Hello Jacopo, I have made a partial revert in r1767866 which now removes the bug that I encountered with the classpath exceptions being thrown. This fueled my hate even more to the custom classloader logic which I'd love to cut from the roots.
So with that being said, I think we can go ahead and prepare a release unless someone else is working on some issues. Taher Alkhateeb On Mon, Oct 31, 2016 at 1:03 PM, Taher Alkhateeb <[email protected] > wrote: > Thank you Jacques, I will work some more towards resolving this and keep > you posted. > > On Mon, Oct 31, 2016 at 12:59 PM, Jacques Le Roux < > [email protected]> wrote: > >> Hi Taher, >> >> At least here it did not work, same issue with same files when debugging, >> after clearing the all Gradle cache and doing a build >> >> Jacques >> >> >> >> Le 28/10/2016 à 12:44, Taher Alkhateeb a écrit : >> >>> Hi Jacques, >>> >>> No I did not try to clear the caches, but that is a good point, perhaps >>> clearing would work. I'll try that next. >>> >>> Taher Alkhateeb >>> >>> On Fri, Oct 28, 2016 at 1:39 PM, Jacques Le Roux < >>> [email protected]> wrote: >>> >>> Hi Taher, >>>> >>>> I currently work in debug mode and I indeed get >>>> java.io.FileNotFoundException for: >>>> >>>> .gradle\caches\modules-2\files-2.1\edu.ucar\netcdf4\4.5.5\67 >>>> 5d63ecc857c50dd50858011b670160aa30b62\jcl-over-slf4j-1.7.7.jar >>>> .gradle\caches\modules-2\files-2.1\edu.ucar\netcdf4\4.5.5\67 >>>> 5d63ecc857c50dd50858011b670160aa30b62\ehcache-core-2.6.2.jar >>>> >>>> several times (3 each IIRW) >>>> >>>> But only those files are concerned. I thought it was only a matter of >>>> clearing Gradle caches to update there. I did not try because it's just >>>> annoying and reloading Internet again is even more. Did you try to clean >>>> the caches? >>>> >>>> Jacques >>>> >>>> >>>> >>>> Le 28/10/2016 à 11:45, Taher Alkhateeb a écrit : >>>> >>>> Hi Jacopo, >>>>> >>>>> I have discovered a regression that I introduced in r1765127. If you >>>>> run >>>>> OFBiz in debug mode it will throw a "FileNotFoundException". This >>>>> happens >>>>> when the ContainerLoader calls the init method of the >>>>> CatalinaContainer. >>>>> I'm not yet exactly sure why, and at worst case scenario I might have >>>>> to >>>>> revert the entire commit (it's a big 2000 lines unfortunately). So I >>>>> will >>>>> do my best in the few coming days to try and debug the root cause. It >>>>> could >>>>> be the class loader but I'm just not sure, the old code was so messy >>>>> it's >>>>> hard to pinpoint the root cause. >>>>> >>>>> I'll keep you posted but please don't release before we take a >>>>> decision on >>>>> this regression. Anyone willing to help is more than welcome to pitch >>>>> in. >>>>> >>>>> On Thu, Oct 27, 2016 at 3:20 PM, Jacopo Cappellato < >>>>> [email protected]> wrote: >>>>> >>>>> Thank you Jacques, I agree that delaying of 1-2 weeks would not be a >>>>> big >>>>> >>>>>> deal. >>>>>> If anyone is planning to contribute improvements in the next 1-2 weeks >>>>>> then >>>>>> please speak up and we could definitely wait for them. >>>>>> >>>>>> Jacopo >>>>>> >>>>>> On Thu, Oct 27, 2016 at 1:40 PM, Jacques Le Roux < >>>>>> [email protected]> wrote: >>>>>> >>>>>> Hi Jacopo, All. >>>>>> >>>>>>> Are there no major improvements that we would wish to have before >>>>>>> cutting >>>>>>> the branch? >>>>>>> >>>>>>> I personally don't remember any. But with the 6 last months turmoil >>>>>>> we >>>>>>> increased from around 0 issues created vs resolved to 1840 created >>>>>>> and >>>>>>> >>>>>>> 1447 >>>>>> >>>>>> resolved. >>>>>>> So we have now a differential of around 400 issues pending for >>>>>>> >>>>>>> resolution. >>>>>> >>>>>> https://issues.apache.org/jira/secure/Dashboard.jspa? >>>>>>> >>>>>>> selectPageId=12310800 >>>>>> >>>>>> I know that most of them are minor subtasks, and it's hard to follow >>>>>>> the >>>>>>> flow. >>>>>>> >>>>>>> Anyway if you wait for an improvement it's the moment to raise a hand >>>>>>> >>>>>>> Thanks >>>>>>> >>>>>>> Jacques >>>>>>> >>>>>>> >>>>>>> >>>>>>> Le 24/10/2016 à 10:40, Jacopo Cappellato a écrit : >>>>>>> >>>>>>> Ok, it is probably time to proceed with this plan: >>>>>>> >>>>>>>> 1) create a new release branch named "release16.10" >>>>>>>> 2) the target goal could be that of issuing a release out of it >>>>>>>> sometime >>>>>>>> in >>>>>>>> the next month ("Apache OFBiz 16.10.01") >>>>>>>> >>>>>>>> Jacopo >>>>>>>> >>>>>>>> On Wed, Sep 7, 2016 at 1:26 PM, Taher Alkhateeb < >>>>>>>> [email protected]> >>>>>>>> wrote: >>>>>>>> >>>>>>>> +1 >>>>>>>> >>>>>>>> On Wed, Sep 7, 2016 at 2:07 PM, Sharan Foga <[email protected]> >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>> That's great news ! +1 >>>>>>>>> >>>>>>>>> Thanks >>>>>>>>>> Sharan >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> On 07/09/16 12:06, Jacopo Cappellato wrote: >>>>>>>>>> >>>>>>>>>> Hi all, >>>>>>>>>> >>>>>>>>>> I think it is the right time to create a new release branch out of >>>>>>>>>>> >>>>>>>>>>> the >>>>>>>>>> >>>>>>>>> trunk. >>>>>>> >>>>>>>> According to our naming conventions the release will be named >>>>>>>>>>> "release16.09". >>>>>>>>>>> >>>>>>>>>>> Any objections or concerns? If not, I will create the branch >>>>>>>>>>> later >>>>>>>>>>> today/tomorrow. >>>>>>>>>>> >>>>>>>>>>> After that we will start, in a separate thread, the discussion >>>>>>>>>>> about >>>>>>>>>>> the >>>>>>>>>>> preparation of our first release out of this branch. >>>>>>>>>>> >>>>>>>>>>> Thanks, >>>>>>>>>>> >>>>>>>>>>> Jacopo >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >> >
