Ubuntu :: Ext4 Is Bleeding Space, Cannot Determine Why
Jul 20, 2010
I am running Ubuntu 10.04 amd64. So I have just returned from a week vacation to see that my / is full. When I had left, the ~80G partition was only consuming around 40G of space. I feel like this is a bug and i haven't tried a custom kernel yet to see if it is kernel level. Here is all the good information though to help debug:
Code:
mount
# mount
/dev/mapper/pdc_cfefedbci3 on / type ext4 (rw,noatime,user_xattr,errors=remount-ro)
proc on /proc type proc (rw,noexec,nosuid,nodev)
[Code].....
I am using a fakeraid with dm-raid on a raid0 for my /dev/mapper discs. I just wonder if anyone else is getting this error or if it is a problem with ext4.
You would think it would be easy to do this (and essential to maximizing and predicting the usage of tapes), but apparently this isn't so. The program MT(1) actually had some commands that give the block positition (if supported by the drive) from which the remaining space could be deduced (even if you couldn't predict exactly how much space the next archive would take up). However I'm using MT(1L) which no longer has such commands.
On one of my servers the "free" command tells me that a lot of swap space are in use. What I'd like to do is to determine which processes have been swapped out. I tried issuing "top" and sort by the "swap" column, but this doesn't seem to provide correct values - when performing the same excersize on another server with close to no pages swapped out, the sum when adding the swap value for each process greatly exceeds the swap usage reported by "free". So how do I go about determining the swap space used for individual processes?
I am using LVM2 and have shrinked my /home partition and extended my / partition but I'm not sure if I used all the free space when growing my / partition. How can I find out? I prefer using the terminal if there is a graphical way to do this but I would like to know both ways if there are two ways.
according to the system manager of my machine, one of my ext4 partitions (home folder) reads 7.6GB free but only 574MB available and the disks "fills" when the 574MB are used, and i'm really needing those 7G right now, so:
why does this happen? are those 7G used in anything? any way to allow the system use that free space?
I used gparted to format a 360GB partition with ext4 and I expected it to have at most several hundred MB used for whatever reason by the file-system, but it said 18GB were used. How come? Are there any file-system settings I should have paid attention to?
I clean installed Ubuntu 10.10 by shrinking my Windows 7 partition slightly. Now that I want to expand my Linux partition, I shrunk my Win 7 partition from Windows OS. From Ubuntu, the partition manager shows /dev/sda1 contains the Win 7 and unallocated partition. /dev/sda2 contains the Linux and swap partitions. I can't seem to expand my Linux partition (ext4) in sda2 with the unallocated space in sda1. I also can't shift the unallocated space in sda1 to sda2. Any idea how to expand my main Linux partition with the unallocated space?
I have a 6x1TB RAID5 set up for testing on ubuntu created with mdadm and formatted with an ext4 fs.
This is being shared over CIFS for windows clients. When looking at the fs from both the file box and the clients, it says 4.47TiB total capacity, and 4.24TiB free space. The only folder is Lost+Found which is empty.
I don't have much experience in Linux filesystems as of yet and I don't understand where this 300 gigs has gone!
My root partition was filling up, with only 500 mbs left, I wanted to resize my root partition from 20 Gb to 40Gb
So I resized my partition by using these steps:
Using Gparted to resize another partition to give space for the EXT4 Using fdisk, deleting the root partition (on /dev/sda2), and creating it again using the new size resize2fs /dev/sda2 Updating grub2
But now the problem is that although I can boot in my new partition and the new partition shows it is 40Gb, but the free size was still 500mb. So I booted from a LiveCD and checked with e2fsck -p /dev/sda2, it reported clean. So I added the -f flag (force check), still, the drive is full.
I have just purchased a 2TB drive for my server and I was trying to get an idea of the differences between these file systems or other file systems out there. What is the amount of space after formatting for ext4, ext3, and ntfs?
This weekend, I installed Debian Squeeze on my server. I've formatted all the hard disks to EXT4, and I'm using kernel version 2.6.32-686-bigmem.When I tried to install the program saidar, it surprised me why it does not show my hard drives under 'mountpoint' [URL] <-- Saidar screenshot) as I could when I ran with Debian Lenny with the same kernel, but where the hard drives were formatted in EXT3. My laptop which has Ubuntu 10.04 as OS and the hard drive is formatted in EXT4 can easily show the hard drive in saidar. I also tried to install PHP SysInfo on the Debian computer, but it does not bother to show anything on the hard disks
I tried to check fstab file and I can see that Debian uses UUID to identify the hard drives, but I've tried to change it to something with /dev/sdx, but it did not help either.[URL] (fstab file)
I know that Debian squeeze is very new, but it would be nice if someone could give me a hint what might be wrong, because I am a little tired of all time to use 'du-hs' command To find out how much space is spent on the various drives, since the command is a little slow, since hard disks are well filled.
Some searches seemed to indicate that it's been awhile since there's been a build thread, so I thought I'd start one. I'm looking to replace a six-year old machine (Athlon XP 3200+ processor (introduced in 2003), integrated graphics with no dedicated graphics card). I'm rocking Windows XP Home Edition (32 bit).
I'm not a "computer hobbyist," far from it. I use my machine mainly for browsing, word processing, and business (QuickBooks and Turbo Tax). It is a tool. I don't really enjoy fiddling around with computers. My current machine is now barely able to support a good research session on the 'Net anymore, the little processor seemingly gagging on the heavy flash content you see everywhere nowadays, the 2GB (the most she'll take) of RAM easily filled.
I'm planning some rough travel international motorcycle trips (see my motorcycle build thread here (don't worry about registering on this motorcycle forum.....they are good folks and I've never gotten any spam from them) and purchased, a while back, an Asus eee PC 901 with the OEM Xandros (a bastardized offshoot of Etch) loaded to use with those trips. From learning that OS, I saw for myself the benefits of Linux (vs. Windows). Of all the distros, Debian is the most attractive to me (even though I'm currently rocking Puppeee (new version) on the 901.....it makes that little machine zoooooom!) as I really like the package management system, the logic, and the stability. I also like the feeling of personal responsibility that comes with Debian.
So, back to the build. I'm currently planning on it being a pure Debian machine. I'm ready to move on from Windows, for reasons I don't have to explain here. I stated that I don't particularly care to "play" with computers, but I'm willing to put in some front end labor in order to enjoy a long-term, low-maintenance relationship with my new machine. With that philosophy in mind, I'm looking to build a machine that will serve my simple needs for ten years(!) So, since the costs will be spread out so far, I'm ready to drop some serious (for me) coin on this project in order to avoid having to go through what is, for me, a rather painful process of research again any time soon.
I have AMD Athlon 3500+ (x86_64) CPU and I am okay with the Ubuntu package maintainer's kernel version (as of now it is 2.6.35-24-generic). What will I gain if I update to latest stable kernel (2.6.36.2)? A fraction of performance?
I hesitated on the title, generally can solve problems w/o this (20+ year *nix/linux experience).
hardware Dell 1505, ATI X1400 (nVidia went south replaced with ATI, used at this stage of life for this laptop).Acer x221w external monitor.
Recently upgraded to 10.04 (should have waited, but hadn't upgraded in a long time).
No real fixes found for the ati X1400 in searches (fixed it on 9.04 and 9.10).
So to troubleshoot, wanted to go to console mode, and start X manually and hopefully debug the config to get the thing working.
So, I get ready to take it down off of X only to find, there are no more consoles???
To work this out and to even come to an end result, when X does not work, you need a console to debug it. There is code in /etc/init.d/console-screen.sh for starting consoles...
Yet they do not start, notices some comments about doing it 'clever' in there, all I can say is 'clever' is more risky than workable, it worked before, why change it? It did not make it better, it made it harder, so now, to get it to work, I have to go debug someone else's 'clever' code? BTW, there are 'getty' running - but still no console.
Not so clever... We need a fix to get (at least at this level), the consoles running, so, where are you 'clever coder', tell us how to turn them back on.
Why in /etc/init/tty*.conf do they not start?
I really find it hard to say anything critical to something that is free, in general, I will not even comment, but just go ahead and find the fix myself, But to even address this, we (IMO) need pretty much a fix for the broken console access, before we can even address X in certain configurations.
A long time user of RH, even back to Yygdrasil, Soft Landing days, I was VERY impressed with ubuntu over the others for a desktop (got it on the Dell), but now, at 10.04, it has become cumbersome and broken.
It is a good product, but we need to get it back to where it is not so bleeding edge and broken.
I'm trying to compile the bleeding-edge compat-wireless because I know there are some ath9k fixes in it because I use in on my Debian and Ubuntu machines; however, when I try and run make, I receive this error:
Code: make: *** /lib/modules/2.6.38.8-35.fc15.i686/build: No such file or directory. Stop. make: *** [modules] Error 2
I got a problem booting ubuntu 10.4 RC but i solved it by replacing root partirion uuid in grub boot menu then I disapled totally uuid passing to linux from /etc/default/grub . but something else i noticed why grub choosed insmod=ext2 why not ext4 specially I use now ext4 .I tried by editing the grub boot menu replacing "insmod=ext2" by "insmod=ext4" it booted and the three lines error during booting that i used to see them science ubuntu 9.10 totally disappeared . really I dont understand can anybody explane for me.and if what i did was right ,can anybody tell me how to make grub always and permenantly detect ext4 as ext4 not as ext2.
I had 5.4 machine. Upgraded to 5.5 today via yum upgrade. All went fine. Rebooted. Wanted to convert root partition to ext4 (I have three partitions: /boot, / and swap). All of them on software RAID 1 (root is /dev/md2). I did the following for converting
yum install e4fsprogs tune2fs -O extents,uninit_bg,dir_index /dev/md2 nano /etc/fstab # I indicated here that my /dev/md2 is of ext4
Logical Memory Space of 4GB is divided in to 3GB User Space and 1GB Kernel Space. Always. Correct?
1. How can we change it? (just changing value of PAGE_OFFSET is okay?)
2. If system have only 256MB of memory (embedded system) and suppose Kernel Modules eat away all the memory during boot. User space will be left will no memory. Is this case possible?
was waiting to install the "cutting edge" Fedora 11 and I did it immediately after its official release. The 1st day I installed the Fedora 11 ...... the system declared a kernel problem. I don't know what was its problem. Anyway I continued with this OS and I was taken day by day all the official updates. I could see everyday that the system was not stable enough. Problems - problems - problems..... Sometimes there was a critical ext4 nfs error but my hard disk is new and ok. Fedora 11 was the only OS into this hard disk. So....I installed the "bleeding edge" Fedora 11 once again - I took again all the official updates and I tried again. Nothing better. Today Iam sure that the ext4 file system is the main problem here, it is quick but not secure enough. That is my conclusion. Yesterday Okt 11 2009 I installed again the Fedora 10 OS with the old ext3 file system and Iam so good again.
I have linux and windowsxp on one machine. I have only 3gigs free on the windowxp machine and 20gigs free on the linux machine. I want to transfer space from the linux box to the windows machine.Is this possible and what steps would I need to follow to do this?
i used gddrescure to clone an 80gb harddrive and this is the result ROFL.i guess you can only do this making sure the target drive is the same size, you see i didnt know lol so..i now have THIS problem.can anyone tell me how to turn my unallocated space into a usable 'free' space? i could play with gparted right now but i dont wanna do anything wrong, so if theres anyone who can tell me how to do this.
i made space by shrinking my window partition and so i have unallocated and would like to add to sda2 to have more space. Check out this pic. How can i do this?
I installed Ubuntu 10.10 with wubi and i have been enjoying my Ubuntu experience a lot. I installed quite a bit of programs and spent a couple hours customizing my machine. The problem is im running out of disc space. Any ideas on how i can add more space. I have gparted but i dont know where to move the free space to because wubi installed it.
my home partition is an extended one, and when i want to create an unallocated space the space will stay in that extended partition. but there is also an 7 gb unallocated space which i want to merge with the other unallocated space. I also cannot extend that partition over that 7 gb. how can i overcome that problem?
i m also uploading a screenshot of gparted.[URL]..
I am developing a I2C CDROM client driver. The CDROM firmware supports TOC information read through a I2C command. It sends the TOC information in burst ( Interrupts a GPIO pin when it is ready ) and my CPU does a I2C read to read the TOC. When the CDROM firmware finishes sending the last data burst , it informs my CPU that it is done with the TOC, by a flag in the last data burst. I would like to know, which is the most efficinet way I can send these TOC information to userspace?
I am writing device driver in which i have to call callback function from kernel space, which are saving my data. But the callback functions are in userspace. While accessing them i am getting segmentation fault.
I wanna write a file in kernel space but from my searching I can to know that instead of writeing file in kernel space ,I can write data to user space by copy_to_user space.
But link is missing ...I dont know how will my user space will access kernel space means my function in kernel space which will do copy_to_user /....How my user space function will call my kernel level function ..
Can any one of you provide me with some example file which are doing this .I know every char driver is using it ...but i could not trace back how user level function is accessing it ...i m confused between user space and kernel space.
