eCSStender.org crashes Firefox
AyKay
I would really like to try this thing out, but I'm seeing the main.js
script time out on every page at the eCSStender site. Makes me wonder
if it's really safe to use this thing in production or not?
Aaron Gustafson
I occasionally see a hang on Google Analytics, but, apart form that, I'm not experiencing the issue you're seeing. Can you provide me with a few answers so I can ferret out the issue?
1) What Firefox version are you using?
2) Which OS are you using?
3) Are you seeing any failed tests when you visit http://test.ecsstender.org
Thanks in advance.
AyKay
Sorry for not posting my setup. I'm running Firefox 3.6.6 on Windows
XP. I am not getting any failed tests on your test page.
A dialog box pops up, giving me the option to stop the script,
indicating that something is wrong in main.js (main.js:20 it says).
On the spoon demo page (http://ecsstender.org/demos/spoon/), the error
occurs in eCSStender.js:9
Best Regards,
Anders
On Jun 30, 2:03 pm, Aaron Gustafson <aaron.easydesi...@gmail.com>
wrote:
Aaron Gustafson
On 30 Jun 2010, at 10:19 AM, AyKay wrote:
> Sorry for not posting my setup. I'm running Firefox 3.6.6 on Windows
> XP. I am not getting any failed tests on your test page.
> A dialog box pops up, giving me the option to stop the script,
> indicating that something is wrong in main.js (main.js:20 it says).
> On the spoon demo page (http://ecsstender.org/demos/spoon/), the error
> occurs in eCSStender.js:9
I just ran both under the same build of FF on WinXP and am not seeing the issue. Perhaps its a plugin conflict? Try disabling things like Firebug, etc. and refreshing (note: I was able to run Firebug without issue, but I've seen it and other plugins cause problems). Hopefully we can suss out the issue.
AyKay
delete the cookies and clear the cache, the site loads properly once
(quite slowly though, around 10 seconds). Seems to choke on google
analytics. When I reload the page, the script timeout is back.
On 30 Juni, 21:02, Aaron Gustafson <aaron.easydesi...@gmail.com>
wrote:
Aaron Gustafson
> I have disabled all my Firefox plugins, but the problem remains. If I
> delete the cookies and clear the cache, the site loads properly once
> (quite slowly though, around 10 seconds). Seems to choke on google
> analytics. When I reload the page, the script timeout is back.
Looking into it. Until I can replicate though, it may be difficult.
Try this: in the Firebug console, run window.localStorage.clear(); after the page has loaded the first time. Then see if the second reload still hangs. If it doesn't, it may be a caching issue.
AyKay
trying this made the browser not hang on the reload!
Best regards,
Anders
Aaron Gustafson
> trying this made the browser not hang on the reload!
Ok, so it's a caching issue. In this session, had you visited the demos? I'm wondering if perhaps they are set to cache and that cache is causing a conflict with the main site cache.
I can set the demos to disable cache, which may resolve this, but first try this for me: after using the technique I showed you for clearing the localStorage cache, reload a few of the normal pages on the site and see if you experience the same issue. If not, open a demo page (http://ecsstender.org/demos/spoon/ for instance) and refresh that before going back to the main site and refreshing there. If a crash occurs, there's our culprit.
I'd do this, but for some reason I'm not experiencing the issue.
red
Documentation and started reading from 2nd step onwards and main.js
kept hanging on all the time, it would not crash FF, but freeze it for
a good 30 seconds or so. Using FF 3.6.3 on Ubuntu 10.04 with a lot of
plugins that have yet to conflicted on any other page, mainly web
developer stuff. Upon trying again everything seems to work, but still
thought I'd dip in since there clearly is some sort of an issue. Good
luck to finding it :/
Radu Tanasescu
actually testing with http://ecsstender.org/ to eliminate the
possibility of me getting the code wrong.
I'm using Windows 7 64bit and viewing the http://ecsstender.org/
website in 3 different browsers yields 3 different results, I've
attached some screenshots to point this out:
Google Chrome Version 5.0.375.125
http://radutanasescu.ro/ecsstender-chrome5.0.375.125-win7-64bit.jpg
Firefox Version 3.6.8
http://radutanasescu.ro/ecsstender-firefox3.6.8-win7-64bit.jpg
Internet Explorer 8 Version 8.0.7600.16385
http://radutanasescu.ro/ecsstender-IE8.0.7600.16385-win7-64bit.jpg
As you can see, it only works on Google Chrome, it partially works on
IE8 (though the shadow is applied wrong and it's not just under the
header but also on the left side of the header) and Firefox utterly
crashes.
Hope this helps somehow, it would be grate for eCSStender to work
properly.
Aaron Gustafson
> I get en error message in firefox as well, but it's not just that, I'm
that there was an issue with the caching mechanism. I've rewritten a
bunch of it and had it do a bit more tidying up after itself, which
should reduce the memory footprint. It seems to have sped things up
across the board.
marie-france
others, I get a "busy or not responding" error for the script :
http://ecsstender.org/j/main.js:20. I delete all cache and history,
disable all plugins and still get it on the second load of a page on
the site. On the first load the page freeze for a bit but there is no
error message.
Aaron Gustafson
version. Is that better?
Alcuadrado
It works (but slow) when cache is clear too
Aaron Gustafson
> I still having the same problem under Firefox 3.6 - Linux x86 version.
>
> It works (but slow) when cache is clear too
issues. I don't have a Linux install and FF 3.6 on Mac and Windows
appear fine for me.
brian stegner
Tested the new 'push' in Ubuntu Linux this morning.
The eCSStender works fine in Opera, Firefox, Galeon, Epiphany. I'm
running Ubuntu in the Gnome desktop, so didn't try Konqueror, but
they have the khtml thing pretty well sorted out, as far as I know.
But I have to say, what use is something like "border-radius," for
example, in Linux, when you're stuck with some of the most awful
fonts ever committed? It's a sad, fragmented state of affairs over
there.
Just my 2 Cents. As for 'speed', page reloads, initial page draws,
etc., it was just as snappy as always, before and after installing
the cSStender in a decent web site test environment.
Off-topic; What's the story with Opera's seeming inability to honor a
second "class" selector?
Example: I use border-radius on specific images, AND, I have a
separate selector (.roundCorners) that I can add to other images with
the addition of the class selector, right?
This works in most browsers, and would be handier in, say, having the
ability to add a single instance (in the css file) of border-radius,
-moz-border-radius, etc., rather than having the same attribute
re-written under every specific image. In other words, instead of
using #content img, and #sidebar img, and #header img, just adding
"roundCorners" to the html, right. but Opera just ignores the second,
generalized approach, and only draws the radius borders on images
that have been specifically selected. Weird.
Best regards,
Brian
Alcuadrado
I've tested it many times, and the same is happening :(
A detail that might help you is that in FF it stays a big time with
the message "waiting for images.eccstender.org" and always crashes
(with the firefox pop-up to stop the script) when saying "transferring
data from analytics.google...". And it's not due to a slow connection,
as in chrome it doesn't even take a second to download and render the
full page after clearing the cashe.