On 14 August 2014 21:42, Gordon Sim <[email protected]> wrote: > On 08/12/2014 07:19 PM, Justin Ross wrote: > >> Please try the >> beta in your test environment, in scenarios important to you, and report >> what you find, posititve or negative. >> > > I tested the following tarballs, no blocking issues, couple of minor > niggles with both c++ and java: > > (1) qpid-cpp-0.30-beta.tar.gz > > - builds and installs fine, broker starts up > - minor niggle is that there are a couple of test failures due to not > finding the qpid-ha tool, even if its on the path, because it's looking in > a relative location that doesn't exist > - another minor niggle, thee examples come with a cmake file but its not > obvious how to get this to work against a non-standard install location > > (2a) > qpid-python-0.30-beta.tar.gz > qpid-qmf-0.30-beta.tar.gz > qpid-tests-0.30-beta.tar.gz > qpid-tools-0.30-beta.tar.gz (python part only) > > All install fine, tests all pass against c++ broker, qpid-stat, > qpid-config etc run as expected. > > (2b) > qpid-tools-0.30-beta.tar.gz (java part only) > > Builds fine, packages produced and once I figured out what I was looking > for was able to run some of the command line tools and start the console > against the c++ broker. > > This is a really nice addition to the release artefacts and will be even > better when split out on its own. I'd encourage anyone who hasn't used this > to give it a try! > > (3) qpid-java-0.30-beta.tar.gz > > - tests all pass, packages produced > - tested 0-10 and 1.0 client examples > (I couldn't get the mvn build to work for these; probably something I'm > doing wrong though) >
What issue did you see? I mentioned in the readme that you need to do a local install of the clients to test locally-built artifacts, but in this case because the source is still versioned 0.30-SNAPSHOT the examples should resolve against the remote snapshots repo too. I have suggested that the next cut has the source without a snapshot version, at which point you will definitely need to install locally to get them working during the release process. For end users with the final release, the clients will be pulled down from maven central. > - all the packages have 0.30-SNAPSHOT subdirs; I assume the 'SNAPSHOT' > part gets removed in the final versions? > I proposed shifting the version to 0.30-beta etc to allow testing the full deployment process out, but although the deployed binaries have that, it isnt reflected in the source currently. I have suggested that the next cut has the source without a snapshot version, e.g qpid-0.30-rc1. Essentially everything should be a snapshot until we actually release it or at least cut a staged RC of it, and only once we are decided its going out should it really become versioned 0.30. Some even take this to the extreme that if we voted on a 0.30 versioned set of artifacts and then decided we had to do a respin before the vote passed and we released, the respin couldnt be called 0.30 anymore but rather 0.30.1 What we are doing currently is a little odd in part because our release process isnt particularly typical of maven usage. The maven release plugin (whch we arent using currentlly) can be used to manage all the versioning, branching, tagging etc for you, but when using that its typical to e.g. tag each RC/release straight off your trunk/version-branch, going from X-SNAPSHOT to X[-classifier] as part of the tagging process and bumping the trunk/version-branch to Y-SNAPSHOT at the same time. I expect this stuff to become clearer when we actually start looking at truly independent releases. > - started broker, ran clients against it > - the 1.0 example fails with a NPE until you create a queue named 'queue' > Yes, I mentioned creating the queue in the readme to help try people along. Would be nice if it didnt NPE of course. - tested with c++ client from 0-10 and 1.0 > - sending message with c++ 1.0 sender to c++ 0-10 receiver I get an > exception from the broker[1] > - sending message with c++ 1.0 sender to jms 0-10 receiver the broker > exits with what looks like the same exception[2] > > > [1] internal-error: Exception processing command: > java.lang.IllegalArgumentException: > unable to resolve type: class org.apache.qpid.amqp_1_0.type.UnsignedInteger, > 1 [caused by 8 \x00:\x00] > > [2] java.lang.IllegalArgumentException: unable to resolve type: class > org.apache.qpid.amqp_1_0.type.UnsignedInteger, 1 > at org.apache.qpid.transport.codec.AbstractEncoder. > encoding(AbstractEncoder.java:305) > at org.apache.qpid.transport.codec.AbstractEncoder. > writeMapEntries(AbstractEncoder.java:362) > at org.apache.qpid.transport.codec.AbstractEncoder. > writeMap(AbstractEncoder.java:351) > at org.apache.qpid.transport.codec.BBEncoder.writeMap( > BBEncoder.java:35) > at org.apache.qpid.transport.MessageProperties.write( > MessageProperties.java:382) > at org.apache.qpid.transport.codec.AbstractEncoder. > writeStruct32(AbstractEncoder.java:288) > at org.apache.qpid.transport.codec.BBEncoder.writeStruct32( > BBEncoder.java:35) > at org.apache.qpid.transport.network.Disassembler.method( > Disassembler.java:214) > at org.apache.qpid.transport.network.Disassembler.command( > Disassembler.java:172) > at org.apache.qpid.transport.network.Disassembler.command( > Disassembler.java:48) > at org.apache.qpid.transport.Method.delegate(Method.java:159) > at org.apache.qpid.transport.network.Disassembler.send( > Disassembler.java:73) > at org.apache.qpid.transport.network.Disassembler.send( > Disassembler.java:48) > at org.apache.qpid.transport.Connection.send(Connection.java:407) > at org.apache.qpid.server.protocol.v0_10.ServerConnection.send( > ServerConnection.java:543) > at org.apache.qpid.transport.Session.send(Session.java:602) > at org.apache.qpid.transport.Session.invoke(Session.java:772) > at org.apache.qpid.server.protocol.v0_10.ServerSession. > sendMessage(ServerSession.java:263) > at org.apache.qpid.server.protocol.v0_10.ConsumerTarget_ > 0_10.send(ConsumerTarget_0_10.java:304) > at org.apache.qpid.server.queue.QueueConsumerImpl.send( > QueueConsumerImpl.java:476) > at org.apache.qpid.server.queue.AbstractQueue.deliverMessage( > AbstractQueue.java:1248) > at org.apache.qpid.server.queue.AbstractQueue.attemptDelivery( > AbstractQueue.java:2003) > at org.apache.qpid.server.queue.AbstractQueue.processQueue( > AbstractQueue.java:2170) > at org.apache.qpid.server.queue.QueueRunner$1.run(QueueRunner. > java:77) > at java.security.AccessController.doPrivileged(Native Method) > at javax.security.auth.Subject.doAs(Subject.java:356) > at org.apache.qpid.server.queue.QueueRunner.run(QueueRunner. > java:68) > at java.util.concurrent.ThreadPoolExecutor.runWorker( > ThreadPoolExecutor.java:1145) > at java.util.concurrent.ThreadPoolExecutor$Worker.run( > ThreadPoolExecutor.java:615) > at java.lang.Thread.run(Thread.java:724) > > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
