Re: [announce] Release Wicket 1.4.9
The ITL's value is copied when a thread is created, so it will either: a. Never get copied to the pooled threads. b. Get copied one time into the pooled threads (the pooled thread is started during a request thread for some reason). If there are multiple Wicket applications running in the same webapp that share the thread pool, then they could potentially see the wrong application object bound to their task threads. On Fri, May 28, 2010 at 4:14 PM, Alex Objelean wrote: > > I'm not insisting on bringing it back, but I don't understand on what is > based your conclusion that it doesn't work for the desired use-case (thread > pools) ? > > Alex > -- > View this message in context: > http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2235138.html > Sent from the Wicket - User mailing list archive at Nabble.com. > > - > To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org > For additional commands, e-mail: users-h...@wicket.apache.org > > - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
I'm not insisting on bringing it back, but I don't understand on what is based your conclusion that it doesn't work for the desired use-case (thread pools) ? Alex -- View this message in context: http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2235138.html Sent from the Wicket - User mailing list archive at Nabble.com. - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
As I said when I determined the same thing (I attached a quickstart to the JIRA issue), the fact that it doesn't "leak" isn't really the biggest negative, IMHO. The fact that it doesn't work for the desired usecase (thread pools) is a huge negative and a reason it should be backed out (which it has already been). On Fri, May 28, 2010 at 3:55 PM, Alex Objelean wrote: > > I had the same results with tomcat 6... We could try to see what happens with > other web servers with combination of different jdk versions. Until now, > there was no prove of the memory leak. > -- > View this message in context: > http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2235109.html > Sent from the Wicket - User mailing list archive at Nabble.com. > > - > To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org > For additional commands, e-mail: users-h...@wicket.apache.org > > - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
I had the same results with tomcat 6... We could try to see what happens with other web servers with combination of different jdk versions. Until now, there was no prove of the memory leak. -- View this message in context: http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2235109.html Sent from the Wicket - User mailing list archive at Nabble.com. - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
If it's of any use, I put together a very, very simple 2D Quickstart for my own edification. I've deployed, exercised and un-deployed builds of application using Wicket 1.4.8 and 1.4.9 to Weblogic 10. In the former case I do not see, via YourKit, the Java2D Disposer thread holding a reference to the WicketApplication in ThreadLocalMap inheritableThreadLocals after the application has been undeployed, in the latter I do not. In the 1.4.8 test the only reference to WicketApplication is to the Class, not an instance as I'm seeing in the 1.4.9 case. Ed. On Tue, May 25, 2010 at 7:21 AM, James Carman wrote: > It is *not* a theoretical objection. Please see: > > http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6489540 > > This is a real problem that Sun (now Oracle I guess) has identified > and is "working on". > > On Mon, May 24, 2010 at 5:20 PM, Alex Objelean > wrote: > > > > I don't want to insist to much, I'm not absurd, but we are technical > people. > > Don't you think that any theorem should be proven? The least we can > achieve > > is to learn a new thing about how ITL are related to memory leaks. > > > > I know it isn't easy to prove, but aren't there enough tools to help us? > Is > > it really that hard to make a simple example with a worst case scenario > to > > check if the presumption is indeed valid? I know it is much simpler to do > > nothing to avoid potential problems, but still do you find it a good way > to > > deal with problems? Don't you believe in presumption of innocence > principle? > > > > I don't want to bother the community, if there won't anybody willing to > > prove the problem, eventually I will spend some time to bust the myth... > > > > Alex > > > > -- > > View this message in context: > http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2229213.html > > Sent from the Wicket - User mailing list archive at Nabble.com. > > > > - > > To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org > > For additional commands, e-mail: users-h...@wicket.apache.org > > > > > > - > To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org > For additional commands, e-mail: users-h...@wicket.apache.org > >
Re: [announce] Release Wicket 1.4.9
Hi Sven! See the task description details: https://issues.apache.org/jira/browse/WICKET-2846 Alex -- View this message in context: http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2229836.html Sent from the Wicket - User mailing list archive at Nabble.com. - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
It is *not* a theoretical objection. Please see: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6489540 This is a real problem that Sun (now Oracle I guess) has identified and is "working on". On Mon, May 24, 2010 at 5:20 PM, Alex Objelean wrote: > > I don't want to insist to much, I'm not absurd, but we are technical people. > Don't you think that any theorem should be proven? The least we can achieve > is to learn a new thing about how ITL are related to memory leaks. > > I know it isn't easy to prove, but aren't there enough tools to help us? Is > it really that hard to make a simple example with a worst case scenario to > check if the presumption is indeed valid? I know it is much simpler to do > nothing to avoid potential problems, but still do you find it a good way to > deal with problems? Don't you believe in presumption of innocence principle? > > I don't want to bother the community, if there won't anybody willing to > prove the problem, eventually I will spend some time to bust the myth... > > Alex > > -- > View this message in context: > http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2229213.html > Sent from the Wicket - User mailing list archive at Nabble.com. > > - > To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org > For additional commands, e-mail: users-h...@wicket.apache.org > > - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
Sorry, what was the use case after all? Sven -- View this message in context: http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2229772.html Sent from the Wicket - User mailing list archive at Nabble.com. - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
On Mon, May 24, 2010 at 4:20 PM, Alex Objelean wrote: > > I don't want to insist to much, I'm not absurd, but we are technical > people. > Don't you think that any theorem should be proven? The least we can achieve > is to learn a new thing about how ITL are related to memory leaks. > > I know it isn't easy to prove, but aren't there enough tools to help us? Is > it really that hard to make a simple example with a worst case scenario to > check if the presumption is indeed valid? I know it is much simpler to do > nothing to avoid potential problems, but still do you find it a good way to > deal with problems? Don't you believe in presumption of innocence > principle? > > I don't want to bother the community, if there won't anybody willing to > prove the problem, eventually I will spend some time to bust the myth... > > I'd be happy to see you bust it. But, like you said - we are technical people - which is why we can postulate certain things without writing a complex test case to prove them. -- Jeremy Thomerson http://www.wickettraining.com
Re: [announce] Release Wicket 1.4.9
I don't want to insist to much, I'm not absurd, but we are technical people. Don't you think that any theorem should be proven? The least we can achieve is to learn a new thing about how ITL are related to memory leaks. I know it isn't easy to prove, but aren't there enough tools to help us? Is it really that hard to make a simple example with a worst case scenario to check if the presumption is indeed valid? I know it is much simpler to do nothing to avoid potential problems, but still do you find it a good way to deal with problems? Don't you believe in presumption of innocence principle? I don't want to bother the community, if there won't anybody willing to prove the problem, eventually I will spend some time to bust the myth... Alex -- View this message in context: http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2229213.html Sent from the Wicket - User mailing list archive at Nabble.com. - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
I find the argument: a thread started outside the control of the programmer by the JRE or a library to do some processing (e.g. rendering an image), which inherits Wicket's thread local, causing a leak of our thread locals rather convincing. The result would be redeploy problems, ultimately causing the dreaded OutOfMemoryError: permgen. Is this hard to prove? Sure it is, therefore I don't need proof to see if this is worse than the (limited) convenience of InheritableThreadLocal. Martijn On Mon, May 24, 2010 at 10:46 PM, Alex Objelean wrote: > > When comparing a small feature (but still feature) proven by an use case > (limited but still an use case) and a NON problem proven only with > theoretical presumption (with also very limited use case), would you still > choose reverting it? Same question for all who voted against it... > > Alex > -- > View this message in context: > http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2229159.html > Sent from the Wicket - User mailing list archive at Nabble.com. > > - > To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org > For additional commands, e-mail: users-h...@wicket.apache.org > > -- Become a Wicket expert, learn from the best: http://wicketinaction.com Apache Wicket 1.4 increases type safety for web applications Get it now: http://www.apache.org/dyn/closer.cgi/wicket/1.4.8 - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
When comparing a small feature (but still feature) proven by an use case (limited but still an use case) and a NON problem proven only with theoretical presumption (with also very limited use case), would you still choose reverting it? Same question for all who voted against it... Alex -- View this message in context: http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2229159.html Sent from the Wicket - User mailing list archive at Nabble.com. - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
On Mon, May 24, 2010 at 3:24 PM, Alex Objelean wrote: > > Hi Jeremy! > Thanks for the effort put in this release. Though there is a lot of > discussion around WICKET-2846, instead of deciding to revert it in the next > release, I would rather suggest to take the chance for those who voted for > reverting it to prove that it is indeed the issue. > > I am still skeptical about the problem and I think that reverting a non > could create a precedent which would encourage users to do the same without > needing to prove nothing in the problem. > > Thanks! > Alex > Thanks for your input - please also add it to the vote (vote for reverting 2846) thread. The community is overwhelmingly saying that they want it reverted. I am still +0 for it - meaning I don't care. While you're right that nobody has proven that it *creates* a bug, we do know that it will exacerbate an existing bug that you may already have. That doesn't really bother me as much as some say that it should. But, I also see very little benefit to the InheritableThreadLocal since it really only serves a very small subset of cases. If it was more advantageous to a wider set of use cases, I'd be in favor of keeping it. -- Jeremy Thomerson http://www.wickettraining.com
Re: [announce] Release Wicket 1.4.9
Hi Jeremy! Thanks for the effort put in this release. Though there is a lot of discussion around WICKET-2846, instead of deciding to revert it in the next release, I would rather suggest to take the chance for those who voted for reverting it to prove that it is indeed the issue. I am still skeptical about the problem and I think that reverting a non could create a precedent which would encourage users to do the same without needing to prove nothing in the problem. Thanks! Alex -- View this message in context: http://apache-wicket.1842946.n4.nabble.com/announce-Release-Wicket-1-4-9-tp2228179p2229134.html Sent from the Wicket - User mailing list archive at Nabble.com. - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
2010/5/24 Michał Letyński > Hi. > W dniu 2010-05-24 03:59, Jeremy Thomerson pisze: > > ** Improvement >> * [WICKET-2790] - wicketTester.executeAjaxEvent method does not check >> if >> form is multiPart >> >> > Is this improvment a part of 1.4.9 ? Since on jira page this improvment is > postponed to 1.4.10. > Sorry, that is a mistake. WICKET-2790 is NOT a part of 1.4.9. It looks like someone set the "fix version" when they first created the issue rather than the "affects version". I've updated JIRA to reflect that it is not in a fixed version yet. -- Jeremy Thomerson http://www.wickettraining.com
Re: [announce] Release Wicket 1.4.9
Hi. W dniu 2010-05-24 03:59, Jeremy Thomerson pisze: ** Improvement * [WICKET-2790] - wicketTester.executeAjaxEvent method does not check if form is multiPart Is this improvment a part of 1.4.9 ? Since on jira page this improvment is postponed to 1.4.10. - To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org For additional commands, e-mail: users-h...@wicket.apache.org
Re: [announce] Release Wicket 1.4.9
Note that there seems to be some delay on some mirrors (including the maven repos). I'm working out the details now, but most of the mirrors (excluding maven) have the artifacts available. In the meantime, if you really need 1.4.9, you could use the maven repo that was sent in the vote message (temporarily): http://people.apache.org/~jrthomerson/releases/apache-wicket-1.4.9/m2-repo/ -- Jeremy Thomerson http://www.wickettraining.com On Sun, May 23, 2010 at 8:59 PM, Jeremy Thomerson wrote: > Wicket 1.4.9 is released! > > This is the ninth maintenance release of the 1.4.x series and brings > over fifteen bug fixes and improvements. > > Tag: http://svn.apache.org/repos/asf/wicket/releases/wicket-1.4.9/ > Changelog: > https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12310561&styleName=Html&version=12314962 > (and pasted below) > > To use in Maven: > > > org.apache.wicket > wicket > 1.4.9 > > > To download: http://www.apache.org/dyn/closer.cgi/wicket/1.4.9 > > It is worth noting that WICKET-2846 will almost definitely be reverted in > the next release because the community has shown a preference for this. So, > it would be wise not to take advantage of the InheritableThreadLocal that > was put into version 1.4.9. > > Best Regards, > > -- > Jeremy Thomerson > http://www.wickettraining.com > > PS - Release Notes - Wicket - Version 1.4.9 > > ** Bug > * [WICKET-2741] - non-performant Collections.synchronizedMap() should > be replaced with ConcurrentMap > * [WICKET-2843] - Palette is incompatible with ListMultipleChoice in > its use of the model > * [WICKET-2853] - ListMultipleChoice/CheckBoxMultipleChoice do not > retain selected but disabled items > * [WICKET-2856] - PackageStringResourceLoader.loadStringResource() > causes NullPointerException when used in a Class within the root package > (i.e. it has no package declaration) > * [WICKET-2858] - WicketSessionFilter: > java.lang.IllegalArgumentException: Argument application can not be null > * [WICKET-2859] - Wrong package names in Examples > * [WICKET-2860] - Wrong name for swiss Application.properties > * [WICKET-2861] - getConvertedInput() returns null and > selectedValues.addAll tries adding it > > ** Improvement > * [WICKET-2790] - wicketTester.executeAjaxEvent method does not check > if form is multiPart > * [WICKET-2840] - Remove final on > AbstractRequestTargetUrlCodingStrategy#getMountPath() > * [WICKET-2846] - Store Application in InheritableThreadLocal instead > of ThreadLocal > * [WICKET-2855] - Constructor of RedirectRequestTarget does not > validate URL > * [WICKET-2869] - RangeValidator should use getMinimum and getMaximum > * [WICKET-2870] - Fix hungarian translation for Wizard > * [WICKET-2879] - delegate isVisible in PanelCachingTab > >
[announce] Release Wicket 1.4.9
Wicket 1.4.9 is released! This is the ninth maintenance release of the 1.4.x series and brings over fifteen bug fixes and improvements. Tag: http://svn.apache.org/repos/asf/wicket/releases/wicket-1.4.9/ Changelog: https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12310561&styleName=Html&version=12314962 (and pasted below) To use in Maven: org.apache.wicket wicket 1.4.9 To download: http://www.apache.org/dyn/closer.cgi/wicket/1.4.9 It is worth noting that WICKET-2846 will almost definitely be reverted in the next release because the community has shown a preference for this. So, it would be wise not to take advantage of the InheritableThreadLocal that was put into version 1.4.9. Best Regards, -- Jeremy Thomerson http://www.wickettraining.com PS - Release Notes - Wicket - Version 1.4.9 ** Bug * [WICKET-2741] - non-performant Collections.synchronizedMap() should be replaced with ConcurrentMap * [WICKET-2843] - Palette is incompatible with ListMultipleChoice in its use of the model * [WICKET-2853] - ListMultipleChoice/CheckBoxMultipleChoice do not retain selected but disabled items * [WICKET-2856] - PackageStringResourceLoader.loadStringResource() causes NullPointerException when used in a Class within the root package (i.e. it has no package declaration) * [WICKET-2858] - WicketSessionFilter: java.lang.IllegalArgumentException: Argument application can not be null * [WICKET-2859] - Wrong package names in Examples * [WICKET-2860] - Wrong name for swiss Application.properties * [WICKET-2861] - getConvertedInput() returns null and selectedValues.addAll tries adding it ** Improvement * [WICKET-2790] - wicketTester.executeAjaxEvent method does not check if form is multiPart * [WICKET-2840] - Remove final on AbstractRequestTargetUrlCodingStrategy#getMountPath() * [WICKET-2846] - Store Application in InheritableThreadLocal instead of ThreadLocal * [WICKET-2855] - Constructor of RedirectRequestTarget does not validate URL * [WICKET-2869] - RangeValidator should use getMinimum and getMaximum * [WICKET-2870] - Fix hungarian translation for Wizard * [WICKET-2879] - delegate isVisible in PanelCachingTab