public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Neil Horman <nhorman@tuxdriver.com>
To: Scott James Remnant <scott@ubuntu.com>
Cc: linux-kernel@vger.kernel.org, akpm@linux-foundation.org,
	earl_chew@agilent.com
Subject: Re: [PATCH] exec: Make do_coredump more robust and safer when using pipes in core_pattern
Date: Sun, 2 Aug 2009 19:50:05 -0400	[thread overview]
Message-ID: <20090802235004.GA11903@localhost.localdomain> (raw)
In-Reply-To: <1249220996.3638.0.camel@wing-commander>

On Sun, Aug 02, 2009 at 02:49:56PM +0100, Scott James Remnant wrote:
> On Sat, 2009-08-01 at 20:22 -0400, Neil Horman wrote:
> 
> > On Sat, Aug 01, 2009 at 07:28:52PM +0100, Scott James Remnant wrote:
> > > On Sat, 2009-08-01 at 09:41 -0400, Neil Horman wrote:
> > > 
> > > > > > Not without additional work.  If init crashed in the initramfs, I don't think
> > > > > > theres a way to handle that.  If it crashes at some later time, I think it just
> > > > > > gets restarted IIRC.  I'm sure you can change that behavior, but this patch
> > > > > > doesn't address that.
> > > > > > 
> > > > > When the system init daemon crashes, the kernel PANICs.  When not using
> > > > > core_pattern, this is ok, we get a core file - when using apport, as far
> > > > > as I can tell it never waits for apport to finish so we don't get the
> > > > > crash.
> > > > > 
> > > > This is non-sensical.  If init crashes, and the kernel panics, you'll only get a
> > > > core by sheer luck and good fortune.
> > > > 
> > > Or by being a bit clever.  Upstart catches the SIGSEGV and the signal
> > > handler forks a child process, unmasking the signal in that child
> > > process with no signal handler installed.
> > > 
> > I don't see how this works.  How is upstart (which by definition is a child of
> > init (pid 1)) going to catch a SIGSEGV from its parent?  How would any process
> > catch a signal targeted to its parent?
> > 
> Upstart *is* /sbin/init (pid 1)
> 
Ah, so basically, you catch sigsegv, and in the handler fork a child, and return
from the handler in the pid, so that the child crashes.  Yeah, I don't see why
that won't work with this patch.  If pid 1 waits for its child to crash, then
you should serialize on the collection of the core via the pipe.  Of couse, pid
1 will have to make sure that all the sysctls are set appropriately before it
encounters a crash.
Neil

> Scott
> -- 
> Scott James Remnant
> scott@ubuntu.com



      reply	other threads:[~2009-08-02 23:50 UTC|newest]

Thread overview: 94+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-06-22 17:28 [PATCH] exec: Make do_coredump more robust and safer when using pipes in core_pattern Neil Horman
2009-06-25 23:30 ` Andrew Morton
2009-06-26  1:49   ` Neil Horman
2009-06-26 10:48   ` Neil Horman
2009-06-26 16:20     ` Andrew Morton
2009-06-26 17:30       ` Neil Horman
2009-06-28 19:31       ` Andi Kleen
2009-06-28 20:52         ` Andrew Morton
2009-06-28 21:00           ` Andi Kleen
2009-06-28 21:18             ` Andrew Morton
2009-06-28 21:50               ` Eric W. Biederman
2009-06-28 21:35           ` Eric W. Biederman
2009-06-28 21:48             ` Andi Kleen
2009-06-28 22:06               ` Eric W. Biederman
2009-06-29  9:15                 ` Andi Kleen
2009-06-28 21:52             ` Andrew Morton
2009-06-26 18:00   ` Neil Horman
2009-06-26 18:02   ` [PATCH 1/2] exec: Make do_coredump more robust and safer when using pipes in core_pattern: recursive dump detection Neil Horman
2009-06-26 16:59     ` Oleg Nesterov
2009-06-26 20:24       ` Neil Horman
2009-06-26 19:14         ` [PATCH 0/2] do_coredump: misc cleanups Oleg Nesterov
2009-06-26 19:14           ` [PATCH 1/2] do_coredump: factor out put_cred() calls Oleg Nesterov
2009-06-26 22:40             ` Roland McGrath
2009-06-26 20:33               ` Oleg Nesterov
2009-06-26 19:16           ` [PATCH 2/2] do_coredump: move !ispipe code into "else" branch Oleg Nesterov
2009-06-26 20:18             ` Q: do_coredump() && d_unhashed() Oleg Nesterov
2009-06-26 22:57           ` [PATCH 0/2] do_coredump: misc cleanups Neil Horman
2009-06-26 19:37     ` [PATCH 1/2] exec: Make do_coredump more robust and safer when using pipes in core_pattern: recursive dump detection Andrew Morton
2009-06-26 20:17       ` Neil Horman
2009-06-26 18:03   ` [PATCH 2/2] exec: Make do_coredump more robust and safer when using pipes in core_pattern: wait for core collectors Neil Horman
2009-06-26 16:48     ` Oleg Nesterov
2009-06-26 20:20       ` Neil Horman
2009-06-29  0:33   ` [PATCH 1/2] exec: Make do_coredump more robust and safer when using pipes in core_pattern (v3) Neil Horman
2009-06-29  0:35   ` [PATCH 2/2] " Neil Horman
2009-06-28 22:24     ` Oleg Nesterov
2009-06-28 23:24       ` Oleg Nesterov
2009-06-29  2:36       ` Neil Horman
2009-06-28 23:32         ` Oleg Nesterov
2009-06-29 10:21           ` Neil Horman
2009-06-30  0:06             ` Oleg Nesterov
2009-06-29  0:32 ` [PATCH 0/2] " Neil Horman
2009-06-30 17:38 ` [PATCH 0/3] exec: Make do_coredump more robust and safer when using pipes in core_pattern (v4) Neil Horman
2009-06-30 17:42   ` [PATCH 1/3] exec: Make do_coredump more resilient to recursive crashes (v4) Neil Horman
2009-06-30 17:43   ` [PATCH 2/3] exec: let do_coredump limit the number of concurrent dumps to pipes (v4) Neil Horman
2009-06-30 17:46   ` [PATCH 3/3] exec: Allow do_coredump to wait for user space pipe readers to complete (v4) Neil Horman
2009-07-01  5:52     ` Oleg Nesterov
2009-07-01 10:31       ` Neil Horman
2009-07-01 12:25         ` Oleg Nesterov
2009-07-01 14:12           ` Neil Horman
2009-07-01 14:48             ` Oleg Nesterov
2009-07-01 15:26 ` [PATCH 0/3] exec: Make do_coredump more robust and safer when using pipes in core_pattern (v5) Neil Horman
2009-07-01 15:30   ` [PATCH 1/3] exec: Make do_coredump more resilient to recursive crashes (v5) Neil Horman
2009-07-01 15:34   ` [PATCH 2/3] exec: let do_coredump limit the number of concurrent dumps to pipes (v5) Neil Horman
2009-07-01 15:37   ` [PATCH 3/3] exec: Allow do_coredump to wait for user space pipe readers to complete (v5) Neil Horman
2009-07-01 16:06     ` Oleg Nesterov
2009-07-01 18:19       ` Neil Horman
2009-07-01 18:28 ` [PATCH 0/3] exec: Make do_coredump more robust and safer when using pipes in core_pattern (v6) Neil Horman
2009-07-01 18:31   ` [PATCH 1/3] exec: Make do_coredump more resilient to recursive crashes (v6) Neil Horman
2009-07-01 18:32   ` [PATCH 2/3] exec: let do_coredump limit the number of concurrent dumps to pipes (v6) Neil Horman
2009-07-01 18:37   ` [PATCH 3/3] exec: Allow do_coredump to wait for user space pipe readers to complete (v6) Neil Horman
2009-07-02  8:29     ` Oleg Nesterov
2009-07-02 10:29       ` Neil Horman
2009-07-02 11:36         ` Oleg Nesterov
2009-07-02 14:44           ` Neil Horman
2009-07-02 15:37             ` Oleg Nesterov
2009-07-02 17:53               ` Neil Horman
2009-07-03 10:10                 ` Oleg Nesterov
2009-07-02 22:57 ` [PATCH 0/3] exec: Make do_coredump more robust and safer when using pipes in core_pattern (v7) Neil Horman
2009-07-02 22:59   ` [PATCH 1/3] exec: Make do_coredump more resilient to recursive crashes (v7) Neil Horman
2009-07-02 23:00   ` [PATCH 2/3] exec: let do_coredump limit the number of concurrent dumps to pipes (v7) Neil Horman
2009-07-02 23:01   ` [PATCH 3/3] exec: Allow do_coredump to wait for user space pipe readers to complete (v7) Neil Horman
2009-07-03 10:16     ` Oleg Nesterov
2009-07-03 10:44 ` [PATCH 0/3] exec: Make do_coredump more robust and safer when using pipes in core_pattern (v8) Neil Horman
2009-07-03 10:50   ` [PATCH 1/3] exec: Make do_coredump more resilient to recursive crashes (v8) Neil Horman
2009-07-07 16:14     ` Neil Horman
2009-07-03 10:51   ` [PATCH 2/3] exec: let do_coredump limit the number of concurrent dumps to pipes (v8) Neil Horman
2009-07-07 16:15     ` Neil Horman
2009-07-03 10:52   ` [PATCH 3/3] exec: Allow do_coredump to wait for user space pipe readers to complete (v8) Neil Horman
2009-07-07 16:19     ` Neil Horman
2009-07-07 16:35       ` Oleg Nesterov
2009-07-07 16:13   ` [PATCH 0/3] exec: Make do_coredump more robust and safer when using pipes in core_pattern (v8) Neil Horman
2009-07-20 15:49   ` [PATCH 0/3] exec: Make do_coredump more robust and safer when using pipes in core_pattern (v9) Neil Horman
2009-07-20 16:27     ` [PATCH 1/3] exec: Make do_coredump more resilient to recursive crashes (v9) Neil Horman
2009-07-20 16:29     ` [PATCH 2/3] exec: let do_coredump limit the number of concurrent dumps to pipes (v9) Neil Horman
2009-08-07 17:08       ` Randy Dunlap
2009-07-20 16:32     ` [PATCH 3/3] exec: Allow do_coredump to wait for user space pipe readers to complete (v9) Neil Horman
2009-07-29 15:13 ` [PATCH] exec: Make do_coredump more robust and safer when using pipes in core_pattern Scott James Remnant
2009-07-29 20:18   ` Neil Horman
2009-07-31 20:20     ` Scott James Remnant
2009-08-01 13:41       ` Neil Horman
2009-08-01 18:28         ` Scott James Remnant
2009-08-02  0:22           ` Neil Horman
2009-08-02 13:49             ` Scott James Remnant
2009-08-02 23:50               ` Neil Horman [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20090802235004.GA11903@localhost.localdomain \
    --to=nhorman@tuxdriver.com \
    --cc=akpm@linux-foundation.org \
    --cc=earl_chew@agilent.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=scott@ubuntu.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox