From: Jonathan E Brassow <jbrassow@redhat.com>
To: LVM general discussion and development <linux-lvm@redhat.com>
Subject: Re: [linux-lvm] "System ID" entry missing in metadata (LVM2) ?!
Date: Wed, 18 Oct 2006 11:40:18 -0500 [thread overview]
Message-ID: <ef27ef6e007de550b667566332420b44@redhat.com> (raw)
In-Reply-To: <45365026.90006@anykey.de>
I think it works in the reverse...
vgexport adds a generic tag to the volume groups metadata, basically
saying "ignore me". Doing a 'vgs' on an exported volume group shows
the 'x' attribute; and trying to activate that volume group results in
"Volume group "<vgname>" is exported". So, after performing this
operation, no-one can use the volume group (until vgimport is run).
vgimport removes the generic tag, allowing the VG to be activated and
used again. One this command is run, anyone that can see the volume
group can use/alter the volume group.
Think of it as "import/export from the set of usable volume groups".
If you want to share the VGs, you have two options:
1) Use clustered LVM2. This is really the best option.
2) Set up your logical volumes on one machine (you should only use
linear or stripe in this scenario - never mirror or snapshot). Never
change the logical volume layout after creating it unless the other
hosts have deactivated the volume groups being shared. Run 'vgchange
-ay' on all machines that have access to the devices.
Clustered LVM2 makes sure that all changes to a shared volume group are
serialized to prevent corruption and makes sure to activate/deactivate
volumes on a cluster-wide basis. If you are never going to change
anything (no risk of corruption or inconsistencies), you might be able
to get away with using LVM2 as it is.
If you need more specialized access, you can use tags.
Note, if you are sharing a logical volume, the application (or file
system) sharing that volume must be cluster-aware.
brassow
On Oct 18, 2006, at 11:02 AM, Andreas Octav wrote:
> Hi,
>
> thanks for your response Jonathan, but I want to share the VGs between
> the hosts. So I hoped that there is something like this functionality:
> -> "vgimport VG" writes some kind of hostid (system_id?) in the
> metadata
> -> other hosts can�t access the VG
> -> "vgdeport VG" removes the ID, so anyone else can import the VG
>
> My C knowledge isn�t very good, but the sources seem to include a
> functionality like the one mentioned above.
>
>
> Kind regards,
> Andreas
>
> Jonathan E Brassow schrieb:
>>
>>>> Hi,
>>>>
>>>> i�m new to LVM2 and wondering if it�s possible to restrict access
>>>> to a
>>>> Volume Group to a single server (e.g. like under vxvm (vxdg
>>>> import/deport)).
>>>> If I import a VG by using vgimport it is still possible to access
>>>> the
>>>> VG
>>>> on another node in a shared SAN environment. Can I prevent this
>>>> somehow?
>>>>
>>>> I�m using lvm2-2.01.14-3.6 on servers running SuSE SLES9 SP3 x86_64.
>>>>
>>
>> You can use tags to achieve this, or you can specify specific volume
>> groups and logical volumes in lvm.conf under "volume_list".
>>
>> brassow
>>
>>
>> _______________________________________________
>> linux-lvm mailing list
>> linux-lvm@redhat.com
>> https://www.redhat.com/mailman/listinfo/linux-lvm
>> read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
>>
>
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@redhat.com
> https://www.redhat.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
>
next prev parent reply other threads:[~2006-10-18 16:37 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1591887.1161167319269.OPEN-XCHANGE.WebMail.wwwrun@eu.main.anykey>
2006-10-18 12:42 ` [linux-lvm] "System ID" entry missing in metadata (LVM2) ?! Andreas Octav
2006-10-18 15:33 ` Jonathan E Brassow
2006-10-18 16:02 ` Andreas Octav
2006-10-18 16:40 ` Jonathan E Brassow [this message]
2006-10-18 17:21 ` Andreas Octav
2006-10-18 17:25 ` Alasdair G Kergon
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=ef27ef6e007de550b667566332420b44@redhat.com \
--to=jbrassow@redhat.com \
--cc=linux-lvm@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).