For as long as I have used Firefox on openSUSE it has always had the occasional crash (at least a daily occurrence).
But it seems with the combination of a clean install of openSUSE (13.2 KDE x86_64) and Firefox 33.0-2.1 (new profile) I am seeing an increase in these crashes.
In a typical day (approx 2 - 3 hrs of usage) I am seeing about 6 - 10 random crashes.
I’ve tried a new profile with no extensions/addons which doesn’t seem to have made any difference to the occurrence.
There also doesn’t appear to be a consistent trigger to the crashes (I can be on a page reading it for about five minutes before a random crash).
I have always just kind of accepted Firefox crashes but the frequency now is a little bit annoying.
What are other people experiencing with the 13.2 & FF 33 combination?
I’ve tried comparing it with Chromium and Midori (same usage time saw one crash in chromium and none in midori)
No extensions/addons also mean no flash? That and video drivers is what I would suspect now you have tested a clean profile. Is sensitive stuff, why Mozilla have a wiki about blocked drivers https://wiki.mozilla.org/Blocklisting/Blocked_Graphics_Drivers Not blocked is not the same as brilliant so may be yours is weird.
Hardware acceleration on vs. off setting under “Advanced” also tested?
If there are known issues with video card vs. Firefox it should be possible to search for them. There will be other victims.
May be you should state card is the one you list in signature.
If you want some speculation here it comes. Your ram might be either bad or set up wrong. Is sure way of getting crashes. Been through that in openSUSE and Firefox recently - Firefox did seem to be most crashing prone part. Probably just logic, it use much memory hence greater chance to run in to a bad chunk of ram. But not necessarily. “Code” does not crash equally 32bit/64bit/Windows/Linux vs. random ram issue will not give you same outcome. Some conditions more forgiving than others. Useless info but not if you have bad ram. So test that, wont hurt anything and then you can exclude. If test fails you can try reseating ram modules, check settings are ok regarding timings, voltage if there even are any in bios.
More unlikely idea: you have not noticed that among the pages you visit an evil one is a browser killer. They do exist though rare. Some can also send Firefox in to a spin where it will just accumulate memory usage. I used to have list of those long time ago. Typically involving handling of images, media - and mostly a solved problem. Used to be a lot worse is my impression. Also stupid to rule out weird event with Firefox
yes, driver is the one in my sig - GMA X4500 (which should be fine according to the list you linked to).
Flash is installed and not disabled but crash can occur on non flash sites.
I’ll try disabling hardware acceleration in FF settings and see how that goes.
Seeing the same behaviour across two identical systems so hard to believe it would be a ram problem (but I guess nothing is impossible)
There are some mile long “Firefox diagnotic” pages available at mozillaZine http://kb.mozillazine.org/Knowledge_Base but somehow I do not think they help. Most of issues are on Windows and/or running corrupted profile.
If you were on Windows so called “3rd party” of which there can be loads would be interesting. “Security” software for example. All sorts of stuff can be an issue. May be even preinstalled junk. Not sure relevant on Linux, on new install.
non flash sites = or so you think Flash can be run silently or may be plugin mess up just being active because it hate your video drivers. Is guessing 100% of course. Hopefully someone with same card can confirm it is fully working on 13.2 KDE.
Better than you reading changelogs for minor kernel updates in hope of a fix I did that yesterday because bored, is amazing… Cant rule it out as they do make all sorts of wacky fixes for single items, motherboards etc. https://www.kernel.org/pub/linux/kernel/v3.x/ If you get annoyed you might sprint ahead but probably best to get some more qualified help than me posting link. If there is logic and hope left you should not have to mess with kernel updates to avoid Firefox crashes.
I have had at most one firefox crash since installing opensuse 13.2 (on Nov. 1).
I have typically two or three “rekonq” crashes per day. When “rekonq” crashes twice on the same site, I open that on “firefox” and all is well. I use “rekonq” for sites that I access via “akregator” (RSS reader), and “firefox” for almost everything else.
With “rekonq”, most of my crashes are at sites that are heavy with advertising. I don’t see that with “firefox”, but that’s probably because I am using “noscript” which block a lot of those ads. The advertising sites rotate their ads, which is probably why the same page will sometimes crash “rekonq” and sometimes not.
From your description, it sounds to me as if it may be ads causing your “firefox” crashes, perhaps due to bad javascripting at the ad sites. You might want to give “noscript” a try to see if it helps.
I allow scripts from most of the sites that I regularly visit (the “noscript” whitelist). But that still blocks the scripting that comes from advertising sites, and is probably part of why “firefox” rarely crashes. I temporarily allow scripting for sites that I occasionally visit. Admittedly, a few sites are hard to read with “noscript”, because they start the content that I want via advertisements at sites that are blocked. So I have to “allow scripts globally” to see those sites. And that’s what I was doing when “firefox” most recently crashed (probably the only crash in the last year).
I also use “flashblock” with “firefox”, though I’m not sure how significant that is. “Firefox” uses flash via a plugin-container, and a bad flash applet should only crash the container and leave “firefox” running.
In summary, I suggest that you give “noscript” a try.e
That is horrible ashley194! And you seem happy too, geeez.
No really, if Firefox crashes on ANY website and it is reproducible then run to their Bugzilla and report.
Should not resort to noscript or other trick, like hmm YesScript https://addons.mozilla.org/en-US/firefox/addon/yesscript/ to make normal usage work. Flash I am not sure about, was not when I used Windows. Using old version on Linux does not change that.
Have to say I do not trust Mozilla for 1 second with most things so for example I also disable webrtc
Should not do a thing if not active but I see it as limiting attack surface, from Mozilla bugs
I also disable their health/telemetry reporting which I can see some would LOVE to use for tracking They, or the devs. getting the GO GO, already talked of this long time ago when ads got “introduced”. Now ads are official game is on… at best it is matter of time. Ads needs to evolve, people must get trained first.
It installs as separate Firefox with own profile so it does not mess up existing installs. No need to set up profile so more than 1 Firefox can run and all that.
If zero issues with that I guess you have found unspecified Firefox bug when used on KDE/openSUSE 13.2.
I do not know what you are doing differently – system or otherwise – that is giving you such a painfull problem. Wish I could help, but it is not happening here.
I had it crash on me pretty reliable with the old MB and only 2 gig of memory when I also ran a VM. FF really really like lot of real memory and does not seem to play well forced to swap. Now with new processor ( AMD 6 core) and MB plus 8 gig of memory it is very very happy .
Plenty of memory (free) here.
System has 8G of ram.
Not sure what to make of these firefox crashes. They were occurring over two systems that had a clean install of 13.2
Disabled hardware acceleration and no crashes for over 24 hours. Re-enabled hardware acceleration and still no crashes in the couple of days have been using (on both systems).
Anyway - seems all good now.
I did not have any with the old MB with 2gig unless I also was running Virtual box set to use about 512 meg. Then is crashed if you only looked at. LOL When the VM was not running no problems at all.
32bit/64bit/Windows/Linux vs. random ram issue will not give you same outcome. Some conditions more forgiving than others. Useless info but not if you have bad ram. So test that, wont hurt anything and then you can exclude. If test fails you can try reseating ram modules, check settings are ok regarding timings, voltage if there even are any in bios.