All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christophe Varoqui <christophe.varoqui@free.fr>
To: device-mapper development <dm-devel@redhat.com>, pcaulfie@redhat.com
Subject: Re: Fwd: [Next problem with multipath-tools]
Date: Tue, 15 Feb 2005 22:48:44 +0100	[thread overview]
Message-ID: <42126E3C.2080306@free.fr> (raw)
In-Reply-To: <20050214145026.GB13778@tykepenguin.com>

Patrick,

remember me why you can't start the daemon later in the boot process ?
Last time when talked about that I guess we concluded the startup 
position didn't really care ...

Anyway, if you maintain the need, several solution might be evaluated :
o a cmdline flag to disable pidfile creation, for those distro that 
manage that in the init script
o a different default pidfile location, or a flag to point to a custom 
location
o rip the pidfile logic altogether and go for a pidof() thing in the tool

Concerning /var/cache/multipathd, I guess the overlay doesn't matter as 
the CLONE_NEWNS flag should protect our namespace.

Please comment on your prefered ways.

Regards,
cvaroqui

>
>	Hello Patrick,
>
>I installed the new version of libsysfs1 which fixed problem in location of 
>libsysfs in /lib (instead in /usr/lib). But I think that this bug fix didn't
>solve problem with multipath-tools. After reboot multipathd is able to start
>correctly when server is coming up. Unfortunately if /var is located on
>separate volume and is mounted at startup sequence after multipathd start, the
>new mounted file system will overlay directory structure (located in /var -
>i.e. /var/run and /var/cache/multipathd) for multipathd and daemon will stay
>nonfunctional. Multipathd is running but do nothing from this moment. It is
>necessary to restart multipathd after mounting of /var file system. I have no
>idea how to solve this problem (maybe it is necessary to modify multipathd to
>store information elsewhere than in /var - i.e. somewhere on root partition or
>in ram disk).
>  
>
>


--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel

      reply	other threads:[~2005-02-15 21:48 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-02-14 14:50 Fwd: [Next problem with multipath-tools] Patrick Caulfield
2005-02-15 21:48 ` Christophe Varoqui [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=42126E3C.2080306@free.fr \
    --to=christophe.varoqui@free.fr \
    --cc=dm-devel@redhat.com \
    --cc=pcaulfie@redhat.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 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.