Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.user > #201238 > unrolled thread

disk encryption causing kernel panic

Back to article view | Back to linux.debian.user

Contents

  disk encryption causing kernel panic Bill <bill3@uniserve.com> - 2018-10-14 23:20 +0200
    Re: disk encryption causing kernel panic Reco <recoverym4n@enotuniq.net> - 2018-10-15 05:30 +0200
    Re: disk encryption causing kernel panic Pascal Hambourg <pascal@plouf.fr.eu.org> - 2018-10-15 20:10 +0200

#201238 — disk encryption causing kernel panic

Hi folks,

I'm unable to boot my previously stable workstation after attempting to 
install rinetd through systemd. Although rinetd worked fine prior to 
systemd it doesn't work now and caused /var to saturate (10Gb). This may 
or may not be related, but I mention it as prologue. I have access to 
the system via a rescue key.

The problem seems to be that there is a bug in cryptsetup? which causes 
a kernel panic. In the output below, md2 is (encrypted) swap and md8 is 
/text - an encrypted partition. /text currently has only a few files 
which I could sacrifice, although I'd rather not.

I successfully enter the password for md2 and then don't even see the 
request for md8_crypt (ie a prompt for input) but get the following 
(condensed) output below.

I'm hoping someone has seen this before and knows a workaround. I'd 
rather not do a complete reinstall if I don't have to. What would I do 
if there were a lot of files?

Thanks much,

	Bill

========

Please unlock disk md2_crypt:
cryptsetup (md2_crypt): set up successfully
Please enter passphrase for disk md8_crypt on /text!
[   51.450202] BUG: unable to handle kernel NULL pointer dereference
at    (null)
[   51,450255] IP: [<    (null)>]   (null)
[   51.450285] PGD 0 [   51.450296]
[   51.450311] Oops: 0010 [#1] SMP
[   51.450329] Modules linked in: ..........
[   51.451083] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.9.0-3-amd64 
#1 Debian 4.9.0-2+deb9u5
[   51.451123] Hardware name: Supermicro X9SAE/X9SAE, BIOS 2.0b 07/10/2013
[   51.451156] task: ffffffff8e40e500 task.stack: ffffffff8e400000
[   51.451185] RIP: 0010: [<0000000000000000>]  [>     (null)>]   (null)
...
...
...
[   51.451524] Stack:
...
...
[   51.451680] Call Trace:
...
...
[   51.452376] Code: Bad RIP value.
[   51.452403] RIP  [<        (null)>]     (null)
[   51.452432]  RSP  <ffff9b799dc03e08>
[   51.452451] CR2: 0000000000000000
[   51.452471] ---[end trace cd08982bd438f7e2 ]---
[   51.452495] Kernel panic - not syncing: Fatal exception in interrupt
[   51.452561] Kernel Offset: 0xc800000 from 0xffffffff81000000 
(relocation range: 0xffffffff80000000-0xffffffffbfffffff)
[   51.452621] ---[ end Kernel panic - not syncing: Fatal exception in 
interrupt

end of output from boot attempt

-- 
Sent using Icedove on Debian GNU/Linux.

[toc] | [next] | [standalone]

#201249

	Hi.

On Sun, Oct 14, 2018 at 01:32:31PM -0700, Bill wrote:
> I successfully enter the password for md2 and then don't even see the request for md8_crypt (ie a prompt for input) but get the following
> (condensed) output below.

The most interesting part of those kernel BUGs are stack traces.
The only useful things that I can see in your abridged dmesg are:

1) You're using an old kernel. The current one's called 4.9.0-8.

2) The problem seems to be related to the swap - it's swapper/0 that's
triggering the bug.

3) There are multiple CPUs/CPU cores/CPU threads there.

But stack trace is needed to proceed further.

Reco

[toc] | [prev] | [next] | [standalone]

#201271

Le 14/10/2018 à 22:32, Bill a écrit :
> 
> The problem seems to be that there is a bug in cryptsetup? which causes 
> a kernel panic.

AFAIK bugs in userland programs don't cause kernel panics. Kernel bugs 
do. Userland programs may only trigger kernel bugs which cause a kernel 
panic.

> In the output below, md2 is (encrypted) swap and md8 is 
> /text - an encrypted partition. /text currently has only a few files 
> which I could sacrifice, although I'd rather not.
> 
> I successfully enter the password for md2 and then don't even see the 
> request for md8_crypt (ie a prompt for input)
> 
> Please unlock disk md2_crypt:
> cryptsetup (md2_crypt): set up successfully
> Please enter passphrase for disk md8_crypt on /text!
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
What then is this ?

> [   51.451083] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.9.0-3-amd64
As Reco wrote, the panic seems to be related with the swap, which was 
unlocked just before the panic. Did you try to :
- not unlock md2_crypt but only md8_crypt
- unlock md2_crypt but not activate the swap within

You can use a live system or the Debian installer in rescue mode to do 
the tests.

> What would I do if there were a lot of files?

Restore from backup, as usual.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.user

csiph-web