How can I quickly "unclog" firefox when it runs out of memory (with 1000/2000 tabs)
How can I quickly "unclog" firefox when it runs out of memory (with 1000/2000 tabs)
Hi, Once in a while I try to clean up my tabs. First thing I do is use "merge all windows" to put all tabs into one window.
This often causes a memory clog and firefox get stuck in this state for 10-20 minutes
I have recorded one such instance.
I have tried using the "discard all tabs" addon, unfortunately, it is also getting frozen by the memory clog.
Sometimes I will just reboot my PC as that is faster.
Unfortunately, killing firefox this way, does not save the new tab order, so when I start firefox again, it will have 20+ windows open, which I again, merge all pages and then it clogs again !
So far the only solution I have found is just wait the 20 minutes.
Once the "memory clog" is passed, it runs just fine.
I would like better control over tab discard. and maybe some way of limitting bloat. For instance, I would rather keep a lower number of undiscarded youtube that as they seem to be insanely bloated.
In other cases, for most website I would like to never discard the contents.
In my ideal world, I would like the tabs to get frozen and saved to disk permanently, rather than assuming discard tabs can be reloaded. As if the websites were going to exist forever and discarding a tab is like cleaning a cache.
Maybe don’t have a THOUSAND tabs
I will, not, do that.
Then you will have software that doesn’t work. This is not a Firefox problem, or a problem of extensions, or anything but a user problem.
If your 1998 Toyota Camry is struggling to haul a cargo container up a hill it’s not the car’s fault. You’re doing it wrong. Whatever tasks you’re trying to do with 1000 tabs, a web browser is the wrong tool for the job.
I'm using the wrong tool in the wrong way and won't stop!!!
Help me!!!
ROFL...
Well then we can't help you
You're not likely going to get any real help since you're insisting on using the browser in an extreme and unconventional way. Your little world is just one browser/OS crash from losing all of those tabs.
I can save all the tabs easily
Here I show how to save 1775 tabs with one click
Ordering and sorting them, that is an impossible task That's why i keep them open
Here is an alternative Piped link(s):
https://piped.video/YaSl1Bb750Y
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source; check me out at GitHub.
What is amazing to me is how some people will come out of the woodwork to tell a person when they think they're using their browser "wrong". Just let them be if you have nothing to contribute.
If someone is trying to achieve a goal through (what they might not know are) impossible means, "letting them be" isn't going to help them.
Although it might not seem very helpful (and indeed there are better ways of helping) pointing out the flaws in the approach is contributing more than "letting them be". Doesn't cost a thing to be civil about it though.
You are objectively using it wrong. Its is like asking how to make your minivan break the sound barrier because you want to get to work faster.
They ARE contributing, in this case the correct answer is "don't do that"...
Feel free to use your browser how you want, but I will feel free to not help you troubleshoot your problem because it won't help you in the end.
I'm not going to tell you that you're managing your information wrong. I would physically die if I had ever more than 20 tabs (my ADHD couldn't handle it).
But I think you might be using the wrong tool. A browser (like Firefox) is not really designed as an information manager. It's primary purpose is navigating and visualizing web pages. So when you talk about "a few megabytes of text and images" thats not what your browser sees. Your browser handles more than just the text and images. It also handles fetching and prefetching, a browser history for every tab, a JS context and much much more.
What you want is some kind of personalized archiving system that processes websites into machine processable (ie searchable) structures. Firefox is not that. Maybe data hoarder communities will have the answers you seek.
I die at anything more than 5
Well so far, it would be too much friction and extra labour to export each tab to external software.
I'm not even sure what software other than a browser would display live web pages in a more organized manner than firefox ?
I'm pretty sure I just hit a bug that's causing firefox to wake up too many tabs and not handle tab discarding correctly. Firefox does seem like the best tool still even if it's not working right.
What I would like instead is a browser that treats tabs more like virtual machines that you can roll back, suspend to disk and resume. Little package of data that get frozen in time and are externally searchable.
Anyway, here's my setup
What I would like instead is a browser that treats tabs more like virtual machines that you can roll back, suspend to disk and resume. Little package of data that get frozen in time and are externally searchable.
Maybe look at ArchiveBox. IIRC it has pretty much everything you ask for including an import from your browser history and bookmarks.
The solution is to see a psychotherapist because dude is there something strange happening in your brain and it really needs fixing.
I think the machine built to handle hundreds of trillions of operation per second should be better at handling a few gigabytes of text and images.
Fuck, you're welcome to create your own web browser if it’s that easy.
Oh sorry, i forgot you have a lot of shit going on.
Yes, it's a disease called "having a lot of shit going on and not wanted to spend my afternoon sorting tabs" It is cured by "throwing all tabs in the bin and starting over" because today's computer are so incredibly weak they can't handle a few megabytes of text anymore.
Okay I know people are being rude. You have to understand its not just text. Your browser sends a request to a server for a webpage and it downloads that webpage, all media included. Its not just text. The only solution here is disabling all of your addons and going one by one until the merge all works. Or finding a work flow that doesn't involve the goal of reaching 20k tabs. Browser are not designed to search through tabs. Firefox has bookmark tags and keywords to search or instantly open a link. But tabs are not meant to be this repository of where you've been.
because today's computer are so incredibly weak they can't handle a few megabytes of text anymore.
I mean, sites today are more richer compared to earlier 2000s. We have css, more complex js scripts, embedded fonts, embedded videos etc. I'm sure you understand that it takes more than a few megabytes of RAM.
You have thousands of things going on?
If you need quick access to this many pages I suggest organizing bookmarks. As this is what they are meant for. Tabs are meant for active pages you are working with. So anytime you get that many tabs with any browser its gonna run like shit.
Nah, FF handles thousands of tabs just fine. I literally have just as many if not more tabs than OP and have never seen this issue. It’s either from the merge they’re doing or something else. It would be better if y’all just worked under the assumption that this does work and something is otherwise wrong with op’s setup.
The issue is parsing all that. There is no way you can keep that many tabs readily accessible like tabs are meant to be. Which is why these addons were born and are not official parts of Firefox. This is one of those just because you can doesn't mean you should situations. I get they've adopted this workflow, but reading through this it sounds more like daily driving than actual work. Which makes this even more bizarre, you can't read them all, they have to reload when you open them after a while (ie download again) so all points are moot. You aren't saving the page, you are holding onto a shell that will request the page again when you wake it up. If the server went offline never to be seen again your tab will not hold the information.
With this workflow, it might be better to have a crawler dump everything into folder hierarchies that are content searchable, and then search that like google using specialized software. I dont see any other reason you could even have 1k tabs open efficiently, you aren't searching through that, might as well google again and follow the purple links.
I find organizing bookmarks incredibly tedious. I have bookmarks folder with thousands of tabs in and it's just easier to use google again to re-find the information than to pick them out of bookmarks. Also tabs just keep the title and URL so you can't even search the text inside. So, organizing a library of tabs is like a much worse version of google without previews. I also use the session manager addon but again, when you open thousands of tabs, it clogs up the memory almost instantly. It's taking multiple gigabytes of ram, just to display a few kilobytes of text ! I wish the browser would just render the page as a static searchable text and image and then ditch all the javascript garbage.
That takes too long. Organizing tab is the computer's job !
Literally isn't, which is why you're hitting a wall.
Have you tried bookmarking things instead of leaving them open as tabs?
Yes, I find that it identical to closing a tab. I never go in the bookmarks manager after. It is very clunky to use, it adds extra steps compared to keeping the tab open. At that point, it's usually easier to use google to find it again, since at least google can search text inside the page, not just the title. I do occasionally dump my thousands of tabs into the bookmarks managers, in a single unusable folder. It hasn't yet happenned that one of these tabs was retreived. But I hope in the future that I could dump all these tabs into another piece of software that will fetch all the tab's body data and allow me to search it all with a local LLM based search like "using my bookmarks, create one browser window with all URLs on the topic of the 7 megahertz maser" We're close but not there yet.
You might be interested in this then, it's an app that uses AI to auto-tag saved bookmarks: https://hoarder.app/
geez, just press Ctrl+W when you're done with a tab, or if the tab is older than a couple hours
I don't understand why some are so attached to tabs. Search your history if you need it again.
I tried closing tabs, I have to finish reading them, make sure I got everything and that whatever reason I had for opening that tab was done. The result is that I spend all most all my time trying to close and sort and order tabs instead of doing what I was trying to do in the first place. And then the browser freezes for 10 minutes.
Something is very wrong that 64GB is nowhere near enough to handle a few megabytes of text. And searching text inside of all tabs is an unthinkably difficult operation ?
Where did the web go wrong !?
Rather than try and force Firefox to deal with thousands of tabs, it’d be easier to use an add-on like SingleFile to download the tabs as self-contained HTML files. After that, you can search their contents using free tools like Agent Ransack or DocFetcher.
If you prefer to keep the data in your browser, then how about using a service like Instapaper that lets you save pages for reading/referencing later as well as search their contents?
Why do you have so many tabs open?
A tab suspender extension might help some, but there's only so much you can do to minimize the impact of thousand(s) of tabs. Cleaning out old tabs more frequently is probably a better habit.
I tried a tab suspender, but it would replace tabs with a moz:// address that would end up breaking all my tabs when I copy & pasted them from a text file. Also tab suspender doesn't work once firefox gets into that state. I think the internal scheduler is trying to load tabs and discarding them as fast as possible. What I need is a big "stop button" that stop it all from at least trying to load new tabs.
I think what's happenning is when I merge all windows, many gets get woken up and, like the youtube tabs they seem to gobble up 2 to 4 gb of ram while initializing to that freezes everything.
It's making it really hard to get to 10k 20k tabs when it really falls apart like that with not even 2k tabs.
This makes the browser experience really bogged down where most of the time is spent finishing and closing tabs instead of just getting on with the actual task.
I would really like to spend less time fiddling with my browser and it "just working".
I've read this entire thread like three times and watched all the videos you've posted, and I still don't understand your workflow at all.
If searching bookmarks/history is harder than using Google to just find the thing you want to get back to, why do you need to keep the things you want to get back to open rather than just using Google to find the page again later? Or when you want to get back to something you (think you?) have left open, do you find it just by scrolling through all your tabs until a title/favicon looks like what you're looking for?
Your last paragraph makes it seem like maybe you want to keep the tabs open so if the page/content gets deleted off of the server, you don't lose it. Is that correct? I'd imagine that doesn't always accomplish that, though, right? (Particularly for something like YouTube.) If that's a significant part of why you keep the tabs open, though, maybe that bit at least is a good question for a data hoarder community.
I haven't been able to find any "discard all tabs" addon for Firefox by Googling. And I can't guess what exactly it does. (Does it save tab states to disk and suspend - but also leave open - all tabs or something?) Are you sure that's the name of the addon you're using?
why do you need to keep the things you want to get back to open rather than just using Google to find the page again later?
If it's already in memory, that's one few step to reach it.
My tab manager can't search google
do you find it just by scrolling through all your tabs until a title/favicon looks like what you’re looking for?
I search my live memory with Tab Manager Plus
Sadly, it can't search tab body text, only tab titles.
if the page/content gets deleted off of the server, you don’t lose it. Is that correct?
My software should not discard data without my permission. When it runs out of RAM it should dump to disk cache, not delete. But browser have the builtin assumption that the web remembers everything, which is false. I also think bookmarks should save all tab data, all text, all images, all code, all video, and the code should remain as functional as possible. That's a long way off, currently the only way to do that is freeze the tab with its browser and operating system inside a virtual machine live snapshot.
I haven’t been able to find any “discard all tabs
I believe this one can do, discard selected tabs, but not discard all tabs
https://addons.mozilla.org/en-GB/firefox/addon/discard/
Discard is the action where all tab content is deleted, keeping only URL, title and favicon
I haven't found discard all tabs either.
I would like "stop all tabs" to work
https://addons.mozilla.org/en-US/firefox/addon/stop-all-button/
But it will only work after firefox has cleared the clog, it currently freezes with the rest of the browser.
If it's already in memory, that's one few step to reach it.
I search my live memory with Tab Manager Plud
Oh, so you're doing something like Googling just to find the page title and then rather than clicking the link in Google, (closing the Google results page, I hope and) searching through your tab titles with Tab Manager Plus to find and switch to the open tab where you already have the page in question open?
Though, I still don't understand why you keep the tab open in the first place rather than juat closing the tab when you're (at least for the moment) done with it and then Googling to find the content again and clicking the appropriate link to get that same content in a new tab when you do need it again. I asked whether the reason was so that if the content is removed from the server, you didn't lose it, but I don't think anything you said in your last post answered that question. You did say:
My software should not discard data without my permission. When it runs out of RAM it should dump to disk cache, not delete.
Which wasn't quite a direct answer to my question. And you then directly admit that the browser doesn't even keep content that's open in a tab:
But browser have the builtin assumption that the web remembers everything, which is false.
So that must not be why you keep content open in tabs, right?
Is it maybe something like if you keep something open in a tab, the presence of that page title in your tab manager gives you confirmation when you later Google to find the page title that such-and-such particular result in the Google results is indeed the thing you're looking for and not a different page than the one you were looking for?
Just as an aside, my web browser use is probably atypical as well. I have my browser forget all cookies, history, cache, etc (basically everything but my bookmarks) every time I fully close it. And I close it every time I switch activities to keep my online personas isolated from each other. (So I'm never logged into my Google account and my Amazon account at the same time, for instance. To reduce targeted ads and such.)
Also, I'm wondering if something more like a caching proxy with maybe page searching capabilities and finegrained control of what is cached and what isn't might fill part of your use case, but I still don't have a firm grasp on your use case.
I’m sorry I don’t have an answer
I'm really curious about the workflow you have that needs that many tabs. How does the History and Bookmark functions fall short of what you need?
It's easier to use google than the bookmarks manager, which can't even find text inside the pages. I do often dump all those thousands of tabs into a bookmarks folder. And it has never happened that I went back into that enormous pile to fetch something that would take hours to find again. I have no use for the history either. A gigantic, alphabetic ordered list of everything I have seen in the last 7 days. Again, easier to just use google.
The one thing that is better and faster than google, is not closing those tabs that may contain the stuff I need.
Of course, it's not really possible to search the text body of open tabs, unless you search them one by one.
But I'm going to ask for only one computing miracle at a time !
What I'd recommend, based on the insistence that seeing to not change your workflow, is to locally download the pages you have open with httrack, wget or a similar application. This would allow you to locally search all your tabs and their contents very quickly without Google, they will load faster because of lack of needing to redownload them, which if I understand correctly Firefox is trying to do at some level.
Why don't you just open less tabs? Just close the browser when you are done using it
I am not sure what you’re working on but from your answers I’ve read you seems to need access to a lot of information with a few keystrokes, like searching for a keyword or tag.
In my opinion you are using the wrong tool for that. Ditch the browser and learn about the Zettelkasten way of working. It is really powerful for plenty of applications like science, studies, dev, or even the way I use it, author repository of ideas/concepts/stuff I need when writing a book.
You can do that with several software but I like obsidian for that (and because of all its plugins you could probably find something to automatically copy webpage content)
On the downside side :
- You’ll have to learn Zettelkasten, Obsidian etc
- Obviously do the work of writing (or copy pasting) your vault.
But on the plus size :
- You’ll have all the information you need at your fingertips, searchable with keywords, tags, associations etc.
- Everything is basic text MD files so it will still be readable by any text editor or terminal in the next century.
- You can have images, run code, do some mathlib, jupyter etc inside.
- Text is light, easy to store, backup and retrieve.
- If you do good enough you can have a satisfying visual representation of your new brain, kinda mindmap (which is also possible)
Cool I would love to navigate my data in a manner similar to this. However not obsidian, I am in the process of de-googling and I have severe cloud fatigue. But maybe QOwnNotes
I'm hoping something like Archivebox or squid or some other software can help me, autodump everything in a way that will become accessible to these second party data management software. Hopefully in a manner as transparent as opening a tab.
QOwnNotes is good but the UI is a little bare. Consider logseq
Can you please tell me about that obsidian, google and cloud relationship ? Cause I don’t know anything about that and I’m curious.
I don’t use any cloud except my own, self hosting FTW!
From a practical standpoint, it's hard to imagine what you could possibly be doing where it's beneficial to have a thousand tabs open.
If I'm writing a research paper, I might want 5 or 10 tabs open at a given time. Let's say I'm a little chaotic so I get up to 20. And then limitations on my working memory kick in, and having any more open tabs actually makes me worse off.
But then let's suppose it's a thesis that's 50 pages long. So I might be relying on 40 or 50 references. I'm not relying on them all at the same time, right? So I definitely don't want to keep those tabs open all at the same time.
What I could do, and what you could consider, is either bookmarking things or using archive.org to make a backup of the pages.
In one of the other comments you mentioned Facebook. That has me a little concerned again with your objectives. If it's something private on Facebook that can't be recovered later, and you need something reliable, then you have no choice but to do long screenshots or scrolling videos. If it's not reliable, then why do you care so much to keep the window open? Just close the window, remember whatever you remember, and move on with your life.
Whatever you do, here's a few rules of thumb... Your web browser is not an archiving tool. Printing to PDF is one way to archive things. There are other ways to archive things too. You don't actually need to archive as much as you might think you need to archive. Most of the things that we think might be important now actually won't be useful at all three months from now. Rarely would one actually want to have a thousand sources of information for any given task.
Zotero is a citation manager, with a firefox extension to save an article (but really, a tab) with one click.
It also has fulltext search. You can search snapshots of everything you save.
"But I can't save all my tabs at once"
(There are some solutions, but nothIng official)
Save as you go. Computers simply don't have enough ram for 2000 tabs.
Anyway, it also seems to be able to run javascript plugins, and I saw you have some experience with that.
It also has support for folders, so you can organize it a bit better than tabs work for that.
Zotero
I like the sound of that, thanks !
I have no direct solution to you exact problem but your usage of tabs sounds like a nightmare.
A while back I found Omnivore which works like a charm if you want to "freeze" the contents of a website to read them later. You can also self host it if you like.
I took it a step further because I love Obsidian as personal knowledge management and I want to have everything in one place. There's a plugin to sync all your saved pages from Omnivore to Obsidian. In the template for it I then have my marked highlights, the links to the version in Omnivore and the original URL and also the whole content. So I have all of that in markdown which is really nice to work with.
Maybe that's a solution you too could be happy with.
Thanks, never heard of Omnivore
"Distraction free. Privacy focused. Open source"
They do hit the right notes. I was going to try QOwnNotes but I'll put that on my list.
I terminate Firefox and reopen it any time it's chewing up my RAM, but I usually don't have more than 500 tabs open at any one time. My tabs persist when Firefox starts again, but tabs don't fully load until I click on them again. This saves my memory from getting chewed up immediately, and can usually go a week or so before I need to do it again.
What? Even 500 tabs? I don't understand this. I get about 10 open and I can't read what they are. Please share a pic of what it has to look like with that many tabs open because I totally do not get this? I feel like this would be akin to asking "I can't see out of my car windshield because I have completely covered it with sticky notes. How can I get to where I need to go?" This is not how browsers were designed to work.
I often have 100+, so I set a fixed width for tabs so I can see more and they don't get too small. To find tabs, I use the drop down to see a scrollable list. But honestly, the biggest win is the "switch to tab" feature when typing in the URL bar.
I see about 20 at a time, and they're usually all related to the same topic because I opened them around the same time.
When I'm done with a project, I "close tabs to the right" and it's clean again.
Here is a quick scroll through my "tab manager plus" view, zoomed out hopefully this doesn't contain doxxing information
I have over a thousand just like OP and it works fine. Use a tree style tab browser and it’s much more usable than chrome or anything like that. OP’s problem is not having too many tabs.
You could just set Firefox to clear everything on close.
Reduce to a sane number. Like less than 20.
You have 64GB RAM and that's still not enough for your browser. Wow.
I've come away from this with only more questions. What does your Downloads folder/Filesystem look like? Do you have notebooks or any real world allocation of information? What's that like? What kinds of things do you keep in a junk drawer?
Absolutely fascinating.
I do not waste time sorting, emails, downloads and bookmarks
For my linux ISOs, which I have approx 60 terabyte of, I use dedicated sorting software and it does a really good job of keeping it all organized. I also make liberal use of symlinks and hardlinks to keep the original alive while also keeping things organized.
As for notes, I have notepad++ with an endless series of titled untitled text files of everything I ever want to remembered. Shared accross computers using a local git server
On my phone I have google keep which has a list of notes that has long since become far too long to scroll to the bottom off of. I am in the process of degoogling and I want to switch to a selfhosted file centric markdown note taking web app, not decided on which but this video is probably going to be one of them.
I don't have a junk drawer, my stuff is sorted into bins, here is a glimpse of that
Those ISOs must go back YEARS! Same with the files! What sorting software helps keep track of all that?
Notepad++ surely has some type of global search feature to help find the thought you saved for later, right? I'm utterly impressed with how much stuff you seem to have around, yet can still find and make sense of it. I would have long since buried myself under it all and given up.
Why do you need so many open tabs to begin with? What is your usecase?
You are manually caching web content. Were you aware that (a) your browser does that for you; (b) the internet does that for you ?
I'm as guilty of this as anyone and can tell you from experience that it's sutpid.
Install Onetab
i use auto tab discard and it works great
It blows my mind how many tabs people open. I rarely use more than 4.
I feel like Firefox should just start hard limiting the number of tabs.
Here is an alternative Piped link(s):
https://piped.video/Tggq4jPPZaA
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source; check me out at GitHub.
I came in here knowing exactly what the comments would look like, and I'm still disappointed. "Just don't use so many tabs" is not an answer. If you don't have anything constructive to say, just move on instead of getting uppity about...not using browsers very heavily or understanding other use cases.
Yeah, thousands of tabs seems extreme. But "you should dedicate a larger amount of time and effort all day, every day to make the computer's job easier" is a bad take. That's obviously worse than OP's existing workflow.
Sorry OP, I don't have a real answer either. You might find Arc Browser's tab system to suit you better, but since it's chromium-based I suspect performance might be worse.
Edit: out of curiosity, how much memory does your PC have, and how much is Firefox using during these freezes? I wonder how much of the delay is caused by swapping.
Just don't use so many tabs" is not an answer.
Yes it is. If somone is holding a knife upside down and complaining it doesn't cut their steak, are you rude and ignore them? or comment that they're doing it wrong? OP has the knife upside down.
FF (or any browser AFAIK) is not designed to do this , OP is doing it wrong is a valid answer.
On the flip side OP wants FF to change so it can do what they want, which is also valid. After all, a sensible person adapts to the environment around them, an insane person expects the environment to adapt to them, therefore all progess is made by insane people. Have at it OP :)
Yes it is.
It is worse than OP's existing workflow, even though the existing workflow sucks. "Do this thing that sucks even more" is not an answer. "I don't have this problem, so you must be mentally ill" is also not an answer.
an insane person expects the environment to adapt to them, therefore all progess is made by insane people
LOL. I love this.
FF was designed to do this. There have been hundreds of bug reports that have been fixed over the years to literally make FF handle thousands of tabs just fine. If instead you had operated under the assumption that something was wrong you literally could help OP resolve the issue which is most likely something like their swap mem or the extension itself being written badly.
Frankly the answer is to not use so many tabs.
I think it’s crazy to need more than 10 active tabs open, let along thousands. I’m a software developer who will regularly go down rabbit holes and I’d never dream of opening so many tabs.
The fact is OP isn’t using the browser in a way that it was designed for. Plus they’re being unreceptive and rude in some of their replies.
Ff literally was designed for this. It has had a significant number of bug reports over the years that improved this exact thing, opening thousands of tabs. That’s not the problem (I also have thousands of tabs open). If instead you had operated under the assumption that that wasn’t the problem maybe you could have helped OP find the actual problem, which I bet is probably that they disabled swap mem or the extension they’re using isn’t written well.
Yeah, thousands of tabs seems extreme. But "you should dedicate a larger amount of time and effort all day, every day to make the computer's job easier" is a bad take.
Computers don't magically make things easier, it just does as it is told(as instructed by code). Computers don't come out as a self-made human assistant that adapts to your personal needs and can magically do anything you want.
If it's so much of a burden that smarter people haven't figured it out, go at it. You might just fix it.
But "to make computer's job easier" is the dumbest shit I've heard in a while.
How could one manage to open more than 100 tabs is beyond me. Just close the browser. This is insanity. I don't know how you keep track of all those pages.
I am going to keep beating that drum until firefox gets better It's already improving, I used to struggle at 700 tabs now I almost make it to 2000. Of course it is mostly artifice as most tabs get fully discarded and what I want is all tab texts in live memory and the ability to search all tab text. Maybe even text search in all pictures in all tabs using object recognition, but clearly we're not anywhere near that yet !
