On Sun, May 18, 2014 at 7:08 PM, Gene Cumm <gene.cumm at gmail.com> wrote:>> > On 05/16/2014 01:35 PM, Tal Lubko wrote: >> > > For some reason when I use kernel 3.7.1 and above the system crashes > during boot and I have no reason why. >> Consider a lack of init or otherwise improper initrd (cpio format, > compression, etc). Consider trying your old kernel with new initrd and/or > new kernel with old initrd to start diagnosis. >> >> --Gene > > [ 0.000000] Kernel command line: root=/dev/sda1 initrd=rootfs.ext2 > console=ttyS0,115200 BOOT_IMAGE=bzImage > > [ 0.297992] Unpacking initramfs... > [ 0.298619] Initramfs unpacking failed: junk in compressed archive > [ 0.299718] Freeing initrd memory: 5636k freedIf GPT used instead of MBR, no initrd required... Regards, Alexey
On May 18, 2014 3:12 PM, "Alexey Orishko" <alexey.orishko at gmail.com> wrote:> > On Sun, May 18, 2014 at 7:08 PM, Gene Cumm <gene.cumm at gmail.com> wrote: > >> > On 05/16/2014 01:35 PM, Tal Lubko wrote: > >> > > For some reason when I use kernel 3.7.1 and above the systemcrashes> > during boot and I have no reason why. > >> Consider a lack of init or otherwise improper initrd (cpio format, > > compression, etc). Consider trying your old kernel with new initrdand/or> > new kernel with old initrd to start diagnosis. > >> > >> --Gene > > > > [ 0.000000] Kernel command line: root=/dev/sda1 initrd=rootfs.ext2 > > console=ttyS0,115200 BOOT_IMAGE=bzImage > > > > [ 0.297992] Unpacking initramfs... > > [ 0.298619] Initramfs unpacking failed: junk in compressed archive > > [ 0.299718] Freeing initrd memory: 5636k freed > > If GPT used instead of MBR, no initrd required... > > Regards, > AlexeyGPT versus MBR makes no difference. Storage adapter and its driver location is more critical, followed by tools for decryption. [ 0.710918] scsi 0:0:0:0: Direct-Access ATA QEMU HARDDISK 1.4. PQ: 0 ANSI: 5 [ 0.712510] sd 0:0:0:0: [sda] 40960 512-byte logical blocks: (20.9 MB/20.0 MiB) [ 0.983947] VFS: Mounted root (vfat filesystem) readonly on device 8:1. Sounds like the user meant to have init in the initrd and possibly not have root on sda1 but the initrd is in an inappropriate format (ie not cpio.gz that's typical for initramfs). --Gene
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Op 2014-05-18 om 17:35 schreef Gene Cumm:> On May 18, 2014 3:12 PM, "Alexey Orishko" <alexey.orishko at gmail.com> wrote: > > On Sun, May 18, 2014 at 7:08 PM, Gene Cumm <gene.cumm at gmail.com> wrote: > > >> > On 05/16/2014 01:35 PM, Tal Lubko wrote: > >>>> > For some reason when I use kernel 3.7.1 and above the system > >>>> > crashes during boot and I have no reason why. > > >> > > >> Consider a lack of init or otherwise improper initrd (cpio format, > > >> compression, etc). Consider trying your old kernel with new initrd > > >> and/or new kernel with old initrd to start diagnosis. > > >> > > > > > > [ 0.000000] Kernel command line: root=/dev/sda1 initrd=rootfs.ext2 console=ttyS0,115200 BOOT_IMAGE=bzImage > > > > > > [ 0.297992] Unpacking initramfs... > > > [ 0.298619] Initramfs unpacking failed: junk in compressed archive > > > [ 0.299718] Freeing initrd memory: 5636k freed > > > > If GPT used instead of MBR, no initrd required...:-)> > > > Alexey > > GPT versus MBR makes no difference. Storage adapter and its driver location > is more critical, followed by tools for decryption. > > [ 0.710918] scsi 0:0:0:0: Direct-Access ATA QEMU HARDDISK 1.4. PQ: 0 ANSI: 5 > [ 0.712510] sd 0:0:0:0: [sda] 40960 512-byte logical blocks: (20.9 MB/20.0 MiB) > > [ 0.983947] VFS: Mounted root (vfat filesystem) readonly on device 8:1. > > Sounds like the user meant to have init in the initrd and possibly not have > root on sda1 but the initrd is in an inappropriate format (ie not cpio.gz > that's typical for initramfs). >More information about initrd at http://lxr.linux.no/#linux+v3.14.4/Documentation/initrd.txt It says initrd should have a path as value. A path that the bootloader (syslinux) can find / access. http://lxr.linux.no/#linux+v3.14.4/Documentation/initrd.txt#L56 And I agree both with Gene and hpa. Gene is right with "it is our community, it is up to us to improve it" hpa is right with "it not syslinux issue" Groeten Geert Stappers - -- Leven en laten leven -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) iEYEARECAAYFAlN6QGkACgkQOSINbgwa/7sjLwCeKKHuZsv5iIbyOWBIAaL3LZxE q4UAn1xvEDZDlPJ4I96soHLipWj8h0iK =NEJ8 -----END PGP SIGNATURE-----