Bug: Console SmartGWT 2.4 Release

[jvmvik]

I thinks there is a big issue in SmartGWT 2.4.

1. SmartGWT 2.4 latest released with SmartClient 8.0, and Firefox 3.0, 3.5, 3.6

2. pure client side issue (SmartGWT console)

3. if you select one RPC, then another one. The browser becomes really slow, then crash or ask if I would to stop running script.

4. I can't provide any trace stack at this level. But, it's very easy to reproduced. Go to to SmartGWT showcase (http://www.smartclient.com/smartgwt/showcase/#main). Open the developer console, and select RPC tab. At same time, you have to open examples that create server side call. Now, you just have to select one RPC then another one. Then, the browser crashes.

5. showConsole() then we need to have at least to calls.

Thanks

[Isomorphic]

Thanks for the report, this has been fixed and the fix will be in the next nightly build.

[michalg]

1. SmartGWT 2.4 latest svn 1592; Firefox 3.6.9

Both in hosted and compiled mode, when using Developer Console rpc tab, when clicked on request row, I have got "script takes to long" error:
Skrypt: http://localhost:8080/SmartGWTplan9000/sc/system/development/ISC_Core.js?isc_version=trunk.js:3847

Logging level in Dev Console is set to Fatal.

Is it the same error, but not fixed yet ?
Thanks for any help to make console usable again.
MichalG

[Isomorphic]

Make sure to clear your browser cache the foolproof way: shutdown all instances of the browser except a blank page, clear the cache, restart the browser.

[michalg]

FF cache cleared, whole project clean&build with the same result - only number of reported line changes in trunk.js.
Also checked with:
<inherits name="com.smartgwt.SmartGwt"/>
<inherits name="com.smartgwt.debug.SmartGwtDebug"/>
<inherits name="com.smartgwt.SmartGwtNoScript"/>

Another thing: FF uses ~50MB memory when application is running. This does not change when starting Developer Console. But going to rpc tab and clicking on one of the request makes FF eating memory up to 1.7GB (4GB whole pc ram).
MichalG

[Isomorphic]

By nightly builds, we always mean these: smartclient.com/builds.

[mazzi]

Hi,

I just want to let everyone know that we have been having the same problem too with every 2.4 releases that we tried so far in both IE8 and FF. I have also tried the very latest ones from the smartclient.com/builds directory with the same result. This is a show stopper issue for us as it is preventing us from moving to 2.4, so it would be really nice if it gets resolved soon.

Thanks.
Mike.

[jvmvik]

I definitely agree that a big issue.

[mazzi]

Hi,

I just wanted to follow up with another post to say that I tried again with the latest 2.5 nightly build from today, and it still happens. I have attached a screen shot of the pop up message that comes up is firefox. I would really appreciate it if someone can confirm to me whether this issue has been fixed in the LGPL nightly builds which are the ones that I am testing with. And if so I would also appreciate it if someone could tell me what am I doing wrong, and what to do to fix it.

Thanks very much,
Mike

[mazzi]

Hi,

Here are some additional information about this bug that should hopefully narrow down the test case to reproduce and address this bug. I was able to reproduce it using the actual Smart GWT Showcase application, but only for specific RPC cases. It seems this console hang/freeze only happens in the RestDataSource, the XSD DataSource, and the XPAth data binding examples, but not in the JSON examples. There is something about these DataSources that cause this console script:

Script: http://www.smartclient.com/smartgwt/showcase/showcase/sc-2010-12-31/system/development/ISC_Core.js?isc_version=8.0-2.4.js:778

to become unresponsive.

I hope this provide more input for resolving the issue.

Thanks,
Mike

[michalg]

My application uses RestDataSource, too.
Had to switch to the Firebug to see rpc request/response.
MichalG

[Isomorphic]

Hi All,
Thanks for the additional information about DataSource types - it gave us enough to figure out what was going on and fix the issue.

We've made a change in our mainline codebase which will resolve this issue. The fix will be present in the next nightly build (Jan 25)

Regards
Isomorphic Software

[michalg]

Confirmed that this issue is fixed in nightly build 2011-01-25 for RestDataSource.
Thanks,
MichalG