- cross-posted to:
- coolguides@lemmy.ca
- cross-posted to:
- coolguides@lemmy.ca
i always thought /usr stood for “user”. Please tell me I’m not the only one
Ken Thompson and Dennis Ritchie created Unix on a PDP-7 in 1969. Well around 1971 they upgraded to a PDP-11 with a pair of RK05 disk packs (1.5 megabytes each) for storage.
When the operating system grew too big to fit on the first RK05 disk pack (their root filesystem) they let it leak into the second one, which is where all the user home directories lived (which is why the mount was called /usr). They replicated all the OS directories under there (/bin, /sbin, /lib, /tmp…) and wrote files to those new directories because their original disk was out of space. When they got a third disk, they mounted it on /home and relocated all the user directories to there so the OS could consume all the space on both disks and grow to THREE WHOLE MEGABYTES. And thereafter /usr is used to store user programs while /home is used to store user data.
source: http://lists.busybox.net/pipermail/busybox/2010-December/074114.html
You’re not the only one 😅 🙋
I thought it was United System Resources.
And I still don’t know what’s the point in separating /bin, /sbin, /usr/bin and /usr/sbin.
Also /mnt and /media
Or why it’s /root and not /home/rootMostly historical reasons, /home was often a network mounted directory, but /root must be local.
And only regular users have their home in /home
And I still don’t know what’s the point in separating /bin, /sbin, /usr/bin and /usr/sbin.
This goes back to the olden days when disk space was measured in kilo and megabytes. /sbin/ and /usr/sbin have the files needed to start a bare bone Unix/Linux system, so that you could boot from a 800kb floppy and mount all other directories via network or other storage devices as needed.
Is there a reason to keep this structure other than „we’ve always been doing it like that“/backwards compatibility?
The structure is changing, many distributions already are merging more and more of the duplicated subdirectories in /usr/ with the counterparts in / but it takes time to complete that and at the moment those subdirectories are often still there but as symlinks to be compatible with older software (and sysadmins).
They hold “system binaries” meant for root user. It’s not a hard distinction but many if not most Linux fundamentals have their roots in very early computing, mainframes, Bell and Xerox, and this good idea has been carried into the here&now. Not sure about the provenance of this one, but it makes sense. isn’t /mnt /media different between distros? These aren’t hard and fast rules - some distros choose to keep files elsewhere from the “standard”.
/bin and /usr/bin, one is typically a symbolic link to another - they used to be stored on disks of different size, cost, and speed.
https://refspecs.linuxfoundation.org/FHS_3.0/fhs/ch03s16.html
https://unix.stackexchange.com/questions/5915/difference-between-bin-and-usr-bin
I think /mnt is where you manually mount a hard drive or other device if you’re just doing it temporarily, and /media has sub folders for stuff like cdrom drives or thumb drives?
Yeah, but why?
You can mount a hard drive anywhere, and why not put all the cdrom and thumbdrive folders in /mnt, too?It gets even more complicated nowadays because most DE will mount removable drives somewhere in folders like /run/$USER/
/media is for removable drives. If you mount something there, file managers like Gnome will show you the “eject” or “disconnect” button.
/mnt drives show up as regular network drives without that “eject” functionality.
/sbin are system binaries, eg root only stuff, dunno the rest but I would guess there are some historical reasons for the bin usr/bin separation
I know the distinction between /bin and /sbin, I just don’t know what purpose it serves.
Historically, /bin contained binaries that were needed before /usr was mounted during the boot process (/usr was usually on a networked drive).
Nowadays that’s obsolete, and most distros go ahead and merge the directories.
I think it originally did under old Unix, it was what /home is nowadays; “Unix System Resources” is a backronym.
deleted by creator
Yup same. I always wondered why there was a user folder when we already have home.
I was just about to post the same thing. I’ve been using Linux for almost 10 years. I never really understood the folder layout anyway into this detail. My reasoning always was that /lib was more system-wide and /usr/lib was for stuff installed for me only. That never made sense though, since there is only one /usr and not one for every user. But I never really thought further, I just let it be.
deleted by creator
deleted by creator
Meanwhile the Linux Standards Base cries in a corner.
wait /usr doesn’t mean user?
/etc has to be the worst name in there
I wonder why that isn’t /cfg? Is there a historical reason?
According to this, it’s been around since the 70’s and was originally just a catch-all for files that didn’t fit in the other default directories, but over time has come to be mostly used for config files. I assume it would cause utter mayhem to try and change the name now so I guess it just sticks. Someone suggested “Edit To Configure” as a backronym to try and make it make more sense if that helps anyone lol.
I too expected it to be “et cetera”.
Try naming a folder “CON” in Windows and learn the magic of old spaghetti code by a multi billion dollar company.
It’s probably the standard in both POSIX and the Single UNIX Specification, so I guess ask Ken Thompson?
deleted by creator
I learned about 16 years ago on a Solaris course that /usr wasn’t “user”, I still say “user”, but I’m happy to see the information spreading that that isn’t what it actually is.
I learned that just now.
It’s going to be TOUGH to mentally replace.
Wow, what an odd coincidence.
I used to pronounce it like yuzr, knowing that it wasn’t user, but not knowing what it was.
Now I have better context. Maybe I’ll go with U.S.R.If you want to confuse people… I pronounce /etc as “ets”, but one of my coworkers recently called it “slash e t c” and I had to ask him to repeat it a couple times before I figured out what he meant…
Well, considering that I am with coworkers who don’t remember when to and not to put the ‘/’ at the start of the file path (despite me explaining it to them multiple times), “slash e t c” is probably the better way.
A pedantic thing to say, surely, but the title really should’ve been: “Linux Directory Structure” – ‘Linux filesystems’ (the title in the graphic) refers to a different topic entirely; the title of this post mitigates the confusion a bit, though still, ‘directory structure’ is the better term.
Sure but for example I understand that /dev and /proc are actually kind of filesystems on their own
Yep, You are right. Done
Right?
I was expecting superiors to the fat & exfat file storage systems
A good first approximation.
So where in this setup would you mount a network share? Or am additional hard drive for storage? The latter is neither removable nor temporary. Also
/runis quite more than what this makes it seem (e.g. user mounts can be located there), there is practically only one system path for executables (/usr/bin)…Not saying that the graphic is inherently wrong or bad, but one shouldn’t think it’s the end all be all.
On debian when i mount an ftp server through my file browser it uses gvfs
This will mount it to /run/user/1000/gvfs/ftp:host=,port=,user=
I always thought /usr was for “user”… TIL
It is, this infographic is wrong. Or I guess technically some other standard could define it like the infographic, but the Filesystem Hierarchy Standard defines it as a secondary hierarchy specifically for user data.
I never understood the title for /usr. Now I do. Thanks!
It’s just short for “user;” “User System Resources” is probably a backronym.
This email explains it in detail: http://lists.busybox.net/pipermail/busybox/2010-December/074114.html
TLDR:
/usrstands foruserThat’s what I’d always thought. Thanks for correcting the bad info from the image. I’d hate to carry that bad info forward.
It feels like
/opt's official meaning is completely lost on developers/packagers (depending on who’s at fault), every single directory in my/optbelongs to standalone software that should just be put into either/usr/libor/usr/sharewith some symlinks or scripts into/usr/bin.I’ve also seen creating there deployment or configuration stack of your choice.
deleted by creator
Visualizing it like this makes it so clear how incredibly outdated this design is.
What is outdated?
Fun fact: you get more accurate info by simply running
man hierI’m pretty sure
sbinoriginally meant static binaries and not system binaries lolWould like an easy way to remember.
- mnt = mount
- opt = optional ?
- etc = etcetera ?
- proc = process ?
- srv = server ?
- var = variable ?
/srv stands probably for
serveas in serving static files like static websites. (Source)More information here: https://refspecs.linuxfoundation.org/FHS_3.0/fhs/index.html
deleted by creator
What’s the reason?
Probably so that you don’t accidentally write to a directory by mistake when it isn’t mounted, and then lose access when you mount something over it, all while services are looking for files that are only there sometimes.
I’ve had exactly this happen to me. It was my own fault but it took a bit of work figure out.
deleted by creator
Where would you mount non-removable media?
deleted by creator
Wow. Talk about ways to skin a cat.
I mount mine to /media using autofs.
I was, at one point, using /mnt but ran in to some situation that Proxmox didn’t like that involved bind mounts (can’t remember what) and shifted them all over to /media.























