Has anyone gone from using FUSE to libgfapi? How much did performance
increase/decrease?
I'm working on some libgfapi code and it is slower then my FUSE based solution,
so I suspect something is wrong. I would like to know what kind of speed
increase others see so I know what to expect.
-bc
Fantastic work, thank you.
-Dan
Dan Mons
Skunk Works
Cutting Edge
http://cuttingedge.com.au
On 6 February 2014 22:50, Kaleb KEITHLEY wrote:
>
> YUM repo at http://download.gluster.org/pub/gluster/glusterfs/samba/ now has
> el6 RPMs for RHEL6, CentOS6; along with RPMs for Fedora
Great work Justin!
- Original Message -
> From: "Justin Clift"
> To: "gluster-devel"
> Sent: Wednesday, 5 February, 2014 4:15:09 PM
> Subject: [Gluster-devel] GlusterFlow (GlusterFS Analytics) 0.02
> Hi all,
> Feeling inspired, so did a marathon coding
> session and created the next
Resending after fixing a typo.
On 02/06/2014 08:13 PM, Lalatendu Mohanty wrote:
On 02/06/2014 06:20 PM, Kaleb KEITHLEY wrote:
YUM repo at http://download.gluster.org/pub/gluster/glusterfs/samba/
now has el6 RPMs for RHEL6, CentOS6; along with RPMs for Fedora 18
and Fedora 19.
Fedora 20 an
On 02/06/2014 06:20 PM, Kaleb KEITHLEY wrote:
YUM repo at http://download.gluster.org/pub/gluster/glusterfs/samba/
now has el6 RPMs for RHEL6, CentOS6; along with RPMs for Fedora 18 and
Fedora 19.
Fedora 20 and Fedora 21 (rawhide) have Samba 4.1.3 by default.
--
Kaleb
Kaleb,
This is awes
Hi all,
The name of the feature changed to 'Barrier' in order to avoid the
confusion between the 'Quiesce' translator.
A much detailed version of the document is updated here
http://www.gluster.org/community/documentation/index.php/Features/Server-side_Barrier_feature.
All comments/suggestio
YUM repo at http://download.gluster.org/pub/gluster/glusterfs/samba/ now
has el6 RPMs for RHEL6, CentOS6; along with RPMs for Fedora 18 and
Fedora 19.
Fedora 20 and Fedora 21 (rawhide) have Samba 4.1.3 by default.
--
Kaleb
___
Gluster-devel maili
As of now, NSR makes use of etcd (which uses the RAFT distributed consensus
protocol) to maintain information (term number, current leader) for the
nodes in a replica group rather than using it in the data replication plane.
On Thu, Feb 6, 2014 at 2:45 PM, Prashanth Pai wrote:
> Hi,
>
> There i
On Thu, 06 Feb 2014 01:12:37 -0800
Joe Julian wrote:
> I said it in IRC, and I'll repeat it here. This is amazingly cool.
Thanks. :)
Going to work on getting it multi-node capable and a bit more
cleaned up/useful over the next few days. Travelling back home
to London from Ghent atm though, so
Hi Prashanth,
Thanks for the info.
Looking forward for a better GlusterFS system :)
Thanks.
On Thu, Feb 6, 2014 at 5:15 PM, Prashanth Pai wrote:
> Hi,
>
> There is already an ongoing effort to use RAFT for synchronous
> replication. You can find the repo here:
>
> http://review.gluster.org
Hi,
There is already an ongoing effort to use RAFT for synchronous replication. You
can find the repo here:
http://review.gluster.org/#/q/project:glusterfs-nsr,n,z
https://forge.gluster.org/new-style-replication
Regards,
-Prashanth Pai
- Original Message -
From: "Sharuzzaman Ahmat Ras
I said it in IRC, and I'll repeat it here. This is amazingly cool.
I can even think of ways that I can use the heat maps I saw Jordan
demonstrating in kibana with this.
On 02/04/2014 07:15 PM, Justin Clift wrote:
Hi all,
Feeling inspired, so did a marathon coding
session and created the next
On 02/05/2014 04:13 AM, Niels de Vos wrote:
On Wed, Feb 05, 2014 at 04:15:09AM +0100, Justin Clift wrote:
Hi all,
Feeling inspired, so did a marathon coding
session and created the next version of
GlusterFlow. It's pretty much "Analytics
for GlusterFS".
https://github.com/justinclift/glus
On Thu, 6 Feb 2014 05:57:12 +0100
m...@netbsd.org (Emmanuel Dreyfus) wrote:
> No clue? It seems it fails because it prefers an IPv6 resolution on a
> machine that does not have it.
Guessing that git blame on the .c files mentioned in the snippet isn't
being helpful? :(
+ Justin
--
GlusterFS P
El 06/02/14 08:51, Vijay Bellur ha escrit:
On 02/06/2014 05:46 AM, Ira Cooper wrote:
Yep... this is an area I am very interested in, going forwards.
Especially sending messages back, we'll need that for any
caching/leasing/oplock/whatever we call it type protocols.
+1. I am very interested i
Hi Avati,
El 06/02/14 00:24, Anand Avati ha escrit:
Xavi,
Getting such a caching mechanism has several aspects. First of all we
need the framework pieces implemented (particularly server originated
messages to the client for invalidation and revokes) in a well
designed way. Particularly how w
Hi Anand,
Yes I tried both things you mention and they work just fine.
Thanks
On Feb 5, 2014 9:29 PM, "Anand Avati" wrote:
> The nop xlator by itself seems OK. Have you tried the stripe config with
> the nop xlator on top? or even without the nop xlator?
>
> Avati
>
>
> On Wed, Feb 5, 2014 at 3
Hi all,
I'm not sure if you all have heard about Raft Consensus Algorithm, but from
the paper, video, and slides on the Internet, I think that this algorithm
can help to solve a lot of distributed cases in GlusterFS.
For example, right now the initial mounting of the filesystem can use any
nodes
18 matches
Mail list logo
