You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@wicket.apache.org by carloc <ca...@yahoo.com> on 2009/01/12 16:43:23 UTC
sessionsize of requestlogger
Hi Everyone,
RequestLogger -
time=103,event=Interface[target:AjaxPagingNavigationLink(userListForm:userGroupContainer:userGroupCheckGroup:ajaxPagingNavigation:navigation:9:pageLink),
page: com.ccti.base.web.usergroup.VedUserGroupPage(2), interface:
IBehaviorListener.onRequest],response=PageRequest[com.ccti.base.web.usergroup.VedUserGroupPage(2)],sessionid=kdrq2wp0eaea,sessionsize=212417,sessionstart=Mon
Jan 12 23:24:44 CST
2009,requests=14,totaltime=7024,activerequests=1,maxmem=66M,total=18M,used=14M
sessionsize=212417
I'm particularly interested of the sessionsize that is being logged by the
requestlogger.
I was wondering what size does it contain?
In One of My List Pages,
It increases by about 72000 when I put in an ajaxpagingnavigator.
I was just wondering what things are put into the session?
One thing that bothers me though is I tried using a LoadableDetachableModel
in my dataview.
It returned the almost the same sessionsize for a dataview in which I used a
CompoundDetachableMOdel.
I tried viewing the page map using eclipse's debug, I can see that it keeps
a reference to the previous page, so does it mean that Wicket always keeps
the last page available in the HttpSession? or does it keep the current page
in the session.
If I try exploring the session, I can see that the loadableDetachableModels
model object cannot be viewed.
And the compoundpropertymodel's attributes can be viewed.
As far as I can see, the wicket components consume significant session size
. Please correct me if I am wrong.
So question is,
What Page Does Wicket Store in its HttpSession? How Many Pages Does It
Store?
What consumes significant session size in wicket?
Is it bad to keep a reference to the previous pages in instance variables of
new pages?
Carlo
--
View this message in context: http://www.nabble.com/sessionsize-of-requestlogger-tp21416562p21416562.html
Sent from the Wicket - User mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org
Re: sessionsize of requestlogger
Posted by Jeremy Thomerson <je...@wickettraining.com>.
Session size really depends on the types of models you use. For instance,
if you have a list of items in some sort of repeating view, perhaps your
model that loads the list is loadable / detachable (good), but you might do
something like "new Model(rowItem)" around each item in, for instance, your
IDataProvider, which then holds each item indefinitely - storing each item
in session (bad).
If you have one page or component that drastically increases session size,
you should use a memory profiler to see what your most prolific objects
are. You might be surprised at what you're still holding on to.
On Mon, Jan 12, 2009 at 9:43 AM, carloc <ca...@yahoo.com> wrote:
>
> Hi Everyone,
>
> RequestLogger -
>
> time=103,event=Interface[target:AjaxPagingNavigationLink(userListForm:userGroupContainer:userGroupCheckGroup:ajaxPagingNavigation:navigation:9:pageLink),
> page: com.ccti.base.web.usergroup.VedUserGroupPage(2), interface:
>
> IBehaviorListener.onRequest],response=PageRequest[com.ccti.base.web.usergroup.VedUserGroupPage(2)],sessionid=kdrq2wp0eaea,sessionsize=212417,sessionstart=Mon
> Jan 12 23:24:44 CST
>
> 2009,requests=14,totaltime=7024,activerequests=1,maxmem=66M,total=18M,used=14M
>
> sessionsize=212417
>
> I'm particularly interested of the sessionsize that is being logged by the
> requestlogger.
> I was wondering what size does it contain?
>
> In One of My List Pages,
> It increases by about 72000 when I put in an ajaxpagingnavigator.
>
> I was just wondering what things are put into the session?
> One thing that bothers me though is I tried using a LoadableDetachableModel
> in my dataview.
> It returned the almost the same sessionsize for a dataview in which I used
> a
> CompoundDetachableMOdel.
>
> I tried viewing the page map using eclipse's debug, I can see that it keeps
> a reference to the previous page, so does it mean that Wicket always keeps
> the last page available in the HttpSession? or does it keep the current
> page
> in the session.
>
> If I try exploring the session, I can see that the loadableDetachableModels
> model object cannot be viewed.
> And the compoundpropertymodel's attributes can be viewed.
>
> As far as I can see, the wicket components consume significant session size
> . Please correct me if I am wrong.
>
>
> So question is,
>
> What Page Does Wicket Store in its HttpSession? How Many Pages Does It
> Store?
> What consumes significant session size in wicket?
> Is it bad to keep a reference to the previous pages in instance variables
> of
> new pages?
>
> Carlo
> --
> View this message in context:
> http://www.nabble.com/sessionsize-of-requestlogger-tp21416562p21416562.html
> Sent from the Wicket - User mailing list archive at Nabble.com.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
> For additional commands, e-mail: users-help@wicket.apache.org
>
>
--
Jeremy Thomerson
http://www.wickettraining.com
Re: sessionsize of requestlogger
Posted by carloc <ca...@yahoo.com>.
Another observation that I noted is that components contain a reference to
their raw markup.
The markup which I used to code it. It is actually stored in the
httpsession.
Is This Correct?
For example, I see this in my debugger.
<html>
<body>
<wicket:panel>
<script language = "Javascript" type = "text/javascript">
var $sxi = jQuery.noConflict();
$sxi(function() {
jQuery('#list1a').accordion({autoheight: false});
})
</script>
<div class="basic" style="float:left;" id="list1a">
<div>
<ul>
<li wicket:id = "firstLevelMenu">
</li>
</ul>
</div>
</div>
</wicket:panel>
</body>
</html>
carloc wrote:
>
> Hi Everyone,
>
> RequestLogger -
> time=103,event=Interface[target:AjaxPagingNavigationLink(userListForm:userGroupContainer:userGroupCheckGroup:ajaxPagingNavigation:navigation:9:pageLink),
> page: com.ccti.base.web.usergroup.VedUserGroupPage(2), interface:
> IBehaviorListener.onRequest],response=PageRequest[com.ccti.base.web.usergroup.VedUserGroupPage(2)],sessionid=kdrq2wp0eaea,sessionsize=212417,sessionstart=Mon
> Jan 12 23:24:44 CST
> 2009,requests=14,totaltime=7024,activerequests=1,maxmem=66M,total=18M,used=14M
>
> sessionsize=212417
>
> I'm particularly interested of the sessionsize that is being logged by the
> requestlogger.
> I was wondering what size does it contain?
>
> In One of My List Pages,
> It increases by about 72000 when I put in an ajaxpagingnavigator.
>
> I was just wondering what things are put into the session?
> One thing that bothers me though is I tried using a
> LoadableDetachableModel in my dataview.
> It returned the almost the same sessionsize for a dataview in which I used
> a CompoundDetachableMOdel.
>
> I tried viewing the page map using eclipse's debug, I can see that it
> keeps a reference to the previous page, so does it mean that Wicket always
> keeps the last page available in the HttpSession? or does it keep the
> current page in the session.
>
> If I try exploring the session, I can see that the
> loadableDetachableModels model object cannot be viewed.
> And the compoundpropertymodel's attributes can be viewed.
>
> As far as I can see, the wicket components consume significant session
> size . Please correct me if I am wrong.
>
>
> So question is,
>
> What Page Does Wicket Store in its HttpSession? How Many Pages Does It
> Store?
> What consumes significant session size in wicket?
> Is it bad to keep a reference to the previous pages in instance variables
> of new pages?
>
> Carlo
>
--
View this message in context: http://www.nabble.com/sessionsize-of-requestlogger-tp21416562p21420242.html
Sent from the Wicket - User mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org
Re: sessionsize of requestlogger
Posted by Bernard <bh...@actrix.gen.nz>.
Igor,
Thanks for referring to Page#getPageReference.
Regarding the reload, I created my own error loading a page.
>On Fri, Apr 17, 2009 at 5:49 PM, Bernard <bh...@actrix.gen.nz> wrote:
>> I prefer not to instantiate a page with a non-default constructor
>> because it breaks if the user reloads it. How do we pass parameters
>> between pages except via query strings which I try to avoid for
>> various reasons?
>
>can you please explain this a bit more? it is perfectly normal to do
>setresponsepage(new edituserpage(usermodel));
>
>wicket will redirect to a GET url which will always properly load that
>instance of the page, not create a new one every time.
>
>you can pass page references around using Page#getPageReference,
>formerly Page#getPageId, and constructing a url to it using
>requestcycle.urlfor() method.
>
>-igor
>
[snip]
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org
Re: sessionsize of requestlogger
Posted by Igor Vaynberg <ig...@gmail.com>.
On Fri, Apr 17, 2009 at 5:49 PM, Bernard <bh...@actrix.gen.nz> wrote:
> I prefer not to instantiate a page with a non-default constructor
> because it breaks if the user reloads it. How do we pass parameters
> between pages except via query strings which I try to avoid for
> various reasons?
can you please explain this a bit more? it is perfectly normal to do
setresponsepage(new edituserpage(usermodel));
wicket will redirect to a GET url which will always properly load that
instance of the page, not create a new one every time.
you can pass page references around using Page#getPageReference,
formerly Page#getPageId, and constructing a url to it using
requestcycle.urlfor() method.
-igor
>
> Or as Daniele asked: Is there "a way to recover the last Page in a non
> restful way"?
>
> Is there any jira issue that tracks this subject?
>
> Bernard
>
> On Wed, 15 Apr 2009 09:09:12 -0700, you wrote:
>
>>On Wed, Apr 15, 2009 at 2:22 AM, Daniele Dellafiore <il...@gmail.com> wrote:
>>> On Mon, Jan 12, 2009 at 5:43 PM, carloc <ca...@yahoo.com> wrote:
>>>
>>>> What Page Does Wicket Store in its HttpSession?
>>
>>the last page accessed from every pagemap is stored in httpsession.
>>
>>>> Is it bad to keep a reference to the previous pages in instance variables of
>>>> new pages?
>>
>>when using diskpagestore there are certain usecases that break. if you
>>keep a page purely for navigation purposes it is fine. but if you
>>manipulate instance of a page from inside a page that holds a
>>reference to it it might lead to unintended sideeffects like your
>>changes to that instance being lost. this has to do with how
>>diskpagestore keeps each page separate and how it rewrites references
>>during serialization.
>>
>>-igor
>>
>>>
>>> I am really interested in an answer about this two questions. I am
>>> using 1.3 so I cannot use Page Reference mechanism.
>>>
>>> Finally, I cannot find a way to recover the last Page in a non restful
>>> way and as I stated in another mail, history.back does not work in all
>>> conditions (i.e. submit that reload the page instead of calling a
>>> setResponse)
>>>
>>> --
>>> Daniele Dellafiore
>>> http://blog.ildella.net/
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
>>> For additional commands, e-mail: users-help@wicket.apache.org
>>>
>>>
>>
>>---------------------------------------------------------------------
>>To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
>>For additional commands, e-mail: users-help@wicket.apache.org
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
> For additional commands, e-mail: users-help@wicket.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org
Re: sessionsize of requestlogger
Posted by Bernard <bh...@actrix.gen.nz>.
It is tempting to use references of previous pages. But I understand
that "passing references between pages and page serialization don't
mix nicely"
http://mail-archives.apache.org/mod_mbox/wicket-dev/200903.mbox/%3C2e1ce7cd0903030143h4a525b8cu46915c799061a8d6@mail.gmail.com%3E
What would be the best Wicket way of letting the current page directly
modify the model of the previous page and then let the current page
show the previous page with the new model?
I prefer not to instantiate a page with a non-default constructor
because it breaks if the user reloads it. How do we pass parameters
between pages except via query strings which I try to avoid for
various reasons?
Or as Daniele asked: Is there "a way to recover the last Page in a non
restful way"?
Is there any jira issue that tracks this subject?
Bernard
On Wed, 15 Apr 2009 09:09:12 -0700, you wrote:
>On Wed, Apr 15, 2009 at 2:22 AM, Daniele Dellafiore <il...@gmail.com> wrote:
>> On Mon, Jan 12, 2009 at 5:43 PM, carloc <ca...@yahoo.com> wrote:
>>
>>> What Page Does Wicket Store in its HttpSession?
>
>the last page accessed from every pagemap is stored in httpsession.
>
>>> Is it bad to keep a reference to the previous pages in instance variables of
>>> new pages?
>
>when using diskpagestore there are certain usecases that break. if you
>keep a page purely for navigation purposes it is fine. but if you
>manipulate instance of a page from inside a page that holds a
>reference to it it might lead to unintended sideeffects like your
>changes to that instance being lost. this has to do with how
>diskpagestore keeps each page separate and how it rewrites references
>during serialization.
>
>-igor
>
>>
>> I am really interested in an answer about this two questions. I am
>> using 1.3 so I cannot use Page Reference mechanism.
>>
>> Finally, I cannot find a way to recover the last Page in a non restful
>> way and as I stated in another mail, history.back does not work in all
>> conditions (i.e. submit that reload the page instead of calling a
>> setResponse)
>>
>> --
>> Daniele Dellafiore
>> http://blog.ildella.net/
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
>> For additional commands, e-mail: users-help@wicket.apache.org
>>
>>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
>For additional commands, e-mail: users-help@wicket.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org
Re: sessionsize of requestlogger
Posted by Igor Vaynberg <ig...@gmail.com>.
On Wed, Apr 15, 2009 at 2:22 AM, Daniele Dellafiore <il...@gmail.com> wrote:
> On Mon, Jan 12, 2009 at 5:43 PM, carloc <ca...@yahoo.com> wrote:
>
>> What Page Does Wicket Store in its HttpSession?
the last page accessed from every pagemap is stored in httpsession.
>> Is it bad to keep a reference to the previous pages in instance variables of
>> new pages?
when using diskpagestore there are certain usecases that break. if you
keep a page purely for navigation purposes it is fine. but if you
manipulate instance of a page from inside a page that holds a
reference to it it might lead to unintended sideeffects like your
changes to that instance being lost. this has to do with how
diskpagestore keeps each page separate and how it rewrites references
during serialization.
-igor
>
> I am really interested in an answer about this two questions. I am
> using 1.3 so I cannot use Page Reference mechanism.
>
> Finally, I cannot find a way to recover the last Page in a non restful
> way and as I stated in another mail, history.back does not work in all
> conditions (i.e. submit that reload the page instead of calling a
> setResponse)
>
> --
> Daniele Dellafiore
> http://blog.ildella.net/
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
> For additional commands, e-mail: users-help@wicket.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org
Re: sessionsize of requestlogger
Posted by Daniele Dellafiore <il...@gmail.com>.
On Mon, Jan 12, 2009 at 5:43 PM, carloc <ca...@yahoo.com> wrote:
> What Page Does Wicket Store in its HttpSession?
> Is it bad to keep a reference to the previous pages in instance variables of
> new pages?
I am really interested in an answer about this two questions. I am
using 1.3 so I cannot use Page Reference mechanism.
Finally, I cannot find a way to recover the last Page in a non restful
way and as I stated in another mail, history.back does not work in all
conditions (i.e. submit that reload the page instead of calling a
setResponse)
--
Daniele Dellafiore
http://blog.ildella.net/
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org
Re: sessionsize of requestlogger
Posted by carloc <ca...@yahoo.com>.
As an added observation, sessionsize of deployment mode is much lower than in
development mode.
Is there any reason to this?
carloc wrote:
>
> Hi Everyone,
>
> RequestLogger -
> time=103,event=Interface[target:AjaxPagingNavigationLink(userListForm:userGroupContainer:userGroupCheckGroup:ajaxPagingNavigation:navigation:9:pageLink),
> page: com.ccti.base.web.usergroup.VedUserGroupPage(2), interface:
> IBehaviorListener.onRequest],response=PageRequest[com.ccti.base.web.usergroup.VedUserGroupPage(2)],sessionid=kdrq2wp0eaea,sessionsize=212417,sessionstart=Mon
> Jan 12 23:24:44 CST
> 2009,requests=14,totaltime=7024,activerequests=1,maxmem=66M,total=18M,used=14M
>
> sessionsize=212417
>
> I'm particularly interested of the sessionsize that is being logged by the
> requestlogger.
> I was wondering what size does it contain?
>
> In One of My List Pages,
> It increases by about 72000 when I put in an ajaxpagingnavigator.
>
> I was just wondering what things are put into the session?
> One thing that bothers me though is I tried using a
> LoadableDetachableModel in my dataview.
> It returned the almost the same sessionsize for a dataview in which I used
> a CompoundDetachableMOdel.
>
> I tried viewing the page map using eclipse's debug, I can see that it
> keeps a reference to the previous page, so does it mean that Wicket always
> keeps the last page available in the HttpSession? or does it keep the
> current page in the session.
>
> If I try exploring the session, I can see that the
> loadableDetachableModels model object cannot be viewed.
> And the compoundpropertymodel's attributes can be viewed.
>
> As far as I can see, the wicket components consume significant session
> size . Please correct me if I am wrong.
>
>
> So question is,
>
> What Page Does Wicket Store in its HttpSession? How Many Pages Does It
> Store?
> What consumes significant session size in wicket?
> Is it bad to keep a reference to the previous pages in instance variables
> of new pages?
>
> Carlo
>
--
View this message in context: http://www.nabble.com/sessionsize-of-requestlogger-tp21416562p21418393.html
Sent from the Wicket - User mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org