Re: 0.6 Release Plans from now onwards
Hi Andrew, Thanks for the update. I've been trying to wade through code reviews/tidy ups to get the JIRAs in a better state. Anyone willing to pitch in on the Java reviews would be much appreciated. Marnie On Tue, Dec 15, 2009 at 10:26 PM, Andrew Stitcher astitc...@redhat.comwrote: As I said in my beta release message I'd like to produce a release candidate at the end of this week. On the closing bugs front there are a couple of issues still (but nothing that looks deadly as far as I can tell): Essentially we have 2 blockers and 2 critical bugs. 2 Java bugs and 2 bugs for the WCF code. QPID-2273 (http://issues.apache.org/jira/browse/QPID-2273), QPID-1830 (http://issues.apache.org/jira/browse/QPID-1830) - Aidan can you give some status on these bugs? Are they going to be finished in the next couple of days? Is there much risk in the fixes? QPID-2260 (http://issues.apache.org/jira/browse/QPID-2260), QPID-2247 (http://issues.apache.org/jira/browse/QPID-2247) - These are more problematic as they don't have anyone assigned to them. It could be that Cliff is working on them, but as he isn't a commiter he can't be assigned the bugs in Jira. Cliff can you comment on these bugs? Are they truly as important as blocker/critical? If so unless there are patches attached to the jiras very soon I think we'll have to omit the specific WCF source code from this release (of course it will still be in the big source tarball). To be honest I'm really sure it's quite mature enough in any case. Assuming that we can clear these issues up (one way or another) in the next couple of days - I plan to release rc1 on Thursday/Friday this week. I will create a (preemptive) subversion tag at that point. From that point I'd like to *freeze* checkins to the tree for 2 weeks taking us effectively to the new year. If we haven't actually voted to release by then I suggest we move the branch tag to whatever the most recent qpid checkin is and do all subsequent release work on a branch, and unfreeze the trunk. This way we should have the minimum frozen time and still be able to progress. Regards Andrew - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org
Re: 0.6 Release Plans from now onwards
On Tue, Dec 15, 2009 at 10:08 PM, Aidan Skinner aidan.skin...@gmail.com wrote: On Tue, Dec 15, 2009 at 11:46 PM, Robbie Gemmell robbie.gemm...@gmail.com wrote: I came up against that a couple months ago. The reason seemed to be that the ConversationFactory is calling toString() on a message before its sent, and the toString() behaviour was changed from previously outputting the hash code to outputting actual content. (r756228 seems to be where you'd want to look). Ah, good spot. Reverting that patch fixes it for me. I'm going to assign the bug to Rajith since just backing it out doesn't seem entirely the right thing. I had a closer look and don't think the above fix is the reason for this issue. I have commented on QPID-1830 with the details. - AIdan (patch review doesn't catch everything) -- Apache Qpid - AMQP, JMS, other messaging love http://qpid.apache.org A witty saying proves nothing - Voltaire - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org -- Regards, Rajith Attapattu Red Hat http://rajith.2rlabs.com/ - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org
Re: 0.6 Release Plans from now onwards
On Tue, Dec 15, 2009 at 10:26 PM, Andrew Stitcher astitc...@redhat.com wrote: QPID-2273 (http://issues.apache.org/jira/browse/QPID-2273), QPID-1830 (http://issues.apache.org/jira/browse/QPID-1830) - Aidan can you give some status on these bugs? Are they going to be finished in the next couple of days? Is there much risk in the fixes? I'm currently working on QPID-2273, it's confined to the protocol initiation so should be low risk. I'll have a fix for that soon. QPID-1830 is more problematic, I don't really have a good handle on what the issue there is yet. I don't *think* it'll be particularly invasive but it's hard to say. It seems to have been caused post-0.5, if anybody would like to claim responsibility for touching BytesMessage in the last few months and have a look that'd be great. ;) - Aidan -- Apache Qpid - AMQP, JMS, other messaging love http://qpid.apache.org A witty saying proves nothing - Voltaire - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org
RE: 0.6 Release Plans from now onwards
It is actually possible to assign JIRA's to non-committers, I believe you just need to add the relevant JIRA account to the Contributors role for the project so that they show up as a possible assignee. Robbie -Original Message- From: Andrew Stitcher QPID-2260 (http://issues.apache.org/jira/browse/QPID-2260), QPID-2247 (http://issues.apache.org/jira/browse/QPID-2247) - These are more problematic as they don't have anyone assigned to them. It could be that Cliff is working on them, but as he isn't a commiter he can't be assigned the bugs in Jira. - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org
RE: 0.6 Release Plans from now onwards
I came up against that a couple months ago. The reason seemed to be that the ConversationFactory is calling toString() on a message before its sent, and the toString() behaviour was changed from previously outputting the hash code to outputting actual content. (r756228 seems to be where you'd want to look). Robbie -Original Message- From: Aidan Skinner snip QPID-1830 is more problematic, I don't really have a good handle on what the issue there is yet. /snip - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org
RE: 0.6 Release Plans from now onwards
QPID-2260 (http://issues.apache.org/jira/browse/QPID-2260), QPID-2247 (http://issues.apache.org/jira/browse/QPID-2247) - These are more problematic as they don't have anyone assigned to them. It could be that Cliff is working on them, but as he isn't a commiter he can't be assigned the bugs in Jira. I assigned them to me. I'll most likely be handling any patches that result. I also assigned myself to QPID-2128 (https://issues.apache.org/jira/browse/QPID-2128) and elevated it to Critical. That code has been under review in the jira and resulting patches are forthcoming I believe. -Steve - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org
Re: 0.6 Release Plans from now onwards
On Tue, Dec 15, 2009 at 11:31 PM, Aidan Skinner aidan.skin...@gmail.com wrote: First sign of madness this, but never mind eh? On Tue, Dec 15, 2009 at 10:26 PM, Andrew Stitcher astitc...@redhat.com wrote: QPID-2273 (http://issues.apache.org/jira/browse/QPID-2273), QPID-1830 (http://issues.apache.org/jira/browse/QPID-1830) - Aidan can you give some status on these bugs? Are they going to be finished in the next couple of days? Is there much risk in the fixes? I'm currently working on QPID-2273, it's confined to the protocol initiation so should be low risk. I'll have a fix for that soon. Patch attached to the Jira. I'd like another set of eyes on this, it works as far as my testing goes but I'm not 100% convinced it gets all the cases. Otherwise I'll commit it Thursday since it's definitely better than what we have. - Aidan -- Apache Qpid - AMQP, JMS, other messaging love http://qpid.apache.org A witty saying proves nothing - Voltaire - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org
Re: 0.6 Release Plans from now onwards
On Tue, Dec 15, 2009 at 11:46 PM, Robbie Gemmell robbie.gemm...@gmail.com wrote: I came up against that a couple months ago. The reason seemed to be that the ConversationFactory is calling toString() on a message before its sent, and the toString() behaviour was changed from previously outputting the hash code to outputting actual content. (r756228 seems to be where you'd want to look). Ah, good spot. Reverting that patch fixes it for me. I'm going to assign the bug to Rajith since just backing it out doesn't seem entirely the right thing. - AIdan (patch review doesn't catch everything) -- Apache Qpid - AMQP, JMS, other messaging love http://qpid.apache.org A witty saying proves nothing - Voltaire - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org
Re: 0.6 Release Plans from now onwards
On Tue, Dec 15, 2009 at 10:08 PM, Aidan Skinner aidan.skin...@gmail.com wrote: On Tue, Dec 15, 2009 at 11:46 PM, Robbie Gemmell robbie.gemm...@gmail.com wrote: I came up against that a couple months ago. The reason seemed to be that the ConversationFactory is calling toString() on a message before its sent, and the toString() behaviour was changed from previously outputting the hash code to outputting actual content. (r756228 seems to be where you'd want to look). The Functions.str() method does not advance the position pointer so calling toString() should not affect the content. I will investigate this tomorrow morning and comment further once I figure out the problem. Ah, good spot. Reverting that patch fixes it for me. I'm going to assign the bug to Rajith since just backing it out doesn't seem entirely the right thing. - AIdan (patch review doesn't catch everything) -- Apache Qpid - AMQP, JMS, other messaging love http://qpid.apache.org A witty saying proves nothing - Voltaire - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org -- Regards, Rajith Attapattu Red Hat http://rajith.2rlabs.com/ - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org
RE: 0.6 Release Plans from now onwards
Cliff can you comment on these bugs? Are they truly as important as blocker/critical? If so unless there are patches attached to the jiras very soon I think we'll have to omit the specific WCF source code from this release [...] Agreed. These were marked as such because they would make a significant difference to early adopters and were thought to be doable *without* holding up the release. Right now, I feel comfortable about resolving QPID-2247 (and even QPID-2128) by tomorrow and QPID-2260 certainly by Thursday. The latter is very easy conceptually and boils down to testing in the various scenarios. Cliff -Original Message- From: Andrew Stitcher [mailto:astitc...@redhat.com] Sent: Tuesday, December 15, 2009 2:27 PM To: Qpid Dev List Subject: 0.6 Release Plans from now onwards As I said in my beta release message I'd like to produce a release candidate at the end of this week. On the closing bugs front there are a couple of issues still (but nothing that looks deadly as far as I can tell): Essentially we have 2 blockers and 2 critical bugs. 2 Java bugs and 2 bugs for the WCF code. QPID-2273 (http://issues.apache.org/jira/browse/QPID-2273), QPID-1830 (http://issues.apache.org/jira/browse/QPID-1830) - Aidan can you give some status on these bugs? Are they going to be finished in the next couple of days? Is there much risk in the fixes? QPID-2260 (http://issues.apache.org/jira/browse/QPID-2260), QPID-2247 (http://issues.apache.org/jira/browse/QPID-2247) - These are more problematic as they don't have anyone assigned to them. It could be that Cliff is working on them, but as he isn't a commiter he can't be assigned the bugs in Jira. Cliff can you comment on these bugs? Are they truly as important as blocker/critical? If so unless there are patches attached to the jiras very soon I think we'll have to omit the specific WCF source code from this release (of course it will still be in the big source tarball). To be honest I'm really sure it's quite mature enough in any case. Assuming that we can clear these issues up (one way or another) in the next couple of days - I plan to release rc1 on Thursday/Friday this week. I will create a (preemptive) subversion tag at that point. From that point I'd like to *freeze* checkins to the tree for 2 weeks taking us effectively to the new year. If we haven't actually voted to release by then I suggest we move the branch tag to whatever the most recent qpid checkin is and do all subsequent release work on a branch, and unfreeze the trunk. This way we should have the minimum frozen time and still be able to progress. Regards Andrew - Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:dev-subscr...@qpid.apache.org