With tar, it is possible to back up your desktop system like so (probably best done from runlevel 1 or a live medium):
# cd / # or wherever your root directory is mounted
# tar --exclude=dev/* --exclude=proc/* --exclude=sys/* --exclude=tmp/* --exclude=backup.tar.gz -czvpf backup.tar.gz .
[code]....
Ok, this just started happening with ubuntu 9.10.
When it first happened, I didnt know what to do so i accessed all my data, backed it up in a diff partition, reformated my drives and reinstalled a clean 9.10. But now, every now and then (i think its when i use frostwire and have other programs running) my ubuntu keeps freezing up, it becomes non-responsive to anything i do, so i have to do a hard shutdown, after which, when i reboot, it is not able to mount /home, so i run FSCK and it fixes it, but ive done this about 3 times now, its starting to get a bit annoying, i dont know if anyone else has this problem, or if its a known bug. Here are my specs:
Running: Ubuntu Karmic 9.10 64 bit
3 Sata HDD: 1 TB, 750 GB, 350 GB
4GB Ram
2 Nvidia 7900GS
with a Pentium core 2 duo @ 3.0GHz
I've been trying to back up my system to a tarball for quite a while now. I recently bought a tape drive, and it works. But I'm having a little bit of trouble getting tar to work--whenever I try to copy the files (either directly to the drive at /dev/st0 or to a tarball), I end up with a "file changed as we read it" error, and tar quits before the archive is done. Is there some way I can either prevent this from happening and/or tell tar to just skip that file and keep the job going?
Code:
$ cd /home
$ sudo tar -czf /dev/st0 soren {soren being my home directory, /dev/st0 the tape drive}
[sudo] password for soren:
tar: soren/.gvfs: Cannot stat: Permission denied
tar: soren/.local/share/Trash/files/From Removable Media/16GB Flash Drive Dump/Cliffs of Incognita/Cliffs of Incognita Music/Audio/Stereo 01_01.wav: File shrank by 7289884 bytes; padding with zeros
[Code]...
when I try to install cpp4.3 it says : cpp-4.3 depends on libc6 (>= 2.11); however version of libc6 on system is 2.7-18
when I try to install libc6 2.11.2-5_i386.deb it says : installing libc6 would break locales, and deconfigoration is not permitted
I run Etch. Yes, it's oldstable, but Lenny and Squeeze annoy the heck out of me. However, I'm trying to install the latest version (5.1) of Google Earth, but it requires >=glibc-2.4. Etch only offers 2.3.something. The earlier version of GE I have (4.3) does work, but is apparently no longer supported by Google as it fails to connect every time I fire it up. I found Lenny does have the required libc6 version. Can I update the libc6 packages by installing them directly (separately downloaded packages) or do I get the dependency hell that I want to avoid?
View 4 Replies View RelatedIssue: I want to backup my system and be able to restore if/when my system crashes.
Setup: Thinkpad T61 running Sidux KDE3.
I want to back up the system files more than personal data. What I have found is that when I apt-get dist-upgrade the upgrade performs without error, but when I restart and login I get a black screen with a mouse. From the 'black screen' I can open yakuake with F12 and then launch applications and such from the command line. The point of this more troubleshooting right now. Later is so I won't need to do a complete reinstall.
What backup prog have you used that will backup the complete system and is easy to restore?
I'm having another issue when trying to install a particular software package.
I'm trying to install a CityEngine 2015.2 trial and when I run the setup script using ./Setup, I get the following error:
Code: Select all[CityEngine 2015.2 Setup Error]
libc6 not detected
I tried reinstalling libc6 to no avail and searching the net for answers but couldn't find any solutions.
I parsed the script for that exception and found the following code:
Code: Select allelif [ "x` which dpkg-query 2>/dev/null `x" != "xx" ]
    then
      DetectedLibcVersion=` dpkg-query -s libc6 2>/dev/null | grep ^Provides: | cut -f2 -d'-' `
      if [ "x${DetectedLibcVersion}" = "x" ]
      then
        error_header
        echo "libc6 not detected" >&2
        exit 1
      fi
I am using my athlon64 machine running Debian squeeze i386 to build some packages optimized for a K6-2 machine. I decided to start with libc6 since most packages depend on it.
I installed the build dependencies via
Code:
apt-get build-dep eglibc
Then, as an unprivileged user, I downloaded the debianized source via
Code:
apt-get source eglibc
I changed to the source directory and ran
Code:
dpkg-buildpackage -rfakeroot -us -us
About 6 hours later, the build failed: [URL]
There were four config.log files:
Code:
./build-tree/i386-xen/config.log
./build-tree/i386-i686/config.log
./build-tree/i386-libc/config.log
./build-tree/i386-amd64/config.log
All of which end with exit status 0. They are also about 1600 lines each so I haven't read them entirely. If someone can tell me which one/ones is/are relevant, I will post it/them to a pastebin.
I'm setting up a Backup & Media server, which will be running debian. I will setup a small HD or SSD/CF card for the OS, and a MD raid for the data drives.The total size of the raid will be either 3 or 4TB, depending. Now, I need to figure out what filesystem to use on top of this raid.My criteria is as follows:
1. Support for large files. I can't imagine anything larger than about 1.2TB, but the 4GB of, say, fat32 just isn't enough.
2. Robust. I don't want it falling apart on me; nothing too unstable.
3. (and this is most important): Good Undelete support. I got burned recently when a software glitch managed to rm -rf my EXT4 drive. All the file data is still there, but all the metadata is gone. I *DO NOT* want that happening with this. I want to be able to do a "RM / -RF", immediately unmount it, and then recover *all* of the deleted data. Obviously, when data is overwritten it's overwritten, but I don't want to lose all my metadata if a "RM -RF" happens. FAT-32 is the model I'm looking at: You can usually recover deleted files if anything happens to the drive.
So, what are my options?EXT2 looks like a possibility. EX4 is *clear out*, unless there's some nice utility/mode that keeps a backup of all deleted metadata etc.
I would like to create a full systembackup to a ISO/IMG-file. I've been searching and found mondorescue.org, but something is wrong with package for debian 6.
View 14 Replies View RelatedI installed a fresh version of squeeze, with kde, and it worked fine. After I untarred my backup /home and /etc, things came to a standstill. I got a final ERROR code after /home. After a restart I got a 'fix your system' warning requiring a Control-D, which I couldn't type. I had used tar -xzvpPf ...... If I had included a -w, which asks for confirmation before overwriting data, could I have prevented this problem.
View 1 Replies View RelatedI have an upstream source tarball which, by coincidence, already contains a directory named "debian", but which has nothing to do with the "debian" directory for packaging. Assume also that renaming this directory would be difficult, for example because many other parts of the code would need to be modified. On an rpm based system, this poses no problem to the packager, since the rpm build process is driven by an external spec file and never has to modify the original source. What would be the best way to package this on a debian system? Is it absolutely necessary to rename the original "debian" directory? Or can I make the dpkg build tools operate out of a different directory?
View 5 Replies View RelatedUpon attempting to uncompress a tarball with
Code:
I just tried installing another version of libc6 (2.10.2-6) with kpackage. It didn't work so I had a broken package on my system. I then wanted to fix it with synaptic package manager which then told me in order to remove it, I had to pretty much uninstall all of linux. I then tried installing the older version of libc6 (the one I'm supposed to have: 2.7-18lenny2). That gave me the error code
dpkg -i '///home/sam/torrents/libc6_2.7-18lenny2_i386.deb' ;echo RESULT=$?
dpkg: `ldconfig' not found on PATH.
dpkg: 1 expected program(s) not found on PATH.
NB: root's PATH should usually contain /usr/local/sbin, /usr/sbin and /sbin.
RESULT=2
I already checked my path and it was: /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin (I found it with echo $PATH)
I guess my path is okay so the problem is in ldconfig but how can I get ldconfig in the path and where could it be now?
Is there any other site where I can download the Bluecurve Debian Tarball? When I try to download from the Freshmeat site it gives me a 404 not found. I search the site, and nothing. Can Anyone Help Please?
View 5 Replies View RelatedI think I need to upgrade my libfaac libraries due to some problems I am having with ffmpeg (the libraries are in the package 'libfaac0'). However, it won't install it because my version of libc6 is really out of date, supposedly (but it is actually the most up-to-date version).Do I need to manually upgrade libc6 or is something else messed up? I don't want to do anything that would put me at risk of screwing up a lot of dependencies.
View 1 Replies View RelatedThis script simply deletes files older than a certain age (in this case 7 days) from a certain location; I use it to purge old backups nightly, and it works as expected:
# delete backups older than 7 days
find /mnt/backup/* -mtime +7 -exec rm -Rf {} ;
The problem is, every morning I get an email with an error message something like this:
find: `/mnt/backup/subfolder': No such file or directory
I am now preparing myself to upgrade lenny to squeeze and decided to do a backup on my system. I used backup-manager to do the job and it worked fine. how do you restore said backup data?
View 4 Replies View RelatedI am trying to restore my system to Ubuntu 10.10, using a system backup made with REMASTERSYS. When I reboot, I get the message: GRUB error:15 I found many threads discussing this issue, most notably here: [URL]
[Code]....
At work we have a linux based ip phone system. Now this system was buggy for a bit but we now have it working perfectly. As it is we clone this hard drive and on backup I clone it back. This works but if for whatever I'm not in the office the backup can't go through because apparently no one can read the VERY simple instructions. What i would like to do is burn a cd with a copy of the hard drive (this configuration will change once every 6-12 months at maximum) as well as a VERY minimal linux install.
The goal is they stick the cd in and reboot. The system loads the super minimal linux then it runs a batch file that clones the portion of the CD that has the hard drive onto the system then it asks the nice moron....err....person...to remove the cd and reboot All I need is a a bootable system to a bash script that will run live from cd. I can make the nice bash script that puts to the screen a nice ascii graphic telling the lucky sap it's running and to come back in a little bit to make sure it's done.
Recently I've backupped my entire /home folder on my laptop with grsynch trough ssh.Since I already had a simply text file containing all installed programs, I figured 'then, when i go from the backup trough my desktop, all my program settings will be restored if i run the txt file to install all those programs first.So i installed al programs, then i went from backup to /home on desktop, logged out and back in. Started up a random program (tried thunderbird and filezilla) and no settings were to be found.In retrospect, docky did start, but didnt have all the launchers i had on my laptop. so that could have been a clue.
View 4 Replies View RelatedI have added a new sata HDD and this caused my fstab to becoming corrupted. I opened the fstab within a live cd and I changed it to the new setup. But I am still getting an error.
My old setup:
sda on PATA
Windows 7
Linux /root
Linux swap
[Code].....
I want to install Ubuntu 10.04 on a second HDD than my windows xp installation. The second drive is on the same ide ribbon as the primary HDD (xp) and is set to slave. When i installed ubuntu on the second HDD it corrupted my primary drives MBR. It was an easy fix but every attempt i make ends in the same result. I also have another HDD that is sata and the sata controller is a via pci expansion card. Installation of ubuntu to this drive as well also corrupts my primary HDD MBR. But when I install to the same HDD the dual boot works fine.
I was thinking of unplugging the power cable from the primary HDD, then installing ubuntu to one of the separate HDDs. Would i still be able to dual boot. And which drive would be optimal for a dual boot installation in this configuration.
This problem started after I moved from ubuntu 9.10 (64bit) dual core to 10.04 (64bit) quad core machine. On both machines, I have svn working directories in an NTFS partition since I dual boot between Vista and Ubuntu - which gives me the opportunity to work on my projects whether I am on vista or ubuntu. My fstab entry is very simple:
UUID=62B81CE6B81CBA8D /media/ntfsdata ntfs-3g defaults 0 0
I use either RapidSVN or RabbitVCS. Whenever I commit those folders to the SVN server(google code), the folders seem to get corrupted. After which a rebuild or a refresh on Eclipse would provoke a 100% use on one of my 4 cpus. It would just hang there and the ntfs partition would not be accessible at all until I killed ntfs-3g process.
Then I would have to boot on vista to repair that partition. The gparted repair utility is USELESS because it cannot spot any partition errors caused by the svn commit. The dual cpu machine had occasional similar trouble but I never needed to repair it. SVN would not kill the ntfs-3g, but eclipse rebuild project after an svn operation would. I could prevent ntfs-3g hangup by checking off build automatically after a svn operation and close all projects and reopen one at a time to refresh.
A refresh would trigger a rebuild. On the phenomenally faster quadcore machine, the speed of operations could be too quick for ntfs-3g to handle (may be??). It looks like ntfs-3g is unable to handle high number of files at high speed operations. I use the same ntfs partition to store high volume bulk copy and bulk downloads with no problems - apparently, it is not the problem with high volume of transfer but the high number of files being thrashed back and forth at high speed by eclipse and svn commit. Does anyone else store their svn working directories on ntfs when developing with Eclipse on ubuntu 10.04?
Further addition: I need to mention that partition is compressed under ntfs. Perhaps, ntfs-3g is not capable of handling compressed ntfs. I uncompressed them - and during uncompress operation, explorer hung at the folder where I had done the svn commit. I rebooted vista and made a fresh uncompressed copy of those svn working folders and deleted the old copy. I'll wait and see what happens on my next commit using rabbitcvs.
I'm using 10.04.2, and I find that whenever I'm trying update (whether using synaptic, or command-line apt-get), the package download starts off fine, but after downloading a few files (usually only 4 or 5), the next partial (ongoing) downloads become corrupt, all the remaining downloads stay in the partial folder and finally apt-get gives a size mismatch error .I'm forced to watch the update progress the entire time and wait for the downloads to corrupt (at this point the progress bar stops at say, "Downloading file 4 of 70" but the details show subsequent files are being downloaded)the update process and clean the /var/cache/apt/archives/partial folder (NOT the archives folder, or all the packages will have to be downloaded again). Then I restart the update and it picks up after the last successfully downloaded .deb.
Effectively, every such iteration downloads 3-5 .deb files successfully, and corrupts the remaining. Needless to say, if I'm upgrading 50 packages, it gets really frustratingI faced this problem even on new installations on my system as well as a friend's. Lucid, Maverik, Natty all have the same problem. By new installations, I mean on the very first boot, I setup the network (I'm behind a proxy server in a university) and that's it. I tried Linux Mint and it had the same problem.
After a fresh boot the first few songs I play with Banshee or Rhythmbox (only used for testing if Banshee was the issue) the songs start to be played badly distored. Like the sound is only being played in very short bursts with gaps between them.
If I restart Banshee the music will usually resuming playing fine for a couple of songs and then the issue repeats. If system load is high the number of songs before the issue appears decreases (as low as 0).
I've attached the output of the alsa-info.sh as I see 90+% of audio issue threads ask for that file in the first reply
I convinced my sister into trying linux instead of windows. When she agreed, I messed things up for her in the install and now her files are gone. I was trying to install Kubuntu. I had backed up all of the data into one partition (hda7 ... which is the last partition) and restructured the other partitions. When applying changes, it crashed in the middle so I had to start again and redo things.
Finally when everything was installed, I tried mounting the NTFS hda7 partition but to no success. Upon checking in fdisk, it was listed as ext3 but even could not mount it with ext3. I checked in windows and the partition is marked as "unknown". what has changed is that instead of hda7 being logical, it is now marked as primary. I have checked testdisk and I am not able to recover any files from it. Any ideas?
1. Testdisk ... did not find any files from that partition
2. Partition magic ... got some messages that something wrong with the partition table, it fixed that, but get Error 17 and partition magic cannot start
3. chkdsk ... did not find anything
I read something about DD but I have never tried it and I am worried I might mess up things.
I switchted my windows friend over to kubuntu and I installed the kubuntu-restricted-extras as well as the ubuntu-restricted-extras and at first amarok was flawless. Now, every time I quit amarok and reopen it, my media library is still there but when I try and play anything it just stops and does nothing.
Then I go to my /.kde/share/apps/amarok directory and clear it and it works again if I open it and reimport the music, but this is a bitch to do every time. I briefly tried rhythmbox also and it seemed to be suffering from the same problem although I didn't try deleting the preferences and testing if it worked after, it also refused to work after quitting. Note: it might get corrupted on reboot only or on quitting
i've tried both the gnome and kde installers and initially it boots fine to the first menu, but if i then select 'installation' or even try to boot to the live cd, the screen goes blank and then reappears as something that looks like the static on your television if you're trying to tune it in. if you let it continue to boot, then it goes blank again and comes back with a different corrupted screen, but this time you have a white block in the middle that is apparently the mouse point as i can then move it around.
i've tried it in VESA mode and even text mode and the same thing happens. the installation media checks out fine, and my graphics card is brand new - a Radeon 3650 HD. I might try an older version of suse and try and upgrade from there
I have installed Open Suse 11.1 on two different machines. One contains an ATI X850 video card, and the other contains an ATI X1950 video card. I updated the ATI drivers on the 850 card and it works like a dream. I updated the ATI drivers on the 1950 card and Suse becomes unusable. System works fine before the update. Before the update I ran it at a resolution of 1280 x 1024. I thought I would lower it to 1024 x 768 to see if that would help, but it does not. I dual boot with WinXP, and card works fine with that setup at 1280 x 1024, so I would suspect this is a specific problem with my Suse setup as opposed to a Motherboard issue.
View 4 Replies View Related