ci.org

it has a very easy setup with github.
I look at it tonight.

Hi,

the server throws the following exception:

ERROR: Unhandled REPL handler exception processing message {:id 7e129e42-acf2-4d05-bc27-beedcf66334b, :op clone}
java.lang.NoClassDefFoundError: clojure/tools/nrepl/StdOutBuffer
at clojure.tools.nrepl.middleware.session$session_out.invoke(session.clj:26)
at clojure.tools.nrepl.middleware.session$create_session.invoke(session.clj:118)
at clojure.tools.nrepl.middleware.session$create_session.invoke(session.clj:114)
at clojure.tools.nrepl.middleware.session$session$fn__657.invoke(session.clj:178)
at clojure.tools.nrepl.middleware$wrap_conj_descriptor$fn__393.invoke(middleware.clj:17)
at clojure.tools.nrepl.server$handle_STAR_.invoke(server.clj:18)
at clojure.tools.nrepl.server$handle$fn__704.invoke(server.clj:27)
at clojure.core$binding_conveyor_fn$fn__4145.invoke(core.clj:1910)
at clojure.lang.AFn.call(AFn.java:18)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)

I wonder if the log messages are a little too intrusive?
I wonder if it wouldn't make sense to make them optional or to turn them off?

I have made changes to the code so now the user has a chance to catch a RuntimeException when start,stop,registerMBean and unRegisterMBean fail.

Why do we need to force the logging of the error if we already throw an exception?

For example is it not intrusive to force the logging of a Sever error when we know we're going to catch it further up and manage it in a way which is not dramatic for the application?

Should the README recommend require instead of use?

I got the following at the REPL, which seems like it could lead to trouble:

WARNING: name already refers to: #'clojure.core/name in namespace: user, being replaced by: #'me.raynes.fs/name
WARNING: parents already refers to: #'clojure.core/parents in namespace: user, being replaced by: #'me.raynes.fs/parents

Great project, thank you for making this available. Is the JAR only on clojars.org? I'd like to use it at work, where we proxy to Maven Central, but not clojars.

Hi Mathieu,

What do you think if jvm-breakglass would use java.util.logging instead of out.println?
I would do the refactoring.
cheers. stan.

I had to re-run the build 3 times to get it to pass.

When it happens I'm getting a socket Exception when the test tries to connect onto the REPL to remmotly executing code.