Re: Deadlock in IOC (ConcurrentBarrier vs JustInTimeObjectCreator)

2020-08-17 Thread Thiago H. de Paula Figueiredo
On Mon, Aug 17, 2020 at 6:42 AM Oliver Kaiser wrote: > sure; that seems to fix it > Thank *you* for doing almost the whole work here: spotting the issue, creating a test case, suggesting the fix (which was exactly what you suggested) then confirming it. :) > > thank you, > Oliver > > -

Re: Deadlock in IOC (ConcurrentBarrier vs JustInTimeObjectCreator)

2020-08-17 Thread Oliver Kaiser
On 08/16/2020 10:20 PM, Thiago H. de Paula Figueiredo wrote: I've committed a fix using your code as a test and uploaded a new Tapestry 5.6.0 snapshot to the ASF snapshot repository at https://repository.apache.org/content/groups/snapshots/. Could you please test on your end too? Thanks in advan

Re: Deadlock in IOC (ConcurrentBarrier vs JustInTimeObjectCreator)

2020-08-16 Thread Thiago H. de Paula Figueiredo
Hello, Oliver! I've committed a fix using your code as a test and uploaded a new Tapestry 5.6.0 snapshot to the ASF snapshot repository at https://repository.apache.org/content/groups/snapshots/. Could you please test on your end too? Thanks in advance. On Sat, Aug 15, 2020 at 3:08 PM Thiago H. d

Re: Deadlock in IOC (ConcurrentBarrier vs JustInTimeObjectCreator)

2020-08-15 Thread Thiago H. de Paula Figueiredo
Hello! It's a very specific corner case. Please create a JIRA ticket and I'll try to fix it. Thanks for posting and investigating! On Thu, Jul 30, 2020 at 1:44 PM Oliver Kaiser wrote: > Hi everyone, > > > we recently found a strange deadlock problem; first seemed to be related > to switching fr

Deadlock in IOC (ConcurrentBarrier vs JustInTimeObjectCreator)

2020-07-30 Thread Oliver Kaiser
Hi everyone, we recently found a strange deadlock problem; first seemed to be related to switching from 5.5.0-beta-1 to 5.5.0, at least in the full application it was easy to reproduce with the latter but (in a few manual tests) not with the beta. after trying to reduce it to a minimal test