All of lore.kernel.org
 help / color / mirror / Atom feed
From: Doug Ledford <dledford@redhat.com>
To: Neil Brown <neilb@suse.de>
Cc: Linux RAID Mailing List <linux-raid@vger.kernel.org>,
	Dan Williams <dan.j.williams@intel.com>
Subject: Re: [Patch mdadm] Add hot-unplug support to mdadm
Date: Tue, 06 Apr 2010 22:24:21 -0400	[thread overview]
Message-ID: <4BBBECD5.6010903@redhat.com> (raw)
In-Reply-To: <4BBBE7D2.6090608@redhat.com>

[-- Attachment #1: Type: text/plain, Size: 1962 bytes --]

On 04/06/2010 10:02 PM, Doug Ledford wrote:
> On 04/06/2010 09:30 PM, Neil Brown wrote:
>> On Mon, 05 Apr 2010 12:40:41 -0400
>> Doug Ledford <dledford@redhat.com> wrote:
>>   would it help do you think to support e.g.
>>
>>     mdadm -If --devnum $major:$minor $name
> 
> As I recall, on a remove event the major minor is already gone too (but
> I could be wrong, the machine has since been blown away and reinstalled
> and the file I had the whole env during a remove event stored off in is
> gone with it).
> 
> And I would be careful about using $env{DEVNAME}, in fact I'm removing
> usage of it in my local rules files right now because I've found it is
> inconsistent (part of the time it includes /dev/ and part of the time it
> doesn't, where as $tempnode is consistent).

OK, so to expand on this a bit further.  $env{DEVNAME} sometimes refers
to devices via a full path (aka, /dev/sda1), and sometimes by just the
last part of the path (aka, md0).  I found this when creating a rule to
do incremental assembly on containers during late udev startup (I need
this because the initramfs created by dracut *only* starts necessary
arrays, and if you have a additional arrays in a container, then the
change event generated during udev startup can be used to trigger
incremental assembly of those additional arrays, but $env{DEVNAME} will
just be md0 or whatever, not a full path name).  So, I switched
everything to $tempnode as it always points to a device special file.
Except when doing removal events, in which case $tempnode doesn't exist
because there is no node to point to.  So, long story short, you're best
off to use $tempnode on any device add|change events, and $env{DEVNAME}
on remove events.


-- 
Doug Ledford <dledford@redhat.com>
              GPG KeyID: CFBFF194
	      http://people.redhat.com/dledford

Infiniband specific RPMs available at
	      http://people.redhat.com/dledford/Infiniband


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

  reply	other threads:[~2010-04-07  2:24 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-04-05 16:40 [Patch mdadm] Add hot-unplug support to mdadm Doug Ledford
2010-04-06 16:26 ` Doug Ledford
2010-04-07  1:30 ` Neil Brown
2010-04-07  2:02   ` Doug Ledford
2010-04-07  2:24     ` Doug Ledford [this message]
2010-04-07  3:07       ` Doug Ledford
2010-04-07  5:32     ` Luca Berra
2010-04-07  6:59       ` Neil Brown
2010-04-08 23:31     ` Neil Brown
2010-04-09  0:33       ` Neil Brown
2010-04-09 20:02         ` Doug Ledford
2010-04-13  9:28         ` Tomáš Dulík
2010-04-13 16:27           ` Doug Ledford
2010-04-13 18:49           ` Doug Ledford
     [not found]             ` <4BC5ADB2.2060705@unart.cz>
2010-04-15  5:24               ` Neil Brown
2010-04-15 13:11                 ` Tomáš Dulík
2010-04-13 19:04         ` Doug Ledford

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=4BBBECD5.6010903@redhat.com \
    --to=dledford@redhat.com \
    --cc=dan.j.williams@intel.com \
    --cc=linux-raid@vger.kernel.org \
    --cc=neilb@suse.de \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.