curmudegeon99, his problem isnt even a mac issue, so your bugs are likely unrelated.

I noticed that. Are you saying there is not a shared code base between the two versions of the app? If that is the case then the FF team is even more lame than I thought. You're saying that you keep two 100% unrelated lines of development in sync with no shared code base?

As I enter my 16tth year as a professional software developer and now architect, I find that baffling. There most certainly is a shared core and it makes sense to allow the devs now working this defect to know that both implementations share the same problem.

Yes, that's one reason why he asked a new thread be made. And if he creates a new thread, he will be given the option to add his Firefox troubleshooting info automatically. This will show us what addons he has installed and show a few more details about his Firefox setup.

Currently, we can't see any of that when someone makes a regular post in another thread. That's the main reason why we ask someone to make a new thread. Not to dilute the reporting of bugs.

So since jhvance is on Windows 7, his problem may be caused by an addon. We'd like to rule that out first instead doing all that troubleshooting here in this thread and making it even longer than it already is. Since everytime a reply is made in this thread, everyone who is subscribed to it receives a email notification. It gets really annoying when it's just about more troubleshooting and not steps to fix the problem. If he finds out his problem is not caused by addons or a bad profile, he can jump back here and post that.

Noah.

Question 1: Is there any shared code base between the Windows Firefox and the Mac one?

If your answer is YES, then I really believe the add on question is a red herring since in my case you clearly saw no add ons were involved in the problem.

Awaiting your answer to question 1.

Day 1 with no answer to the question of do OS X and Windows share a FF code base.

Hi all,

Is there any info we can give about our setup to help you fix this issue ?

I guess there is no proper troubleshooting info because the browser doesn't crash, it just eats all the available memory in the box.

I really don't think the issue is there on Windows though, being a heavy Windows user (and doing support on XP/W7 for a living), to me it is only linked to Mac setups.

But it's clearly a big pain to have to rule out Firefox as a web browser on OSX ... I'm forced to use Google Chrome, and as a web developer I really don't like it ^^

Please tell us is we can upload some meaningful data for you guys

And please, at least acknowledge there IS something wrong with FF+OSX

Please keep up the good work, I'll always promote FF as the best anyway :)

I did begin to ask a "new" question before my earlier post, but the system kept providing opportunities to ascertain if a similar or comparable question had already been asked -- tiring of that game, this thread was the one most alike the problem which I've identified involving memory-leaking or -hogging characteristics.

If you want to know what plugins and addons I've got, then describe a quick-and-easy way to produce a listing that I can cut-and-paste in a future response, or dump to a file that can be attached.

Sodawillow and others interested, try this:

Visit about:memory as soon as you open Firefox and before you begin any browsing. Then, copy and paste everything on the page to http://pastebin.com and return to this site and give us the url to the pastebin.

Then grab another copy of the about:memory data during the halfway point of 10 minutes. Or when you notice you've reached the halfway point of the memory leak (if the highest you've seen is 2GB for example, then get the about:memory @ 1GB).

And lastly, get one more copy of the about:memory data before Firefox becomes unresponsive but is near the 2GB (or whatever) high memory you speak of. This part will be tricky if Firefox is hanging. :/

So you should have about 3 pastebins worth of about:memory data when you're done and I will show that to the memory dev and report back on his findings.

It'd be nice if you could just paste that data here but the first part of it becomes mangled & unreadable due to how the "Main Process" data section is formatted. See my example attempt here: https://support.mozilla.org/en-US/questions/963890#answer-456894

@ curmudgeon99

Question 1: Is there any shared code base between the Windows Firefox and the Mac one?

I asked and the answer I received was "Yes, most of the code is the same".

I then asked: Hmm, well if a memory leak is found on one platform, does that mean that the leak isn't really confined to that one platform but possibly all of them?

Answer I got: Depends on whether the leak is in the shared code.

Which is a good point. The memory leak is still being investigated so no one can say for certain yet if this leak is in the shared code. I've seen countless reports in general about high memory usage on Windows (lots of addons, lots of tabs, etc) and only a few specific reports of high memory usage on mac (which made me take notice and spot a trend). There's lots of reasons why you get high memory but not many solutions to solve them.

I'm just a volunteer and I've tested the best I can on Windows. I've found the best you can do in general is not to install too many addons or visit too many memory hungry sites (Tumblr, Facebook, Huffingtonpost, etc). My browsing session consists of a small amount of tabs (under ~10) and frequent closing and reusing of tabs. I've find that makes a huge difference for my surfing and I usually don't go over 250 k of memory. If I hit a bad site that has tons of images and lots of scripts (javascript), my memory will shot over 300 k and hit high CPU. The site is just as much to blame as Firefox, IMO. Some sites are just horrible. Take Huffingtonpost or Facebook for example.

You might want to get a newer version of Firefox 22.

Well, if "most" of the code is the same, then separating the Windows bugs from the OS X bugs seems to be a distinction without a difference, and it again raises the issue of trying to mask the severity of the issue by diffusing bug reports.

If you had read the prior traffic in this bug report, you will see that I DID upgrade to FF23.0 and the memory leak was WORSE.

The best way to test if it is your personal data is to simply create a new profile.

This can either be done by creating a different user in your OS, or turning by using the profiles options in Firefox.

I find this a useful tool with almost all programs, not just Firefox.

Ruediix, if you have read at the beginning, your suggestion is moot, since at the time it was a brand new install. The same goes with the Mac and PC versions. New users, new accounts, new installs. Memory leak just as bad as before as it is today.

DISCLAIMER This is My personal opinion not an official suggestion from Mozilla.

IF you think there is a genuine memory leak in Firefox on Mac it will be worth filing a bug, but please be aware if trying that I think you will need to

• Have the time to be able to demonstrate this to developers
• Produce about:memory reports from Firefox, rather than the alternative methods of indicating memory use
• In all probability be asked to run Firefox in safe mode in a clean profile
  • (or with just any suspect add-on)

If anyone thinks there is a genuine memory leak then let's try to get developer attention. It does appear a regression in Fx4 was missed. Anything now is likely to only affect a minority of users, but if it affects you then speak up and try to help instead of just saying you are being ignored.

Note also Noah_SUMO already has developer attention on this and may or may not be looking for further information.

What may be interesting is to know who with memory problems is willing to co-operate with developers if required ?

I expect most readers or posters in this thread either will not have the time to help or consider we should fix any problem without your help. Unfortunately those with a problem will likely be a minority group and if you want a fix you need to help instead of just complaining.

Anyone interested in following up on this?

John, I filed that bug two weeks ago and supplied memory dumps of before and after. Please don't just chime in unless you've read all the previous text.

If you had read the previous traffic on this post, you would have known:

• The problem came without any add ons.
• I am a 16-year software engineer and know how to file bugs
• I supplied official memory dumps from FIrefox that showed the before and after memory usage and had the evidence of the problem I saw verified by the Devs who allegedly are working on the bug.

curmuudgeon99

Sorry, I cannot follow all posts on this forum, although I have tried to follow this thread I had not noticed you or anyone else on this thread announcing a bug had been filed and was active.

curmudgeon99, I do not immediately see your bug what is the bug number please ?

bug 896016

OK thanks for the link to Noah's bug. (Going off at a tangent it may be a good idea to be able to post gzip files to the forum)

P.S. thast's a no go Kadir vetoed it.

As I said earlier anyone interested in helping ?

(But this depends on the possibility of the developers needing help to reproduce specific issues; as opposed to an opportunity for discussing related problems)

Disabling Ghostery worked for me. I re-installed it from Mozilla add-ons (different version, I was at 5.0.1 now I'm 2.9.6) So I downgraded.

All works fine now, at least my initial memory leak is gone. Firefox was permanently using 25% cpu and between 600-800mb ram in less than 5 mins, eventually after time I would be using like 3gb.

After disabling ghostery and starting firefox again, it's 0-3% cpu and only 300-400mb ram. It' holding steady and not increasing fast like before.

I may still have a long term memory leak (doesn't currently seem that way...). But unless it takes less than 24 hours to get to 1-2gb or more, it doesn't worry me as I would have closed firefox before then.

So disabling ghostery and closing/opening firefox is worth a try.

I'm assuming the OP has probably already done this, But google brought me here when looking up the problem. So this may help anyone else looking for an answer.