2009/3/28 Sean Davis :
>
> My understanding is that distribute uses a hash to do the distribution. So,
> if you copy the same file, you will get the same result every time; that is,
> it will go to the same server every time. Hashing is deterministic in that
> sense. Vikas or someone else may cl
2009/3/28 Simon Liang :
> Is there a way to use a round-robin or something, as opposed to hash?
>
> The problem with distribute is that it does not know the disk has run out of
> space, until it's starting copying and run out of space.
> Can this be avoided?
As Sean Davis said in another mail in t
Hi.
I aware of two bugs that happen with the latest release:
1) A gradual memory leak happening in the client.
2) The AFR not synchronizing, even with favorite child specified and ls -lR
run multiple times.
Any idea how these 2 serious issues can be reported and resolved by the
developers?
Reg
Hello list
I am having problems with my GlusterFS setup (git version). I have two systems
acting as GlusterFS server and client on the same time. I used AFR on the
server side and the client used RRDNS to connect to the servers. But after
having data loss with that setup I moved to client side
2009/3/30 Steve :
Are you using the latest git (or use a release > RC7).
Did your servers go down and come back?
Is this reproducible?
Vikas
--
Engineer - Z Research
http://gluster.com/
___
Gluster-users mailing list
Gluster-users@gluster.org
http:/
Hallo Vikas
> Are you using the latest git (or use a release > RC7).
>
Latest git.
> Did your servers go down and come back?
>
During the phase where I did the test: No.
> Is this reproducible?
>
In my environment: Yes.
> Vikas
> --
Steve
> Engineer - Z Research
> http://gluster.com/
2009/3/30 Steve :
> Hallo Vikas
>> Is this reproducible?
>>
> In my environment: Yes.
Shortest sequence of steps to reproduce?
Vikas
--
Engineer - Z Research
http://gluster.com/
___
Gluster-users mailing list
Gluster-users@gluster.org
http://zresearch
> Shortest sequence of steps to reproduce?
>
Two systems in my case:
1: hermes
2: tisiphone
On both systems I do:
# /etc/init.d/glusterfs-client.vu-hosting stop
# /etc/init.d/glusterfs-server.vu-hosting stop
# /etc/init.d/fuse stop
# umount /mnt/glusterfs/vu-hosting/
# mkfs.xfs -f -l version=2 -
On Mon, 2009-03-30 at 15:07 +0200, Steve wrote:
> I copy one file to the target directory and everything seams to be
> okay. But then for no reason the file on one of the clients just has 0
> bytes size and after some while the other client has as well the file
> with 0 byte size.
>
> Here the use
Matt,
You can just add bricks to unify subvolumes. Adding bricks should not
break anything, provided you don't remove any of the already existing
subvoumes and the namespace subvolumes.
--
Gowda
On Fri, Mar 27, 2009 at 10:36 PM, Matt Lawrence wrote:
> Just so I don't really screw something up,
Original-Nachricht
> Datum: Mon, 30 Mar 2009 18:08:26 +0100
> Von: John Leach
> An: Steve
> CC: gluster-users@gluster.org
> Betreff: Re: [Gluster-users] Trouble with AFR - bug tracker?
> On Mon, 2009-03-30 at 15:07 +0200, Steve wrote:
> > I copy one file to the target director
So we upgraded to the latest GIT realease.
Still seeing the errors from postgres during restore.
Here is our client log.
=
=
=
=
=
=
=
Version : glusterfs 2.0.0git built on Mar 30 2009 10:40:10
TLA Revision : git://git
12 matches
Mail list logo