So, I've done a lot of stupid things, but this one ranks pretty high. So I'm looking at files here and there, fooling around and tweaking things if I can, but after a while, I get sick and tired of having to fiddle around because a file's owner is root. So, completely ignoring the fact I could start an x session as root, I perform the following command:
chown -hR MY-USERNAME / I'm thinking to myself, "Oh look at me, I'm so smart" until I turn off my laptop for the night and come back this morning, and ubuntu (10.10, 32 bit) says it could not change ICE authority (or something like that) and a few more error messages. Then, I boot into the recovery console, and again, not even bothering to think anything through properly, I chown everything back to root, then chown my home folder back to me. Anyway, I still get the error message, execpt now I can't alt-c to close the first windows that talks about the ICE authority file.
One of my friends has an ubuntu partition, so I can ask him about certain ownerships, but that could take a while and I don't know where to start.
I accidentally changed etc., folder ownership, and now my computer doesn't even start up. I tried starting up in failsafe mode and changing the ownership from the root console, but somehow I wasn't allowed to do that. Then I loaded kubuntu from a live disk and I changed etc ownership to root. I thought that would clear up the mess, but apparently live disk's root is not equivalent to system's: when I try to start up the computer I get a message saying the filesystem is readonly. I'm not too concerned, because I have a complete backup, but I'd rather avoid the time of re-installing all my software again.
It seems I had some kind of intrusion and I found 6 files changed its ownership to user 1035 and group 1035, I don't know how but I need to change them back to its original owner (root) because one of them is the ls command and the other is the ifconfig how can I revert them to its original state? I cant do it with chown.
I have ubuntu 10.04 and winxp dual boot and if you leave magic jack plugged in the usb ubuntu will boot up to the desk top sometimes and freeze and other times nothing. No boot at all. Also had same thing happen with linux lime 9. For some reason the the magic jack when inserted into the usb of computer revents linux form booting. I thought ubuntu was messed up but when I put linux lime on drive by it self found out the magic jack was the cause mentioned this.
After a stint of using Ubuntu exclusively, I'm trying to install Windows 7 again. However, when I select boot from CD/DVD in the boot menu, I just get the GRUB menu.
Ubuntu Lucid 10.04 Beta 2 64-bit Dual-boot with Windows 7 Professional 64-bit Compaq CQ50-215NR laptop with 3GB system memory, 1.9GHz Athlon X2 QL-60 dual-core processor, nVIDIA GeForce 8200M G, 160GB HDD
Today I decided to install the BURG bootmenu on my computer so that I could enjoy nice, pretty graphical bootmenus instead of boring old text-based ones.I installed and configured BURG using this tutorial(after some few snafus wherein I tried to use older tutorials and went through many needless complicated steps that didn't end up working anyway).Restarted, and it worked great - except for one thing. Windows would no longer boot.It would get to the "Starting Windows" screen, then about halfway through there it would flash and restart.It wouldn't start in "Startup Repair Mode" either (the same thing happened). So, I did some frantic googling, and discovered this thread. I followed a set of instructions posted therein to install Lilo, reconfigure my MBR to Windows' liking, and reinstall Grub2.Windows booted after that.Well, I thought that the problem must have been caused by one of the numerous false starts I made while trying to get BURG installed, so I installed BURG again, set it up again, and bang - Windows wouldn't boot any more.So, I re-did the MBR with Lilo again, only this time I forgot to reinstall Grub2 (oops!) and had to boot from my LiveCD and install it thataway.At any rate, I'm fairly certain that installing BURG is what's keeping Windows from booting.
I'm running Ubuntu Server 11.04 on a home PC as a local-only file and web server for some private projects.After a few minutes Ubuntu will stop responding to the network and a cold reboot is required to re-enable the NIC.I've been able to stop the above occurring with a user logged in and constantly pinging the gateway, although this is obviously undesirable.
I read here and elsewhere that a fix is to set acpi=off in grub. I have done this but now instead Ubuntu will not boot and shows the message:
do_IRQ: 1.55 No irq handle for vector (irq -1) Disabling IRQ #34
and hangs there indefinitely. The system will boot fine if I revert the changes but obviously the sleep problem remains.The BIOS does not have an option to disable ACPI, just select S1 or S3.Is there anything else I should be looking for in order to disable sleep? Otherwise the system is working fine, and when it's awake I haven't had any other trouble.
So I recently built a gaming computer, which I typically use windows for. I'm mainly a linux user and I have separate computers for that. But, I wanted to try an experiment with linux on my gaming computer due to it having the best hardware I have.Anyways, I have an AMD athlon II x3, with the 4th core unlocked. Windows will boot up and use it just fine - it considers it a phenom II x4 b40. However, linux will not boot up as long as that 4th core is unlocked. grub is responsive but once it begins booting, it doesn't really do anything. its not just debian, its any linux distro.
i've never heard of linux having problems with core unlocking, and i have a pretty good motherboard (one of the top 5 socket am3 boards). i'm also very surprised to see windows somehow manages to accept this but linux doesn't.
I am using Fedora core 10. I have changed my partition size of Linux from windows. After I finished resizing the partition, I rebooted my system to the Linux platform. While booting it gave me an error: repair filesytem #1: I don't know what to do?
Since installing Lucid, I find that the shutdown button doesn't work (does nothing) if the OpenOffice.org quickstarter is running. To shutdown I either have to kill the quickstarter or run shutdown manually from a terminal window. Hibernation doesn't work either, although suspension does.
I recently installed virtual box on debian and after it had finished my terminal informed me that I could remove some "unnecessary" software by use of sudo apt-get autoremove. When I did this, some of the icons on the desktop changed and all of the icons in the drop down menu on the bar at the top of the screen also changed to ordinary folder symbols. The theme that I was using also went away. I restarted the computer and it booted back into a shell prompt with no GUI. I tried to get back to the GUI using alt+f7 but it didn't seem to exist
In the past month or two I've run into a strange problem where Grub gives error 18 "Selected cylinder exceeds maximum supported by BIOS". Strange because: 1) after a few tries it mysteriously starts working again without any change on my part 2) I am not running a dual boot system, which seems to be the usual cause of this problem 3) though I have SATA hard drives, I have the RAID capabilities disabled in the BIOS, which can also apparently lead to this problem. I'm currently running lucid but it also happened once before I upgraded. It seemed to start happening not too long after I installed a new 1TB drive, though that drive is only mounted under /mnt/backup, and not anywhere else of importance, so it may be a coincidence.
Here are some specs on my machine: BIOS is ALive NF6G-VSTA P1.80 I have 2 SATA drives (no IDE drives except DVD-ROM): /dev/sda is a 200GB Samsung drive with two partitions - sda1 is about 197GB and includes / (including /boot) - sda2 is about 3GB of swap /dev/sdb is a 1TB Western Digital Drive - 1 partition mounted at /mnt/backup If the problem is that I don't have a separate partition for /boot, shouldn't the error show up every time I boot, not just once in a while?
Upgrade calls for ensuring latest updates to Karmic are installed before upgrading to 10.04 LTS. Every attempt to update hangs, yielding the following dialogue:
W: GPG error: http://ppa.launchpad.net karmic Release: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY B152F042D246C25D I guess I am sort of stuck in nowhere land here.
I recently noticed a bug with ndiswrapper in Ubuntu 10.4 which would prevent the mounting of new USB device classes. This bug has been noticed somewhere else, in particular in at least one older Ubuntu version, but has been carried over. This doesn't seem to be widely discussed, or the solution worked anecdotally, but the problem is actually more general in nature. At the moment I found a workaround that always works, but I don't have a permanent fix.
The problem kicks in only after ndiswrapper module is loaded into the kernel. Devices of the same USB *types* that were recognized *before* ndiswrapper is loaded will continue to work. For example, an USB memory stick will be mounted fine if an USB hard disk had already been used at some point before ndiswrapper was loaded. However, if one plugs in any other new type of device, for example, an USB mouse, after ndiswrapper is loaded, the new USB device will not be mounted. The workaround solution is to disable ndiswrapper before introducing a new type of USB device by doing:sudo modprobe -r ndiswrapper
Then, plug in the USB device. After it is successfully mounted, do: sudo modprobe ndiswrapper
The corollary is that if ndiswrapper is loaded into the kernel upon boot-up, only USB devices which were present at the boot stage will be recognized. But performing the above kludge will allow new devices to be recognized.It'd be nice if someone can come up with a more elegant fix.
I am running ubuntu 10.10. I recently enabled the firewall and installed Firestarter to configure it. Bad decision apparently. I can't connect to the internet using Firefox unless I first stop the firewall using Firestarter. After I do that, Firefox connects to the internet just fine.
If I uninstall Firestarter, will the ubuntu firewall function as it did originally, before I configured it? Or will it continue to function the way it does right now, which doesn't allow me to connect to the internet?
I put my two SATA hard drives in a new computer (new mb, cpu, no pci SATA adapter, built-in nVidia graphics), and I thought everything was going well. The Mythbuntu logo came up, but the bar at the bottom did not move. It eventually dropped me out to a (initramfs) prompt.
When I used the recovery mode, this is where it had problems:
[ 4.756000]ata1:SATA link up 3.0 Gbps (SStatus 123 SControl 300) [ 5.068000]ata2:SATA link down (SSTatus 0 SControl 300) [ 5.552000]ata3:SATA link up 1.5 Gbps (SStatus 113 SControl 300) [ 5.864000]ata4:SATA link down (SSTatus 0 SControl 300)
[Code]....
I had a SATA PCI card in my old computer, would adding that to the new system make it work? Is there someway to update the UUID numbers (I think I have done this before).
Prior to posting this I entered search arguments, "Busybox," then "initramfs" to attempt to find another user who may have experienced the same thing; if the solution is imbedded in the message threads, I must have missed it. On hand is a Ubuntu Server 8.04.3 installer CD. What works: Installs to a Mac Pro under both Parallels, and then under Fusion both were successful first try.
Install to a newly purchased 2 gig Acer Aspire One seems to have succeeded, but upon launching from Grub, Busybox/initramfs intercepts and halts the launch with an error I don't understand:
The new Acer-PC Aspire One was partitioned into four 80 gig partitions using the Paragon software to do the partitioning. XP is on the first, and Ubuntu Server on the next 80 gig partition. Install was "successfully" done from a Lite-On external CD/DVD over USB.
During the install I changed the default name from Ubuntu to something else (probably a bad idea), and I naively entered "/" root to answer the mount question (finding nothing to tell me what to do). Ubuntu required a disk driver choice from a long list - finding no reference or guidance as to what to enter, I naively entered IDE Generic, and it took (I guess it "took").
FWIW the Ubuntu installer didn't require me to indentify the mount entry or the disk driver during installs on my Mac (Fusion/Parallels).I did connect the Internet during the install. I did checkoff/install LAMP during the install from CD. Goal: To use Joomla in a mock server environment.
I cannot update my system either with Update Manager nor with Synaptic. The system tries to download the updates, and then generates an error message that I have broken packages. So, I click on Fix Broken Packages, etc., and when running get the following error message:
E: /var/cache/apt/archives/libavcodec52_5%3a0.6.3-0.0_i386.deb: short read on buffer copy for backend dpkg-deb during `./usr/lib/i686/cmov/libavcodec.so.52.72.2'
This appears to me to be a system bug. I'm running the 32-bit version of Squeeze. Re-install, or ?
It seems today's kernel update has given a few people some difficulty - but my symptoms seem pretty unique.
* I go through GRUB normally. I 'esc' out of plymouth, and watch my services start. I get message 'mdadm: No arrays found in config file'. Don't remember seeing this before, don't have any RAID devices.
* Next, my file-system gets re-mounted in read-write mode (don't know if this is normal or not). Then, my services begin to start-up. My last service is 'crond'. All services start normally. When it gets to 'crond' the [OK] status is never reported. Visually, the screen flickers a few times, stops, then nothing happens. I can type characters though, and they do show up.
* If I then shutdown my computer by pressing the power button, the shutdown procedure begins. 'crond' shuts down successfully (indicating it started just fine), and the computer goes down normally.
* I don't believe this is an NVIDIA issue. I rescue-moded and changed my XORG.conf to the basic VESA driver to eliminate this possibility.
* Worst of all -- No older kernels work now either. They behave in the same way. When I try to boot to other run-levels, no luck either. This tempts me to think the problem might be file-system related and the re-generation of grub.conf threw something off?
I have a problem every time I start steam trough wine SELinux is blocking it can I switch off SELinux or do something so steam can start. I have already tried to ignore it and that doesnt work does someone know how to fix this.
I'm writing a game for Linux using Xlib whereby when a window becomes active it will take control of the mouse using XGrabPointer. The idea being that if the users switches applications through Alt+Tab the game will relinquish the mouse with a call to XUngrabPointer.
The game is using the FocusIn event to determine when it has been activated and needs to grab the mouse, and likewise uses the FocusOut event to release the mouse. This causes a problem because for some reason an Alt+Tab from the keyboards generates two FocusOut events and then a FocusIn event causing the game to take control and no application switching occurs.
If the game does not grab the mouse, then application switching works fine.
What is needed so that application switching can work while the game has control of the mouse?
I am running Fedora 12 on an Intel Core2 Duo machine. I am also using Compiz as the window manager.
The following article prompted me to ask the question if there is any software that will prevents devolving a computers location? [URL]... Wi-Fi systems on smartphones and laptops may be used to secretly report the locations of users, their family and friends unless changes are made to their design, warns Ontario's Information and Privacy Commissioner Ann Cavoukian.
I didn't see another thread about Google Picasa being blocked by SE Linux. How do I tell my machine to allow it to run? I have the same problem with a program called Smartboard that I have installed but can't open.
i installed smeegol out of curiosity. and i am planning on keeping it installed as a alternative desktop. so "get rid of smeegol" is not an acceptable answer to my question. my question is also not so much about how can i turn this behaviour off but i wanna understand whats going on. here it is:when i log out of kde and start smeegol i am in offline mode per default. now thats unconfortable but i can live with it. what is however very weired and this is what the question is all about: when i leave smeegol with
init 3 init 5
and i start kde again my network manager doesnt start properly. i tried switching to traditional and switching back. still not working i tried killing the processes and restarting them still not working. i tried restarting my hole pc (i know stupid but it was a desperate move) still not working i tried a dozen other things that are even more desperate the only thing that got my network working again was starting smeegol again and switching to online mode
then ctrl + alt + backspace and back into kde suddenly i am online again
this is confusing how can smeegol my network manager up that hard that even a restart wont get it connected.
Following the instructions, I get to a little after "setting software channels" and a blcr-dkms_0.8.2-13 blacklist error pops-up and ends the installation.
I'm using MATE with Lightdm as display manager in Jessie and I've been using a very simple script run from /usr/share/lightdm/lightdm.conf.d/01_debian.conf to slow down the mouse cursor, something like
Code: Select all#!/bin/bash
xinput --set-prop 10 269 -1
xinput --set-prop 10 267 1.100000 the script was placed in my ~ and called from 01_debian.conf , and so far it used to work.
Lately this script wasn't executed properly, most likely because I had unplugged the mouse from its usb port and then plugged again in a different port, and as a result Lightdm wouldn't start at all.
After much trying, because this by coincidence happened after the update to 8.4, I realized that it wasn't a serious failure of the X server as I suspected (the update included a new version of the flgrx driver), but a simple matter of Lightdm failing because of this script: in fact, I could still manually start the session with startx as regular user, although this only gave me access to a LXDE session instead of MATE (I have also the LXDE environment because when I installed Jessie I did use the LXDE version and then added MATE).
What happened was that the mouse properties were scrambled up, so xinput --set-prop 10 269 -1 had now to be something like xinput --set-prop 10 268 -1 and so on - therefore the script failed and in turns Lightdm failed as well.
-is this supposed to happen when you unplug the mouse and then plug it back in a different usb port? It reminds me of some windows-like behavior, where usb peripherals had to be always in the same port to work properly.
-is Ligthdm supposed to fail because of this, or is it kind of a bug? It looks weird to me that instead of simply not executing the script, the whole display manager fails.
In testing, with an identical setup, as a matter of fact the script failed as well (it's on another partition on the same hardware) but Lightdm started as usual.
-when trying to modify the mouse properties in that script, I couldn't get them right again: what looked correct in the booted system, was once again off at the next reboot (i.e. property 269 was 270 again and so on), therefore I ditched the above setup and placed everything in the .xsessionrc file, and it worked again.
-why startx defaulted to the LXDE environment? where is this setting stored? How could I have specified to start a MATE session instead from command line?
[URL]... It seems like I should be able to install i686 and x86_64 versions of the same package. This same dependency also prevents installing Adobe Reader.
Yesterday I updated my F15 x86_64 system from kernel 2.6.38.6-27 to 2.6.38.8-35. After the reboot, my triple-head system complained of not having enough CRTCs and only 2 screens worked. Other components were also updated, including some X11 components, however selecting a fallback kernel at the grub prompt indicates this follows the kernel (fall back to old kernel has all 3 monitors working). xrandr --verbose on the 2.6.38.6-27 kernel shows each of the 3 monitors assigned to separate CRTC entries, with CRTCs 0 thru 5 available.
Same command on the new 2.6.38.8-35 system only shows CRTCs 0 & 1 available. My system using an XFX Radeon HD5450 card, with 3 interfaces: HDMI, DVI, VGA. The default Fedora open source "radeon" driver allows all 3 interfaces to be usable. However, with the new kernel, the loss of several CRTCs prevents the 3rd monitor from working.What are the thoughts from the community? I've searched various forums here and haven't seen similar symptoms reported. I'm looking for any direction/experience/thingsToTry from others before trying to open a bug report. (Attempts to use the Catalyst driver several months ago resulted in only 2 of the 3 connected monitors working.
