[Gluster-users] cannot create a new volume with a brick that used to be part of a deleted volume?

[John Mark Walker]

This sounds like something we need to post to community.gluster.org - Avati, can you do the honors? 

-JM 

----- Original Message -----

> The current behavior is intentional. There have been far too many
> instances where users delete a volume, but fail to understand that
> those brick directories still contain their data (and all the
> associated book-keeping metadata like self-heal pending changelogs,
> partial dht hash ranges etc.) -- and will not be happy when they
> find that the newly created volume using those stale brick
> directories start misbehaving. This typically happens when a user is
> trying out gluster for the first time (where volume creation and
> deletion is frequent, while trying to get a hang of things) and
> result in an ugly first experience.

> For all you know, you yourself might have possibly ended in a
> situation where you could have created a new volume with all the
> staleness (like the hidden .glusterfs directory as well) from the
> previous volume silently carried over and cause unintended behavior.
> The way I see it, your email report is a positive result of the
> stale brick check having served its purpose :-)

> Avati

> On Tue, Sep 18, 2012 at 11:29 AM, Lonni J Friedman <
> netllama at gmail.com > wrote:

> > Hrmm, ok. Shouldn't 'gluster volume delete ...' be smart enough to
> 
> > clean this up so that I don't have to do it manually? Or
> 
> > alternatively, 'gluster volume create ...' should be able to figure
> 
> > out whether the path to a brick is really in use?
> 

> > As things stand now, the process is rather hacky when I have to
> > issue
> 
> > the 'gluster volume delete ...' command, then manually clean up
> 
> > afterwards. Hopefully this is something that will be addressed in a
> 
> > future release?
> 

> > thanks
> 

> > On Tue, Sep 18, 2012 at 11:26 AM, Kaleb Keithley <
> > kkeithle at redhat.com > wrote:
> 
> > >
> 
> > > There are xattrs on the top-level directory of the old brick
> > > volume
> > > that gluster is detecting causing this.
> 
> > >
> 
> > > I personally always create my bricks on a subdir. If you do that
> > > you can simply rmdir/mkdir the directory when you want to delete
> > > a
> > > gluster volume.
> 
> > >
> 
> > > You can clear the xattrs or "nuke it from orbit" with mkfs on the
> > > volume device.
> 
> > >
> 
> > >
> 
> > > ----- Original Message -----
> 
> > > From: "Lonni J Friedman" < netllama at gmail.com >
> 
> > > To: gluster-users at gluster.org
> 
> > > Sent: Tuesday, September 18, 2012 2:03:35 PM
> 
> > > Subject: [Gluster-users] cannot create a new volume with a brick
> > > that used to be part of a deleted volume?
> 
> > >
> 
> > > Greetings,
> 
> > > I'm running v3.3.0 on Fedora16-x86_64. I used to have a
> > > replicated
> 
> > > volume on two bricks. This morning I deleted it successfully:
> 
> > > ########
> 
> > > [root at farm-ljf0 ~]# gluster volume stop gv0
> 
> > > Stopping volume will make its data inaccessible. Do you want to
> 
> > > continue? (y/n) y
> 
> > > Stopping volume gv0 has been successful
> 
> > > [root at farm-ljf0 ~]# gluster volume delete gv0
> 
> > > Deleting volume will erase all information about the volume. Do
> > > you
> 
> > > want to continue? (y/n) y
> 
> > > Deleting volume gv0 has been successful
> 
> > > [root at farm-ljf0 ~]# gluster volume info all
> 
> > > No volumes present
> 
> > > ########
> 
> > >
> 
> > > I then attempted to create a new volume using the same bricks
> > > that
> 
> > > used to be part of the (now) deleted volume, but it keeps
> > > refusing
> > > &
> 
> > > failing claiming that the brick is already part of a volume:
> 
> > > ########
> 
> > > [root at farm-ljf1 ~]# gluster volume create gv0 rep 2 transport tcp
> 
> > > 10.31.99.165:/mnt/sdb1 10.31.99.166:/mnt/sdb1
> 
> > > /mnt/sdb1 or a prefix of it is already part of a volume
> 
> > > [root at farm-ljf1 ~]# gluster volume info all
> 
> > > No volumes present
> 
> > > ########
> 
> > >
> 
> > > Note farm-ljf0 is 10.31.99.165 and farm-ljf1 is 10.31.99.166. I
> > > also
> 
> > > tried restarting glusterd (and glusterfsd) hoping that might
> > > clear
> 
> > > things up, but it had no impact.
> 
> > >
> 
> > > How can /mnt/sdb1 be part of a volume when there are no volumes
> > > present?
> 
> > > Is this a bug, or am I just missing something obvious?
> 
> > >
> 

> > _______________________________________________
> 
> > Gluster-users mailing list
> 
> > Gluster-users at gluster.org
> 
> > http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
> 

> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20120919/425e1c7e/attachment.html>