• Type: Bug
• Component: javafx
• Sub-Component: web
• Affected Version: 8u20

• Priority: P3
• Status: Closed
• Resolution: Cannot Reproduce

• Submitted: 2014-11-10
• Updated: 2015-06-12
• Resolved: 2015-03-27

JDK 8
8u60Resolved

David, thanks for reporting. Filed it: RT-40369. 27-03-2015 Unfortunately I cannot reproduce the crash at will, but I've had it crashing 7-8 times while browsing around at http://www.aftonbladet.se (Sweden's larges web based newspaper) 26-03-2015 Followup: I'll close it after RT-40055 is in 8u60, which brings the WebKit update into the release. 26-03-2015 Hi David, Thank you for the check. Please, describe the scenario in details which you used to face the crash. Unfortunately, an hs dump generated by a release build version gives only rough notion of what's going on because it doesn't show native stack. This JIRA issue had been splitted into two: 1) broken rendering (the problem tracked here) 2) resource leak (tracked in RT-39585). As the 1st problem has gone, I'm closing this issue. The 2nd one remains open. It's of the same priority and target (8u60). 26-03-2015 I can confirm that the rendering issue (at least) is gone with jfx9b52. I've tested on Mac and Windows. However, I can get the JVM to crash while browsing web pages. Stability seems to improve if I assign more RAM to the JVM. Here's a crash report: http://jalbum.net/download/hs_err_pid17768.log Hoping for a swift resolution to this and that these fixes will be back ported into JDK 8. 26-03-2015 I no longer see the issues (the "outstanding locks" and the rendering artefact) with the latest jfx9-dev, whereas I can easily reproduce them with older jfx versions. Since jfx9b52 WebView has got an updated WebKit: RT-36726. Would be fine if anyone else tries it with jfx9b52+ to confirm the problem has gone. 13-03-2015 Hi David, the bug is in a short-term list of critical issues. The work on it will be resumed as soon as possible... 13-02-2015 Thanks for the update Anton! We're keeping our fingers crossed that you can resolve it within the short future 13-02-2015 Sorry to bother you again, but what's the time estimate on getting this bug fixed? It's holding us back from moving to JavaFX with jAlbum. 27-01-2015 Working on the issues this one depends on. 12-01-2015 Leonid, what is a progress so far? 08-01-2015 What is the current status on this bug. It says "In progress" but I'd appreciate knowing more details and if there's anything we can do to assist. This bug currently stops us from moving to JavaFX. We're currently forced to use the 3:rd party "jxBrowser" alternative which lacks support for the video tag and mp4 playback. 07-01-2015 this is introduced in 8u20 (at least) so isn't a regression in 8u40 05-12-2014 SQE is ok to defer from 8u40. 05-12-2014 Is it a regression? 05-12-2014 pending SQE' OK to defer 05-12-2014 @ingrid, thanks for the info. We were able to reproduce on Intel HD 4000 as well. 19-11-2014 Here is their graphic card information from the CAP member - Video Card: Intel(R) HD Graphics 4000 / integrated - (latest drivers) Processor: Intel Core i5-3570K @ 3.4 GHz Memory: 16 GB OS: Window 7 Home Premium (x64) (I haven't had any problems with my video card or the drivers in the past 2 years, since I own it.) The rendering bugs are exist on OSX too (any browser), though I don't know if the memory problems exist on that platform too. 18-11-2014 Observing the same behavior on Mac. 18-11-2014 Well, this is reproduced when a thumbnail with images (not with video) is selected. So, this doesn't relate to a pipeline. 18-11-2014 I tried it with 8u20 b26 as reported in the bug. I don't reproduce the problems which the attached video demonstrates. So, everything looks Ok to me. The same with the latest 8u40. I suspect this might be related to the GPU used. I have NVIDIA Quadro 410 on Windows 7 x64. 18-11-2014 We should try this with a build of 8u40 that contains the fix for RT-39013 since that fix has greatly reduced the memory footprint. 10-11-2014