Oh, eventually windows reports that Chrome has crashed, though dismissing the error leaves Chrome open (strange). "Developer Tools"/ Scrips shows an error like this every second (the same period as the RPC timer set up in onModuleLoad() ):
Uncaught Error: Error calling method on NPObject! On Dec 7, 3:05 pm, mrpantsuit <[email protected]> wrote: > Sorry, I thought the XMLHttpRequests weren't leaking, but apparently > they still might be. > > Here's what I did: > Google Chrome 4.0.249.22 (Development version) > GWT 2.0.0 RC2 > Created GWT app: webAppCreator -out MemTest memtest.MemTest > Added this to onModuleLoad(): > > Timer timer = new Timer() { > @Override > public void run() { > greetingService.greetServer("xxx", new > AsyncCallback<String>() { > public void onFailure(Throwable caught) { > } > > public void onSuccess(String result) { > } > }); > } > }; > timer.scheduleRepeating(1 * 1000); > > Ran the generated MemTest launcher (OOPHM) and clicked "Launch Default > Browser" (Chrome). > > After a couple hours, the mem was 50MB+ (according to "Task > manager"/"Stats for nerds") and Page/Developer/"Developer Tools"/ > Profiles/"Heap Snapshot" indicated 2500+ instances ofXMLHttpRequest. > > Though, the snapshot indicated that only 5MB or so of memory was being > used, so I don't know what is using the other 45MB+. > > I also opened a OOPHM session in IE8, and the same steady memory > increase was observed (though the growth was actually slower). > > On Dec 4, 3:32 pm, mrpantsuit <[email protected]> wrote: > > > > > Regarding theXMLHttpRequestleak, it seems I just need to wait for > > the garbage collector to run. Does anyone know how to manually > > initiate garbage collection in Chrome (or in the other browsers, for > > that matter)? > > > On Dec 4, 2:41 pm, mrpantsuit <[email protected]> wrote: > > > > I downloaded the Developer Channel version of Chrome, which has some > > > nice heap analysis developer tools. I thought that perhaps my memory > > > leaks were not restricted to IE, but were simply more pronounced on > > > it. Turns out I had several memory leaks. > > > > I changed to use the GWT incubator progress bar instead of the GWT Ext > > > one, which seemed to reduce the IE leakage significantly. I don't > > > know theleakis inherent to the GWT Ext progress bar, or perhaps I > > > was just using it wrong. > > > > I also noticed that I was leaking XMLHttpRequests. I rebuilt with GWT > > > 2.0 RC2, but theleakpersisted. I'm still investigating that one. > > > > I was also leaking HTTPCollections. I have a FlexTable that I rebuild > > > every few seconds after RPCing data from the server. I'm still > > > investigating that one, too. > > > > Thanks, > > > Kevin > > > > On Dec 3, 11:59 am, Chris Ramsdale <[email protected]> wrote: > > > > > A diff on com.google.gwt.xhr.client.XMLHttpRequestshould give you an > > > > idea > > > > of what code has changed and when it changed. That said, Rob points out > > > > earlier in this thread that he is not seeing the issue in 1.7. > > > > > Are you in a position where you could build your app with RC2 to see if > > > > the > > > >leakgoes away? > > > > > On Thu, Dec 3, 2009 at 11:35 AM, mrpantsuit > > > > <[email protected]>wrote: > > > > > > I'm seeing a similar memoryleakin 1.7. Does anyone know if this bug > > > > > predated 2.0, or was it introduced therein? > > > > > > On Nov 14, 11:28 am, Rob <[email protected]> wrote: > > > > > > Seehttp://code.google.com/p/google-web-toolkit/issues/detail?id=4133 > > > > > > andhttp://code.google.com/p/google-web-toolkit/issues/detail?id=4229 > > > > > > > Issue 4133 was raised almost a month ago but doesn't seem to have > > > > > > been > > > > > > acknowledged yet - would anyone from the GWT team care to comment ? > > > > > > > Cheers > > > > > > Rob > > > > > > > On Nov 13, 2:18 pm, Rob <[email protected]> wrote: > > > > > > > > I've been trying out one of my existing apps with GWT 2.0 to see > > > > > > > what > > > > > > > advantages it has and have noticed a majormemoryleakwhen using RPC > > > > > > > with Internet Explorer 8.0. > > > > > > > > Basically my application reads some values from a database and > > > > > > > displays these in the browser every 1 second. I am seeing amemory > > > > > > >leakof 16k per second -i.e.every time I make an RPC request the > > > > > > > browser leaks 16k ofmemory. This eventually leads to an out of > > > > > > > virtualmemorysystem error. If I switch back to using GWT 1.7 then > > > > > > > the problem goes away, there are no leaks at all. > > > > > > > > I have managed to reproduce the problem with a very simple RPC > > > > > > > example > > > > > > > which I can supply if required. > > > > > > > > Has anyone else noticed anything similar ? > > > > > > > > Cheers > > > > > > > Rob > > > > > > -- > > > > > > You received this message because you are subscribed to the Google > > > > > Groups > > > > > "Google Web Toolkit" group. > > > > > To post to this group, send email to > > > > > [email protected]. > > > > > To unsubscribe from this group, send email to > > > > > [email protected]<google-web-toolkit%2Bunsubs > > > > > [email protected]> > > > > > . > > > > > For more options, visit this group at > > > > >http://groups.google.com/group/google-web-toolkit?hl=en. -- You received this message because you are subscribed to the Google Groups "Google Web Toolkit" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/google-web-toolkit?hl=en.
