All of lore.kernel.org
 help / color / mirror / Atom feed
From: Milan Broz <mbroz@redhat.com>
To: LVM general discussion and development <linux-lvm@redhat.com>
Cc: xake@rymdraket.net
Subject: Re: [linux-lvm] lvm, config and commandline
Date: Thu, 27 Jan 2011 17:24:39 +0100	[thread overview]
Message-ID: <4D419C47.202@redhat.com> (raw)
In-Reply-To: <061e0b45500b99a3f38b4bfa88a9a591.squirrel@webmail.rymdraket.net>

On 01/27/2011 10:37 AM, xake@rymdraket.net wrote:
> I have a system with a script at bootup currently running something like
> "lvm pvscan && lvm vgscan && lvm vgchange -ay" all with the option
> "--config 'global { locking_dir = "/dev/.lvm" }'" since when the script
> runs /var/lock is not in a writeable state. --ignorelockingfailures is
> fine, but gives a message on stderr I do not want, but I still want to get
> other error messages so "2>/dev/null" is not an option.

Use --sysinit instead of --ignorelockingfailure, no need to set that
locking dir at all in this phase.

If you have some strange messages paste them here, --sysinit is exactly
here to handle read-only device boot problem.
(seems it is just poorly documented...)

Moreover, pvscan and vgscan are NOOP here, because it updates
lvm cache, not possible on read-only system.

All you want is probably:
/sbin/lvm vgchange -a y --sysinit


> I realized however that if I run this as a script using lvm as interpreter
> (running lvm once and the commands inside of lvm) I cut the time for this
> script in half.
> 
> But with that I have the following problem:
> 
> # echo "pvscan --config 'global { locking_dir = "/dev/.lvm" }'" | lvm

Do not do this. if there is lock dir set inlvm.conf to another directory,
it is just recipe for data corruption.

Do not use --config at all, there are only very special cases when it is needed
system init is not one of them.

> Also, would it not also make sens to have --config an option working
> directly on lvm (something like "lvm --config "something" <command>)?

it works this way for one command: lvm <command> --config "something"
but not for the script mode.

Milan

  reply	other threads:[~2011-01-27 16:24 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-01-27  9:37 [linux-lvm] lvm, config and commandline xake
2011-01-27 16:24 ` Milan Broz [this message]
2011-01-27 17:55   ` xake
2011-01-27 21:34     ` Milan Broz
2011-01-28  8:53       ` Peter Hjalmarsson

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=4D419C47.202@redhat.com \
    --to=mbroz@redhat.com \
    --cc=linux-lvm@redhat.com \
    --cc=xake@rymdraket.net \
    /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.