the page should always be stored. i think we made a bad optimization if it is not. just because the page is not dirty doesnt actually mean it is not changed. eg, ajax requests are "unversioned" because we do not rewrite urls outside the ajax repainted regions. but, just because an ajax url did not mark the page as "dirty" doesnt mean it should not be stored. it should also always be put into the session attribute so it triggers replication if necessary.
-igor On Fri, May 6, 2011 at 12:00 PM, Martin Grigorov <mgrigo...@apache.org> wrote: > It seems I'll have to revert the "improvement" for your ticket (WICKET-3667). > https://issues.apache.org/jira/browse/WICKET-3676 is broken because of > this change. > > AFAIK Wicket should not store the page for Ajax requests (this is what > I know from the other devs) but apparently this is not quite true. > > I tried the app in 3676 in latest 1.4.x and it works fine. So I > debugged it and it seems Wicket 1.4.x actually stores the page after > Ajax request. > So now I'm a confused what to do. > > @Other devs: any advises ? > > On Wed, May 4, 2011 at 5:09 PM, nino martinez wael > <nino.martinez.w...@gmail.com> wrote: >> ahh saw that you continued the commentary on the issue.. Thanks >> >> 2011/5/4 nino martinez wael <nino.martinez.w...@gmail.com>: >>> Any time frame on when you will be able to test it? >>> >>> 2011/5/4 nino martinez wael <nino.martinez.w...@gmail.com>: >>>> Done and reproduced.. >>>> >>>> https://issues.apache.org/jira/browse/WICKET-3667 >>>> >>>> 2011/5/4 nino martinez wael <nino.martinez.w...@gmail.com>: >>>>> Hmm I startet wondering if it could be something with pageversioning >>>>> if we have 12 windows opened AFAIK pageversioning are set to 15 by >>>>> default per session? So all these windows keep refreshing every 2 >>>>> seconds, some of them are bound to go over the limit... >>>>> >>>>> I'll look into doing the quickstart and see if I can reproduce it.. >>>>> >>>>> regards Nino >>>>> >>>>> 2011/5/4 Martin Grigorov <mgrigo...@apache.org>: >>>>>> create a quickstart and describe the exact steps to reproduce it >>>>>> all this in a ticket >>>>>> >>>>>> On Wed, May 4, 2011 at 1:05 PM, nino martinez wael >>>>>> <nino.martinez.w...@gmail.com> wrote: >>>>>>> It did not seem to do a difference... >>>>>>> >>>>>>> 2011/5/4 nino martinez wael <nino.martinez.w...@gmail.com>: >>>>>>>> Ok overriden the defaultpageprovider and set everything to 10x of the >>>>>>>> original implementation that should make a change correct if this is >>>>>>>> the problem? >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> 2011/5/4 Martin Grigorov <mgrigo...@apache.org>: >>>>>>>>> The session is not timed out. >>>>>>>>> You see PageExpirationException. *Page*! >>>>>>>>> That means the page instance is not in the page store. >>>>>>>>> Reasons: >>>>>>>>> - it storing failed so the page never been in the store >>>>>>>>> - it has been removed from the store because of other reasons - like >>>>>>>>> the size for the session related data reached the maximum allowed and >>>>>>>>> the new page instances overwrote the old ones >>>>>>>>> see org.apache.wicket.DefaultPageManagerProvider >>>>>>>>> >>>>>>>>> On Wed, May 4, 2011 at 11:46 AM, nino martinez wael >>>>>>>>> <nino.martinez.w...@gmail.com> wrote: >>>>>>>>>> Ok now we are getting something.. >>>>>>>>>> >>>>>>>>>> I can see that the session are being created (one session for the 12 >>>>>>>>>> browser instances). But I never get the session destroyed log info. >>>>>>>>>> However wicket still claims that the session are timedout.. How can >>>>>>>>>> that be? >>>>>>>>>> >>>>>>>>>> regards Nino >>>>>>>>>> >>>>>>>>>> 2011/5/4 Martin Grigorov <mgrigo...@apache.org>: >>>>>>>>>>> Servlet API - HttpSessionBindingListener >>>>>>>>>>> >>>>>>>>>>> On Wed, May 4, 2011 at 10:33 AM, nino martinez wael >>>>>>>>>>> <nino.martinez.w...@gmail.com> wrote: >>>>>>>>>>>> hmm I did see something about the request logger being enabled.. >>>>>>>>>>>> could >>>>>>>>>>>> be a version mismatch.. Can I get information about when sessions >>>>>>>>>>>> are >>>>>>>>>>>> created/deleted? >>>>>>>>>>>> >>>>>>>>>>>> 2011/5/4 nino martinez wael <nino.martinez.w...@gmail.com>: >>>>>>>>>>>>> but what about the dummyrequestlogger information? >>>>>>>>>>>>> >>>>>>>>>>>>> 2011/5/4 Martin Grigorov <mgrigo...@apache.org>: >>>>>>>>>>>>>> See >>>>>>>>>>>>>> org.apache.wicket.settings.IRequestLoggerSettings.isRequestLoggerEnabled() >>>>>>>>>>>>>> >>>>>>>>>>>>>> On Wed, May 4, 2011 at 10:18 AM, nino martinez wael >>>>>>>>>>>>>> <nino.martinez.w...@gmail.com> wrote: >>>>>>>>>>>>>>> We are seeing a lot of these on the customers server >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> 2011-05-04 10:13:25,843 INFO ["http-bio-8080"-exec-3] >>>>>>>>>>>>>>> (DummyRequestLogger.java:367) - >>>>>>>>>>>>>>> time=31,event=null,response=null,sessionid=034801D1024C79B62FD60ACE7C0B2110,sessionsize=1456,sessionstart=Wed >>>>>>>>>>>>>>> May 04 10:12:08 CEST >>>>>>>>>>>>>>> 2011,requests=131,totaltime=5190,activerequests=0,maxmem=259M,total=19M,used=11M >>>>>>>>>>>>>>> 2011-05-04 10:13:26,484 INFO ["http-bio-8080"-exec-10] >>>>>>>>>>>>>>> (DummyRequestLogger.java:367) - >>>>>>>>>>>>>>> time=16,event=null,response=null,sessionid=034801D1024C79B62FD60ACE7C0B2110,sessionsize=1456,sessionstart=Wed >>>>>>>>>>>>>>> May 04 10:12:08 CEST >>>>>>>>>>>>>>> 2011,requests=132,totaltime=5206,activerequests=0,maxmem=259M,total=19M,used=13M >>>>>>>>>>>>>>> 2011-05-04 10:13:26,703 INFO ["http-bio-8080"-exec-9] >>>>>>>>>>>>>>> (DummyRequestLogger.java:367) - >>>>>>>>>>>>>>> time=47,event=null,response=null,sessionid=034801D1024C79B62FD60ACE7C0B2110,sessionsize=1456,sessionstart=Wed >>>>>>>>>>>>>>> May 04 10:12:08 CEST >>>>>>>>>>>>>>> 2011,requests=133,totaltime=5253,activerequests=1,maxmem=259M,total=19M,used=15M >>>>>>>>>>>>>>> 2011-05-04 10:13:26,703 INFO ["http-bio-8080"-exec-3] >>>>>>>>>>>>>>> (DummyRequestLogger.java:367) - >>>>>>>>>>>>>>> time=47,event=null,response=null,sessionid=034801D1024C79B62FD60ACE7C0B2110,sessionsize=1456,sessionstart=Wed >>>>>>>>>>>>>>> May 04 10:12:08 CEST >>>>>>>>>>>>>>> 2011,requests=134,totaltime=5300,activerequests=0,maxmem=259M,total=19M,used=16M >>>>>>>>>>>>>>> 2011-05-04 10:13:26,968 INFO ["http-bio-8080"-exec-10] >>>>>>>>>>>>>>> (DummyRequestLogger.java:367) - >>>>>>>>>>>>>>> time=31,event=null,response=null,sessionid=034801D1024C79B62FD60ACE7C0B2110,sessionsize=1456,sessionstart=Wed >>>>>>>>>>>>>>> May 04 10:12:08 CEST >>>>>>>>>>>>>>> 2011,requests=135,totaltime=5331,activerequests=0,maxmem=259M,total=19M,used=12M >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> On our demo server they do not occur.. >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> 2011/5/4 nino martinez wael <nino.martinez.w...@gmail.com>: >>>>>>>>>>>>>>>> Yes. >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> We have several browsers open on the same machine, and >>>>>>>>>>>>>>>> suddenly after >>>>>>>>>>>>>>>> a few seconds a couple of the browsers gets the session >>>>>>>>>>>>>>>> timeout page. >>>>>>>>>>>>>>>> It's very strange. >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> What we have tried so far are FF4, IE6 Tomcat 7.11 and jetty >>>>>>>>>>>>>>>> 7.4. >>>>>>>>>>>>>>>> Removing the filterfix for tomcat. Using wicket rc2 and rc3. No >>>>>>>>>>>>>>>> difference there, it acts the same. No information in logs, >>>>>>>>>>>>>>>> they are >>>>>>>>>>>>>>>> set to INFO. >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> However on our demo server it works without any problems, we >>>>>>>>>>>>>>>> do not >>>>>>>>>>>>>>>> get the error there. Is there any log I could enable that >>>>>>>>>>>>>>>> would help? >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> regards Nino >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> 2011/5/4 Martin Grigorov <mgrigo...@apache.org>: >>>>>>>>>>>>>>>>> Can you explain what exactly happens ? >>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>> On Wed, May 4, 2011 at 9:06 AM, nino martinez wael >>>>>>>>>>>>>>>>> <nino.martinez.w...@gmail.com> wrote: >>>>>>>>>>>>>>>>>> Hmm tried with ff and problem reoccurs.. we are doing >>>>>>>>>>>>>>>>>> further testing.. >>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>> 2011/5/3 nino martinez wael <nino.martinez.w...@gmail.com>: >>>>>>>>>>>>>>>>>>> now testing here on our demo server using ff chrome and >>>>>>>>>>>>>>>>>>> IE8.. One >>>>>>>>>>>>>>>>>>> thing springs to mind, we tested on the customers server >>>>>>>>>>>>>>>>>>> which has IE6 >>>>>>>>>>>>>>>>>>> installed, i'll get ff installed tomorrow and retest the >>>>>>>>>>>>>>>>>>> scenario.. >>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>> 2011/5/3 nino martinez wael <nino.martinez.w...@gmail.com>: >>>>>>>>>>>>>>>>>>>> BTW are running on tomcat 7 using the filterfix for non us >>>>>>>>>>>>>>>>>>>> characters.. >>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>> 2011/5/3 nino martinez wael <nino.martinez.w...@gmail.com>: >>>>>>>>>>>>>>>>>>>>> Hi >>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>> I am experience random timeouts during ajax requests(I >>>>>>>>>>>>>>>>>>>>> have a 2 s >>>>>>>>>>>>>>>>>>>>> heartbeat going), no errors in logs or anything.. This >>>>>>>>>>>>>>>>>>>>> it both with >>>>>>>>>>>>>>>>>>>>> rc3 and rc2.. We had 10 browser windows opnened on the >>>>>>>>>>>>>>>>>>>>> machine.. But >>>>>>>>>>>>>>>>>>>>> are now trying with only one browser to see what happens >>>>>>>>>>>>>>>>>>>>> during the >>>>>>>>>>>>>>>>>>>>> night.. >>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>> Has anybody seen something similar? >>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>> regards Nino >>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>> --------------------------------------------------------------------- >>>>>>>>>>>>>>>>>> To unsubscribe, e-mail: users-unsubscr...@wicket.apache.org >>>>>>>>>>>>>>>>>> For additional commands, e-mail: users-h...@wicket.apache.org >>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>> -- >>>>>>>>>>>>>>>>> Martin Grigorov >>>>>>>>>>>>>>>>> jWeekend >>>>>>>>>>>>>>>>> Training, Consulting, Development >>>>>>>>>>>>>>>>> http://jWeekend.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 >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> -- >>>>>>>>>>>>>> Martin Grigorov >>>>>>>>>>>>>> jWeekend >>>>>>>>>>>>>> Training, Consulting, Development >>>>>>>>>>>>>> http://jWeekend.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 >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> -- >>>>>>>>>>> Martin Grigorov >>>>>>>>>>> jWeekend >>>>>>>>>>> Training, Consulting, Development >>>>>>>>>>> http://jWeekend.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 >>>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> -- >>>>>>>>> Martin Grigorov >>>>>>>>> jWeekend >>>>>>>>> Training, Consulting, Development >>>>>>>>> http://jWeekend.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 >>>>>>> >>>>>>> >>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> Martin Grigorov >>>>>> jWeekend >>>>>> Training, Consulting, Development >>>>>> http://jWeekend.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 >> >> > > > > -- > Martin Grigorov > jWeekend > Training, Consulting, Development > http://jWeekend.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
