When I compile a custom kernel with this command: make-kpkg --initrd kernel_image kernel_headers and then install the .deb, there's no initrd in /boot and I have to create it manually. I've thought that the --initrd option should take care about this, but somehow it doesn't.
It behaves like this for about two years at least (since I've compiled my first kernel). Of course, it's no big deal to create it manually, I was just wondering whether do I do anything wrong or whether should I fill a bug report..
I have compiled the linux kernel. My bzImage is 1,14 mb big
Anyway. It's only 2 programs i need and it's Python and Busybox. I have compiled python and busybox and put them together in a folder. Busybox is 146 kb and pyton is 4,4 mb.
But how do i make an initrd file of them? Or should a make an initrd file of them?
I get this error when a runt with a compiled busybox to gz format as initrd, and without initrd.
Quote:
VFS: Cannot open root device "sda1" or unknow-block(0,0) Please append a correct "root=" boot option; here are the available partions: Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)Pid: 1, comm: swapper Not tainted 2.6.38.3 #1
But i would like to move to cpio, because with dd, if you add something new, you might need to change the count. Also cpio is used in distro's like Fedora and Ubuntu.
I need to install Ubuntu on approximately 50-60 netbooks. None of them have CD drives, and I don't want to have to install them individually, walking around with a USB stick. I figured the fastest way to install on so many machines is to use a combination of apt-cacher (http://www.debuntu.org/how-to-set-up...ith-apt-cacher) and netbooting. I have successfully booted one machine to test, but as soon as the kernel comes up, support for the network interface is gone. Specifically, the "atl1c" module is not included on the netboot initrd image. Also, I would like to try to use preseeding, and I need to get that onto the initrd as well.
So, to summarize my question: How can I create a custom install kernel and initrd? I have a feeling it's related to the "debian-installer" category in the package repository, but I have not found any good documentation about doing this.
I am attempting to PXE boot Redhat 5.4 and load the OS into RAM. Here is how far I have gotten so far:
I have successfully configured my DHCP, TFTP, and PXE servers. The PC that is booting up Redhat 5.4 is able to get a DHCP IP address from the server, grab the kernel and initrd from the TFTP server, and boot up from there. I have this setup working if I store the root file system on a NFS share on the server and use the kernel's NFSROOT parameter. Long story short, I now need to get this same setup working without using NFS if possible. I have spent several hours google searching how to create a RAM disk (initrd) that also contains the root file system. I must be searching the wrong keywords. Anyway, based on what I have read so far, I need to do the following:
1. I need to pass different kernel parameters. It needs to look something like:
2. It seems like newer versions of initrd are created using the cpio tool. So would this simply be a matter of booting to my hard drive that has Redhat 5.4 loaded on it and running:
To create the RAM disk with a root file system attached to it? I have a feeling it has to be more complicated than that but I just cannot find any sites that specifically explain what would be involved with creating an initrd file that can also be used as the root file system.
3. Would I need to modify the init script (many sites call it "linuxrc") at all in order to accomplish what I want to do? For example, I found this site:[url]
My understanding of this page is that you have to create a linuxrc script that basically does the work of decompressing the file system into /dev/ram0 and mounting it to /root. Is something like this actually needed?
4. The kernel must have certain parameters set such as enabling RAM disk support.
I have most of this setup working. I just need to figure out how to basically store the contents of the OS on the TFTP server and then tell the kernel to load that OS into memory. Based on what I have read, this should be possible (otherwise, how do Live CD's work?).
I install fedora on my usb-hdd. Sometimes i boot it from real machine, sometimes i boot from virtual machine (kvm). When I boot from real machine, hdd driver is usb-storage, when i boot from virtual machine hdd drivers are ata_generic and pata_acpi.
Everytime i update kernel, mkinitrd only create initrd image contains only hdd driver current running, and I must recreate initrd manual with all preload hdd driver. How to config fedora to everytime update kernel, mkinitrd can make initrd image with all hdd driver I need?
I have an encrypted root and encrypted boot drive... To avoid entering a passphrase multiple times I'd like to add a keyfile for my root drive to my initrd... I've done this in Arch and it was relatively simple but I've hit a road block in debian finding the analog processes...
Code: Select all#Part 1) Generate the keyfile, give it suitable permissions and add it as a LUKS key:
i am using linux for couple of years. suddenly i a question arises in my mind. Before loading kernel during booting linux box initrd image loads necessary real mount point, file system, modules etc. but how initrd loads them ?? is there predefined modules list stored in initrd image or something else....what is the background procedure of it
These kernel sources are usually the sources from debian, with a couple of more patches that I add. It appears that for nvidia-driver package versions higher than 352.79-1, the kernel headers/sources need to be prepared with 'make prepare' and 'make prepare scripts'. It's that simple. I concluded this after the nvidia dkms build failed on my custom kernel, but then succeeded after I pointed it to the full sources, but only after running 'make prepare' and 'make prepare scripts' on them. The problem is that this make-kpkg scheme doesn't appear to do this, or if it does, it doesn't properly include in the headers everything that it should.
I have encountered a bug in the live-build in Debian Live. The persistence does not work it is because of the bug #767195. I cannot recreate another live img file because of bandwidth problem, is there any way to repack the initrd and enable the cryptsetup?
as per suggested: #767195 – cryptsetup needs to be enabled for initramfs inclusion – Debian Bug report logs On Wed, Oct 29, 2014 @ 18:09, Evgeni Golov <evgeni@grml.org> wrote: .... Edit: Line 77 -79 @ # nano /usr/lib/live/build/chroot_hacks
When I installed Debian stable on a headless machine of mine, I configured a partition with LUKS encryption (intended for swap), but told the installer not to use it. After installation, I configured that encrypted partition as swap and mounted it. I wanted my headless machine to boot all the way without manual intervention, so I can log in via ssh and mount my encrypted partitions.However, since the kernel was updated (and the initrd regenerated), the machine now waits during boot for me to enter the swap encryption password, but no others, only the swap.I have been unable to find in my searchings how the initrd is generated with that setting or how I can change it, preferably permanently so future regenerated initrd's don't try activating my encrypted swap on boot. Does anyone know how to configure a Debian style initrd generator to generate an initrd that will not try to activate swap?
So I'm building a custom kernel cuz I want the fbcondecor patch in my kernel. I use the same .config that 2.6.32 debian kernel package comes with. This kernel runs perfectly. I pass --initrd to make-kpkg when building the package but no initrd is built when I install it so I have to make it using "mkinitramfs -o /boot/initrd.img-2.6.33 2.6.33" to get it to boot. Now when I try to build the nvidia drivers it complains it can't find the source. I did build kernel_headers and installed them also the source is in /usr/src/linux. I also tried to specify the path by passing --kernel-source-path= to the nvidia script but no change. What is going on? I've done this fifty times before and never had any problems. Has there been some changes to how debian kernel packages are built? EDIT: Just thought I'd add some info about the steps I took.
Code: tar xjvf linux-2.6.33.tar.bz2 ln -s linux-2.6.33 linux cd linux patch -p1 < ../fbcondecor-0.9.6-2.6.33-rc7.patch cp /boot/config-2.6.32-trunk-686 ./.config
make menuconfig Loaded .config then I removed support for maxtorfb, tile blitting and some sirrusfb thing, nothing thats relevant to my system. Changed cpu from Pentium 4 to Core 2 and added framebuffer decor support from my patch, exited and saved. Then:
[Code]....
EDIT2: I have now tried to build 2.6.32.8 in the same way with the same strange results, anyone have any thoughts as to what I'm doing wrong just throw it at me, I'm getting desperate and running out of ideas. I've checked all the kernel source symlinks and everything looks good.
I'm running Debian testing and after a recent software update, suddenly I can no longer boot into my single Linux kernel image anymore. I did look at the software update before I did it and I don't believe that it removed any software packages. When I try to boot into my linux 2.6.26-1-amd64 image, I get the following message from grub:"error: unknown command `initrd'"..The same error appears if I try to boot into single-user mode. I have a Windows XP partition on the same hard disk and I am still able to boot into that from grub. I've searched the net and haven't really found my same problem replicated anywhere. I would imagine that the problem is just that somehow the initrd package got removed in my last software update, but again I don't recall seeing any software that the update planned to remove.
Any thoughts on how I would go about fixing this problem? Since I'm unable to get into Linux at all, I imagine I'll have to use a LiveCD to get in and fix the problem, but I don't know where to begin looking.
The RHEL 4 update 7 installation CD installs the kernel with version 2.6.9-78.EL and therefore must contain corresponding vmlinuz and initrd.img. However I would like to have an all modules initrd that is used during installation but for a different version (2.6.9-78.0.13.EL). Is there a way in which I could obtain/make such an initrd?
I have installed vmlinuz and initrd.gz (squeeze) in /boot/newinstallation and added to Grub the lines:titleNova Instalaçãoroot(hd0,0)kernel/boot/newinstall/vmlinuzinitrd/newinstall/initrd.gzNevertheless, when I choose this option at the grub, it begins to start the kernel vmlinuz and thus restart the system. The installed Debian Lenny boots in a normal way.
I've had a Centos 5.3 zimbra mailserver set up for about 3 months now.Every 7-10 days or so, the system will do a kernel panic and I've been putting off its resolution due to other matters.I'm now ready to look into it but I have two issues :
1) No logs.
So far, I've manually rebooted when a kernel panic occured and could look at the panic printout and write down relevant info. But since I wanted more info than I had written down, I went hunting in the logs for the kernel panic event (one happened earlier today) but, to my surprise, I noticed the system doesn't log the kernel panic at all. Kdump is currently disabled as I'd like to keep all available memory available to the system but I was under the impression it would log the kernel panic message regardless but would simply leave out more advanced diagnostic information.How would I make sure the kernel panic is logged so I can have more info about it ?
2) The actual panic.Here's a snippet of what's appearing when the kernel panics :
I've also seen f793bee4 instead of the highlighted part above but everything else is always the same from the times I've seen (as I haven't always had the chance to be around when a panic occured).I'm looking further into it at this point but early research seem to indicate this could be a memory issue. I'll run memtest next weekend to check that out (since I don't want to take the system offline for any length of time during office hours) but people here might have other theories based on what I've put up so far. The system has all of the latest patches and updates and is running only what I need as far as software goes (zimbra server + prerequisites, x + GNOME (not loaded on startup), firefox and vnc for windows RDP access). My system also uses a software RAID1 array (boot = md0 and / = md2 with swap files not being mirrored as there's no need to).
PS : In order to further troublehsoot by myself before posting here, I've tried to enable kdump to see if kdump is the culprit for the missing logs but I'm also having an issue with it. I'm getting a "No kdump initial ramdisk found" and "mkdumprd: failed to make kdump initrd" messages at startup. I'm thinking this might be because of my raid array ? I've found an old bug report regarding one of those error messages on centos 5.2 but it was supposedly fixed in the 30 version by RH and I'm using the latest one (56).
I heard somewhere that by creating custom initrd image and default compiled kernel image, we can do PXE linux installation. can anyone please guide me 'what content will be placed inside initrd?'I know the process of creating custom initrd file .
I do not find any initramfs-2.6.33.5-112.fc13.i686.PAE.img in my /boot folder after I updated Fedora 13 after a fresh installation. The vmlinuz file is present in the /boot folder.he grub.conf file does not show any initrd entry too.
I'm running FC 13. I believe that the base install is kernel 2.6.33.6-147.x86_64. That is the last version of the kernel that I have installed that includes an initramfs file. Neither 2.6.34.7-56 nor 2.6.34.8-68 installed an initramfs file. Since my root volume is in LVM, I can't boot with those kernels. I saw in another post that doing a "yum install kernel" would install the missing files. I tried that and it tells me that kernel-2.6.34.8-68.fc13.x86_64 is already installed and latest version.
I ultimately want to upgrade to FC 15 because I've got a new nVidia video card and need driver 270.41.06 for fc15. My fear is that if I upgrade I might not get the initramfs files in the new versions of FC and end up with an unusable system. Is there a way to add the initramfs files for the newer kernels and/or will I have the files after an upgrade?
i upgraded from 11.1 to 11.2. Unfortunately the new kernel does not work with my sata/southbridge (i googled and figured its a known issue). So i tried to boot a 11.1 64bit rescue system to install the older kernel and the corresponding initrd. The problem is, that i cant find the kernel and initrd on the rescue system.
i am using rhel5 and i just wanted to see the contents of the initrd image. I have copied that from /boot/grub and saved in /tmp directory. Then i did the following
when i gave that, it says "mount: you must specify the filesystem type" Then i included "-t ext2" option in the mount command. Now it says,
mount: wrong fs type, bad option, bad superblock on /dev/loop0, missing codepage or other error In some cases useful info is found in syslog - try dmesg | tail or so
Then i created a ext2 filesystem and did the same thing. But there is no difference.
I've always booted Slackware with an initrd until recently. With my new installation of Slack Current (13.37), I didn't bother to create an initrd because of the constantly updating/transient nature of the current branch at the moment.My question is this: when 13.37 does go final in a few days, should I go back to my normal practice of booting with an initrd? Also, I think I understand how the initial ramdisk works, but is it really needed with the more modern kernels?
Why use nash instead of busybox in initrd and initramfs? I'm just looking for pros and cons of both really (and any other applications with similar functionality). I'm currently leaning towards busybox being the better option, why redhat and fedora use nash in their initrd.
I debootstrapped an install into an existing lvm and booted into it and everything is working great except that after initrd hands off to boot the real root, the text font changes and the resolution is unsupported by my old monitor. The box is up and running because I can ssh in. This is more of an annoyance than anything as I only use the console when something is broken, but it does need to be resolved.
I used dpkg-reconfigure console-setup as described in the debootstrap config guide but I don't see an option for changing the "vga" statement before it regenerates the initrd.
