[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: upgrade failed debian 8.7



Mike Nunn <bonnie650@gmail.com> wrote:
> [-- text/plain, encoding 7bit, charset: UTF-8, 24 lines --]

> I have an strace output for the following:

> strace -f -o foo update-initramfs -u -v

> It stopped at the following line as usual:

> Building cpio /boot/initrd.img-3.16.0-4-amd64.new initramfs

> The last few lines of the strace output:

> 12056 open("/usr/lib/locale/locale-archive", O_RDONLY|O_CLOEXEC) = 3
> 12056 fstat(3, {st_mode=S_IFREG|0644, st_size=2855152, ...}) = 0
> 12056 mmap(NULL, 2855152, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7f46dea0e000
> 12056 close(3)                          = 0
> 12056 sync(

> I have tried this twice with the same screen output and strace output.

Can you please check if process 12056 is "/bin/sync"? Search for a line
above looking like

3400  execve("/bin/sync", ["sync"], [/* 23 vars */]) = 0

(3400 being the PID of sync in my example here)

If this is the case, then you build stops at line 176 of
/usr/sbin/update-initramfs. Good news is: at that point the initramfs
has already been created successfully and moved into the right place.
The only step left is to run "update-grub".

But when the sync() system call blocks for such a long time, there is
something very wrong with either the kernel or the underlying medium.

Please test if just calling "strace sync" as root hangs as well.

But right now, I don't think this is a bug in update-initramfs but
somewhere else. Maybe inside the NVMe driver in the Linux Kernel.

Grüße,
Sven.

-- 
Sigmentation fault. Core dumped.


Reply to: