Debian Installation :: Wheezy - Installer Failed On Grub
Nov 23, 2014
I am installing Whezzy and the installer failed to install grub on the MBR of /dev/sda
My disks are
- /dev/sda : the flash drive with the debian ISO
- /dev/sbd : a small SSD for the / partition (sdb1)
- /dev/sdc + /dev/sdd : a software RAID-1 array with LVM for /home, ...
Grub fails to install on /dev/sda which makes sense since this is the flash drive containing the Debian ISO so no MBR.
I already tried to run grub-update manually on /dev/sdb (with chroot /target grub-install --no-floppy --force "/dev/sdb" ).
It works but the system is not directly bootable. I had access to a second PC to read the grub documentation so was able to boot and to fix my system but this is annoying.
Is it normal to see my flash drive as /dev/sda? Could it be that my flash drive is incorrectly detected as a HD?
Is there a way to force the installer to install grub on /dev/sdb?
I am using Wheezy netinstall on a USB stick. The machine has an Intel DH57JG motherboard, USB keyboard, and no optical drive. The installer loads to the initial screen OK, but the keyboard is not active. This problem has been reported for several years and was supposed to be fixed by kernel 2.6.32, yet here it is. What can I do?
One suggestion has been to disable/enable USB legacy support in the BIOS. It is enabled by default. If I disable it, the installer doesn't load. This keyboard works fine on other Linux machines and on the BIOS in this machine. I read there was a problem with Cherry keyboards and the Squeeze installer. This keyboard is not a Cherry but does have Cherry keyswitches.
I'm trying to install Wheezy (amd64) from the daily netinst image with the text-based installer. The PC will lock up randomly at some point in the process (usually after about 5-10 minutes). The video signal cuts out, keyboard lights are unresponsive, and I have to hard-reset the machine.I was previously trying to install Linux Mint Debian Edition and had the same problem. Another user reports that this is due to a bug in the nouveau nVidia driver, but I can't confirm it.URL...
I assume the Debian text installer isn't using nouveau, so it seems unlikely that would be the problem. The random timing does suggest an overheating problem, though, and the fact that the video signal drops out would seem to implicate the video card. I had no lockups at all with the OS that was previously installed (Ubuntu 11.04).
I upgraded from deb7 to deb8, but am no longer able to boot. After passing the grub boot menu, the following messages are displayed:
Code: Select allLoading, please wait... [ 6.065713] systemd-fsck[148]: /dev/sda1: clean, 428644/1310720 files, 410616 9/52442880 blocks [ 7.480551] Error: Driver 'pcspkr' is already registered, aborting... [ 8.692700] systemd-fsck[341]: /dev/sda5: clean, 145485/6176768 files, 17407409/24695552 blocks [ 18.066215] Loading kernel module for a network device with CAP_SYS_MODULE (deprecated). Use CAP_NET_ADMIN and alias netdev- instead _
The screen then clears and an underscore is displayed as the sole character at the top left position of the screen. The system hangs at this point. During installation, I rejected two changed files: /etc/init.d/bootlogd and /etc/libreoffice/sofficerc. For both, I opted to keep the installed version (the default choice of the installer) rather than replacing with the new version. The first might be related to the problem, although it seems to be responsible only for logging the boot process, and I would not expect this to compromise booting.
In case this information is useful, sda1 is mounted at /, sda2 is swap space, sda3 is extended, and sda5 is a logical partition mounted at /home.
I am able to boot into rescue mode, but other than that the system is not usable. Unfortunately, no useful error messages are given to aid in diagnosing the problem.
I've changed my /etc/apt/sources.lst file to use "jessie" repositories instead of "wheezy". I then ran synaptic and updated everything (there were loads of packages, something like 2000 to update).
After this I rebooted. The grub menu shows as usual with the background image I'd set and the operating systems as usual (including Windows 7) however there is no longer a 5 second countdown and when I select *any* menu option, it asks for a username and password.
I don't know what username and password it's asking for as I never used to have one set!!! I did have a username and password set up so that if you wanted to edit a grub menu option so I tried that but to no avail.
I'm trying to install stable Squeeze from HD "!! Scan hard drives for an installer ISO image. The quick scan for installer ISO images, which look only in common places, did not find an installer ISO image, but it may take a long time. Do full search for installer iSO image? [YEs} No kernel modules were found. This probably is due to a mismatch between the kernel Version available in the archive If you're installing from a mirror, you can work around this problem by choosing to install a different version of Debian. The install will probably fail to work if you continue without kernel modules"
I have a VIA Epia M 5000 system with 2 western digital 1TB NAS SATA drives connected through SATA<->IDE adapter. Everything installs and writes as expected except... grub. It never boots, after a message 'Grub loading' I always get 'error: no such disk'. I've tried numerous times and has been attemping to fix the issue for the past 2 days.
/dev/sda 0.999TB 0xfd linux raid autodetect partition 1GB 0xfd linux raid autodetect, logical partition
/dev/sdb exactly the same 0.999TB 0xfd linux raid autodetect partition 1GB 0xfd linux raid autodetect
/dev/md0 RAID1 of /dev/sda1 and /dev/sdb1 marked as ext4, boot point /
/dev/md1 RAID1 of /dev/sda5 and /dev/sdb5 marked as swap
update-grub2 in rescue mode generates grub.cfg with SET root=/mduuid/UUID_OF_SDA1 then after that there's search --no-floppy etc --set root=/mduuid/UUID_OF_MD0
I'm writing this from memory but simply the two uuids are different. Is this correct? I get those UUIDs to compare from blkid. All partitions are marked as bootable. grub-install /dev/sda and grub-install /dev/sdb produces no errors. grub-install /dev/md0 does not work, complains about superblocks or something similar.
Grub.cfg file contains insmod raid mdraid1x and similar lines, so that should be ok. Grub drops to rescue mode with message error: no such disk. Not device, but disk. Google finds many results for 'no such device' error, but I am not getting that error. 'ls' produces (hd0) (hd0,msdos1) (hd1) (hd1,msdos1) ls ANY_VALID_PATH produces empty newline being printed, nothing more.
setting prefixes manually does not work, with error message 'error: file not found'. ls (hd0,msdos1)/boot/grub also produces empty line being printed. Rescue CD and auto-assemble of md0 and md1 works, the files are there, everything okay, except grub.
I try to install linux on hpc mini. Let's crack some keyboards. I have just bought a new pc: hpc mini compaq! I find windows Xp rather fine; but let's try linux. The PC has no cdrom, so, I cannot boot from any ISO of debian. I choosed testing netinstall, and followed the how to given on the debian website from another PC using knoppix iso cdrom. I did placed five files as shown here onto my pendrive as followed in one howto: I found 5 files for testing:
Code: debian-testing-i386-businesscard.iso initrd.gz vmlinuz and Code: 26644480 Aug 28 17:44 ../squeeze-debian-testing-i386-businesscard.iso with : [Code]....
I was installing sqeeze i386 on my laptop VOSTRO 1400 and got this the 'grub-pc' package failed to install into /target/. without the GRUB boot loader, the installed system will not boot.
Im trying to do a frugal install off the hard drive (no usb,cdrom) with unetbootin. It installs grub and then will boot the installer but is still tries to find the removable media. Whats going on????
I'am using Ubuntu 10.10, and wish to install a Debian OS.I downloaded the file "debian-6.0.1a-amd64-netinst.iso.(My motherboard is a 945CG-F7 with INTEL PENTIUM DUAL CPU E2160)I recorded the file on CD as "iso" (following instructions from Internet).Grub tried to read the CD but than started Ubuntu OS.
Toward the end of installing Ubuntu 10.10 32bit (Alt CD) on my iMac 11,1, the installer asked me to type in the location for installing the grub boot loader.
I told it to use /dev/sda3 and it immediately failed. I'm still in the installer. Can anyone suggest a solution?
Here are my partitions on sda:
...from the shell in the installer, there is no grub.cfg in /target/boot/grub.
I installed lilo and it boots. How do I installed grub again. Do I just use synaptic manager to uninstall lilo and install grub?
[code]....
I know grub2 does not work, giving me error: ntoskrnl.exe missing or corrupt. I did install grub-legacy and it worked, but I had to re-install squeeze due to other problems. So, now I want to replace lilo with grub legacy.
HDD with four partitions: Three DOS bootable primary partitions are located in the head, and the residual extended partition is divided into several logical drives.
1st trial: 10.04 installer recognized the last largest logical drive for system installation, but installer truncated the extended partition, and created the "terrible" 5th primary partition at the end of the HDD. GParted and other utilities cannot access to this 5th primary partition. (So, I restored the lost partition table on the HDD manually by MBM. But several OS were broken.)
2nd trial: To avoid making this "terrible" 5th primary partition, I located the largest logical drive for installation at not of the end.10.04 installer recognized this logical drive, but failed again.I tried GParted. but also failed to formating to ext4.Maybe, 10.04 installer failed at this formating step. Logical drives are not supported in 10.04 ?
I recently brought home an old Compaq server from work. I was told that it works fine. When I brought it home, I tried installing Ubuntu 9.10 server from a cd. While it's trying to install, I get the message "Installation step failed. An installation step failed. You can try to run the following item again from the menu, or skip it and choose something else. The failing step is: Load installer components from CD." I turned off the computer and tried it again several times. Each time the same message appears.
I know this is not strictly a Ubuntu question, but I was unsure where to ask this question and since it is about installing Ubuntu Server, I put this question here.
I have used the Universal USB installer to create an install USB device, but I also need to be able to boot to a DOS prompt so I can update the bios if I need to. I would like to have just a single USB drive rather than having to carry two of them around.
Has anybody used the Universal USB installer and is there a way to modify the resulting USB installer so that I could say add a menu program such as grub or something else that would allow me to either access the installer or go to a DOS prompt? Kind of like the way CloneZilla does or perhaps the gparted Live CD.
Just re-installed Wheezy to ssd-drive and noticed that everything works much faster than before. For the older install I made some changes to /etc/fstab (I think) which I found from the web. I don't remember exactly what changes.
Question is: what's the situatation with Wheezys beta4-release and ssd-drive? Do I have to edit something at the moment, or does the installer notice ssd-drive and make everything needed?
I searched the forum about installing the liquorix kernel but I cannot understand why the file /etc/apt/sources.list is not mentioned to invoke the liquorix repository.
Instead, I found: Alternative Performance Kernel for Debian wrote:For anyone interested, just add the following to a sources file (/etc/apt/sources.list.d/liquorix.list): deb [url] or for you scripters, echo "deb [url]
I already have a Notebook with windows (for some reason it needs 3 partitions..) and wheezy. When I installed wheezy I create 2 logical partition inside a primary one: one logical for /home and the other one for filesystem. Now I would like to overwrite wheezy and get get jessie, but without touching windows and home.
I already try to upgrading and it went wrong, so I prefer overwrite, what I have to do during "manual installation" phase?
I'm wondering I've read in some places that if people would like to move from a stable branch of Debian to the testing you can usually just replace the lines in sources.list with the testing release and update and then dist-upgrade. Is this true...and if so is it safe?
I upgraded one of my Squeeze installations to Wheezy, but after selecting it in grub, nothing is displayed on screen: no CLI and no GUI. I tried Ctrl + Alt + F[1-7] and I got nothing. My laptop is Acer Aspire 7715Z. I am attributing this to module as there seems to be disk activity, but without a screen, I cannot be certain.
I have a server in which I'm trying to install Debian into. There's no BIOS, only EFI boot. The vendor locked it down so there's no way to see it or get into it.
The Wheezy 7.8 netinst CD has EFI boot parameters and works on everything I throw it in, except this one server. Booting it up, the code doesn't see the EFI and boots into normal mode, negating every chance to install it (dozens upon dozens of failed installs).
During boot, I press the [TAB] at the Debian Linux installer menu and get the load parameters (/install.amd/vmlinuz vga=788 etc). What parameter can I add to this line to get it to run in UEFI mode so we can install it?
I have a laptop primarily used for a client work. It is running Debian Wheezy, and wish to keep it intact in case I need to do more work for client.
I would like to install Ubuntu 14.04 alongside Debian, and use Ubuntu as a bit of a play/experiment area, etc... HD has lots of space (600GB free), and as far as I can tell Grub is installed in MBR.
I did some searches, and from what I can tell, it sounds like I can just install Ubuntu from ISO file, specify how much space to use (say 400GB), and that's it. This sounds almost too easy. Once I install and restart machine, will there be selection for what Distro to boot.
Without knowing the consequences I added the Sid repository to Wheezy (installed version) in order to install some software. Only much later I discovered that this generated me some mess which does not allow now to install additional software due to library conflicts.
I tried to install some packages needed to build the PhantomJS but here is what I got:
Code: Select allReading package lists... Building dependency tree... Reading state information... build-essential is already the newest version. g++ is already the newest version. g++ set to manually installed.
[Code] ....
Is there a way to clean up the mess that Sid introduced and revert back to the Wheezy versions?
I have an old backup, so it would take me much more time to reinstall/reconfigure certain software, so I am looking to alternatives.
Is there a way to check all the packages and find all the potential conflicts and then a way to restore the original Wheezy content?
is it possible to simulate upgrading a Squeeze installation to a Wheezy or Jessie installation, on a OVH server ?I would like simulate upgrading server, and if not problems, upgrading in real time.I don't do that manipulation, and I don't do mistakes on a production server.
I try to update the packages before upgrade to wheezy,looks like when I run apt-get update,it shows error as per below:
W: GPG error: http://fosiki.com stable Release: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY 379393E0AAEE96F6 rat:~# apt-key adv --keyserver keyserver.fosiki.com --recv-keys 379393E0AAEE96F6 Executing: gpg --ignore-time-conflict --no-options --no-default-keyring --secret-keyring /etc/apt/secring.gpg --trustdb-name /etc/apt/trustdb.gpg --
I have just installed Wheezy after repartitioning and had to set up my sources list myself as (most of it) was not done for me. I have just added the main official repository,security updates and the multimedia repository. To be honest I'm not sure if any of them are working when I do an update in aptitude, I think the multimedia one might be. The CD ROM isn't even working despite the fact I have done apt-cdrom add.I am getting the following when run aptitude update:
Ign cdrom://[Debian GNU/Linux testing _Wheezy_ - Official Snapshot i386 CD Binary-1 20110328-05:16] wheezy InRelease Ign cdrom://[Debian GNU/Linux testing _Wheezy_ - Official Snapshot i386 CD Binary-1
I've created my own wheezy live usb-hdd using:# lb config -a i386 --mode debian --distribution wheezy --linux-flavours 686 -b usb-hdd --bootloader syslinux --memtest none -p xfce P: Considering defaults defined in /etc/live/build.conf P: Creating config tree # lb build The binary.img was created. Now if I try to boot from this image using qemu, I get a message "No bootable device" If I check this main partition using gparted, the flag is "lba" instead of "boot". Even doesn't work if try to change to "boot". How to solve this problem?
I'm trying to install adobe flash player on wheezy (7.4). I've downloaded the package since adobe thinks apt works in a browser (it doesn't on my laptop). Anyway, I've tried installing the adobe flash player package via gdebi and it complains binutils is not installed. fine, I go to synaptic to find binutils and synaptic cannot find it. I'm not even sure why it is not installed. How to get binutils?
Here is my sources.list:
Code: Select all# # deb cdrom:[Debian GNU/Linux 7.3.0 _Wheezy_ - Official i386 NETINST Binary-1 20131215-03:38]/ wheezy main # deb cdrom:[Debian GNU/Linux 7.3.0 _Wheezy_ - Official i386 NETINST Binary-1 20131215-03:38]/ wheezy main
# deb http://ftp.us.debian.org/debian/ wheezy main # deb-src http://ftp.us.debian.org/debian/ wheezy main
