Vijaikumar, Sachin will work on this quota issue.
Pranith
On 01/22/2015 12:46 AM, Pranith Kumar Karampuri wrote:
+ quota devs
Pranith
On 01/07/2015 07:37 PM, Bekk, Klaus (IKP) wrote:
Hi all,
we are running 4 server with Ubuntu 14.04 as glusterfs server with
3.5.2-ubuntu1~trusty1.
When
W dniu 2015-01-28 o 00:45, Franco Broi pisze:
> Glad that you found the problem. I wouldn't have thought that glusterfs
> should return before the filesystem is properly mounted, you should file
> a bug.
>
> https://bugzilla.redhat.com/enter_bug.cgi?product=GlusterFS
Ugh! This is known problem...
h
Hi,
I want to known why net recv 91MB/s, read only 22MB/s when do FIO read
test.
when I set cache-size=2GB or disable open-behind, this phenomenon
disappears.
thanks.
gluster 3.4.5 (default configure)
centos 6.5
replica 2, 2 server
client do FIO test
FIO command:
fio -directory
On Tue, 2015-01-27 at 14:09 +0100, Bartłomiej Syryjczyk wrote:
> OK, I removed part "2>/dev/null", and see:
> stat: cannot stat ‘/mnt/gluster’: Resource temporarily unavailable
>
> So I decided to add sleep just before line number 298 (this one with
> stat). And it works! Is it normal?
>
Glad t
Hi Niels,
I see no kernel NFS service running on the gluster node.
Here are the output
root@glusterprod001:~# rpcinfo
program version netid addressserviceowner
104tcp6 ::.0.111 portmapper superuser
103tcp6 ::.0.11
Ben,
Thanks for the detailed answer! We will look at implementing it and will
let you know how it goes. FYI, for our initial setup (and maybe in the
final setup) we're using distribute only (pure DHT), not replicas. So I
figured our bandwidth wouldn't be cut too much...
Jorge
On 01/27/15 11
No problem! With glusterfs replication is done client side by writing to
multiple bricks at the same time, this means that you bandwidth will be cut
into whatever number of replicas you chose. So in a replica 2 you bandwith is
cut in half, replica 3 into 1/3, and so on. Be sure to remember th
Hi Ben,
Sorry I didn't reply earlier, one week of much needed vacation got in
the way...
Anyway, here's what we're trying to do with glusterfs for now. Not much,
but I figured it would be a test:
Supporting a group that uses genomics data, we have a WHOLE bunch of
data that our users keep
No, there's not. I've been asking for this for years.
On 01/27/2015 10:09 AM, Ml Ml wrote:
Hello List,
i was able to produce a split brain:
[root@ovirt-node03 splitmount]# gluster volume heal RaidVolB info
Brick ovirt-node03.example.local:/raidvol/volb/brick/
Number of entries: 1
Brick ovirt
Hello List,
i was able to produce a split brain:
[root@ovirt-node03 splitmount]# gluster volume heal RaidVolB info
Brick ovirt-node03.example.local:/raidvol/volb/brick/
Number of entries: 1
Brick ovirt-node04.example.local:/raidvol/volb/brick/
/1701d5ae-6a44-4374-8b29-61c699da870b/dom_md/ids
Nu
Hello,
I ran a gluster volume profile while doing a dd to read from /dev/zero and
write to a file and noticed unusually high latency on the FSYNC, INODELK, and
FINODELK operations. This latency seems to correspond with very slow page
loads of our website. I'm looking for any help in finding w
Not elegant, but here is my short-term fix to prevent the issue after a
reboot:
Added 'noauto' to the mount in /etc/rc.local:
/etc/fstab:
#... Note: Used the 'noauto' for the NFS mounts and put the mount in
/etc/rc.local to ensure that
#... glsuter has been started before attempting to mount u
Hi. I'm using glusterfs 3.4 in replica 2 configuration. My main usage of
gluster is for openstack storage backend (glance and shared storage for
nova-compute) and i'm using gluster and i'm using gluster for about 1 year
(not really a production environment, so a limited set of experiments is
possib
In my /etc/fstab, I have the following:
gfsib01bkp.corvidtec.com:/homegfs_bkp /backup/homegfs
glusterfs transport=tcp,_netdev 0 0
gfsib01bkp.corvidtec.com:/Software_bkp /backup/Software
glusterfs transport=tcp,_netdev 0 0
gfsib01bkp.corvidtec.com:/Source_bkp /b
I rebooted the machine to see if the problem would return and it does.
Same issue after a reboot.
Any suggestions?
One other thing I tested was to comment out the NFS mounts in
/etc/fstab:
# gfsib01bkp.corvidtec.com:/homegfs_bkp /backup_nfs/homegfs nfs
vers=3,intr,bg,rsize=32768,wsize=32768 0
Hi, diese Antwort ging jetzt an die Liste
. gluster-users@gluster.org
Von: Bekk, Klaus (IKP)
Gesendet: Montag, 26. Januar 2015 12:38
An: Pranith Kumar Karampuri; gluster-users@gluster.org; Vijaikumar
Mallikarjuna; Sachin Pandit
Cc: Wochele, Doris (IKP)
Betreff: AW: [Gluster-users] Fwd: to mail
Hi,
Can the replica position be configured ?
Regards.
Bin.Yang
ZTE Information Security Notice: The information contained in this mail (and
any attachment transmitted herewith) is privileged and confidential and is
intended for the ex
After shutting down all NFS and gluster processes, there was still an
NFS process.
[root@gfs01bkp ~]# rpcinfo -p
program vers proto port service
104 tcp111 portmapper
103 tcp111 portmapper
102 tcp111 portmapper
104 udp
Hi,
I had a similar problem once. It happened after doing some unrelated
tests with NFS. I thought it was a problem I generated doing weird
things, so I didn't investigate the cause further.
To see if this is the same case, try this:
* Unmount all NFS mounts and stop all gluster volumes
* Ch
I have established the passwordless ssh session rsa keys and have tested that
successfully. And I have tried
gluster system:: execute gsec_create then issuing the the create push-pem which
fails as well. So I can never get to the point where I can actually issue the
start command.
gluster volu
Turning off nfslock did not help. Also, still getting these messages
every 3-seconds:
[2015-01-27 14:16:12.921880] W [socket.c:611:__socket_rwv] 0-management:
readv on /var/run/1f0cee5a2d074e39b32ee5a81c70e68c.socket failed
(Invalid argument)
[2015-01-27 14:16:15.922431] W [socket.c:611:__soc
Hi,
Lalatendu Mohanty, Niels de Vos, and myself will be holding GlusterFS
Office Hours at FOSDEM.
Look for us at the CentOS booth, from 16h00 to 17h00 on Saturday, 31
January.
FOSDEM is taking place this weekend, 31 Jan and 1 Feb, at ULB Solbosch
Campus, Brussels. FOSDEM is a free event, n
Hey,
is the first command working without setting LC_NUMERIC?
Can you check if your problem might be related to the reported issue in
the "[Gluster-users] Mount failed" mailing list thread?
-jhz
On 26/01/15 23:17 +0100, Nicolas Ecarnot wrote:
> Hello,
>
> Many people wrote here that the
On Tue, Jan 27, 2015 at 12:07:35PM +0100, Niels de Vos wrote:
> Hi all,
>
> This meeting is scheduled for anyone that is interested in learning more
> about, or assisting with the Bug Triage.
>
> Meeting details:
> - location: #gluster-meeting on Freenode IRC
> - date: every Tuesday
> - time: 12:
W dniu 2015-01-27 o 09:47, Bartłomiej Syryjczyk pisze:
> W dniu 2015-01-27 o 09:20, Franco Broi pisze:
>> Well I'm stumped, just seems like the mount.glusterfs script isn't
>> working. I'm still running 3.5.1 and the getinode bit of my script looks
>> like this:
>>
>> ...
>> Linux)
>> g
glusterfs-3.6.2 has been released.
The release source tar file and packages for Fedora {20,21,rawhide},
RHEL/CentOS {5,6,7}, Debian {wheezy,jessie}, Pidora2014, and Raspbian
wheezy are available at
http://download.gluster.org/pub/gluster/glusterfs/3.6/3.6.2/
(Ubuntu packages will be availab
Hi Pedro,
On 01/27/2015 12:43 PM, Pedro Serotto wrote:
Hi Xavi,
what happens is that Nova cannot create the file structure which is
needed in order to create the virtual machine.
In //var/lib/nova/i//nstances/, I can find only one directory under a
casual name (/random/), for example
/ca095fd4-1
Hi Xavi,what happens is that Nova cannot create the file structure which is
needed in order to create the virtual machine.In /var/lib/nova/instances, I can
find only one directory under a casual name (random), for example
ca095fd4-1c0c-4fb4-a2d8-a0d7586c0ffa. If I try to list the content, the co
Hi,
two comments here:
1 - I don’t think it’s possible, as libvirt would not accept this, but most of
all it is not needed, as gluster should handle correctly the failover with
gfapi too, at least this is what happens in my case (gluster 3.6.2 at the
moment);
2 - If you want to be able to rest
Hi Pedro,
do you get any specific error ?
From logs, the only thing I see is a warning about a GFID not being
available. This can cause troubles, but I'm not sure if this is the problem.
Anyway, version 3.6.2 has some fixes that improve on this. Is it
possible for you to try the newer versio
Hi all,
This meeting is scheduled for anyone that is interested in learning more
about, or assisting with the Bug Triage.
Meeting details:
- location: #gluster-meeting on Freenode IRC
- date: every Tuesday
- time: 12:00 UTC, 13:00 CET (run: date -d "12:00 UTC")
- agenda: https://public.pad.fsfe.o
W dniu 2015-01-27 o 09:53, Franco Broi pisze:
> What do you get is you do this?
>
> bash-4.1# stat -c %i /mnt/gluster
> 1
> -bash-4.1# echo $?
> 0
---
[root@apache2 ~]# /usr/sbin/glusterfs --volfile-server=apache1
--volfile-id=/testvol /mnt/gluster
[root@apache2 ~]# stat -c %i /mnt/gluster
1
[roo
W dniu 2015-01-27 o 09:43, Franco Broi pisze:
> Could this be a case of Oracle Linux being evil?
Yes, I'm using OEL. Now 7.0, earlier 6.6
--
Z poważaniem,
*Bartłomiej Syryjczyk*
___
Gluster-users mailing list
Gluster-users@gluster.org
http://www.gluste
Dear all, I try the disperse volume on 3 node cluster, but my instances never
build.
This is the log:
root@node-1:/var/log/glusterfs# tail -f var-lib-nova-instances.log [2015-01-27
10:03:11.510702] I [client-handshake.c:1415:select_server_supported_programs]
0-nova-vm-client-1: Using Program Glu
Hi, all.
I'm curious how we're supposed to find out about new releases. There are
several sources, each of which give either no information, or different
information.
Let me attempt a list of places I've tried just now in an attempt to find
the latest version:
- On GitHub I see there is a v3.6.2
Hi all,
As it happens every year, the processes for GSOC begin in February,
starting with the organization registration, which runs this year from
February 9th to 20th. Rest of the timeline can be viewed at [1].
Last year we participated in GSOC under the Fedora organization, we hadn't
registered
On 26/01/15 21:35, Rosemond, Sonny wrote:
I have a RHEL7 testing environment consisting of 6 nodes total, all
running Gluster 3.6.1. The master volume is distributed/replicated,
and the slave volume is distributed. Firewalls and SELinux have been
disabled for testing purposes. Passwordless SSH
What do you get is you do this?
bash-4.1# stat -c %i /mnt/gluster
1
-bash-4.1# echo $?
0
On Tue, 2015-01-27 at 09:47 +0100, Bartłomiej Syryjczyk wrote:
> W dniu 2015-01-27 o 09:20, Franco Broi pisze:
> > Well I'm stumped, just seems like the mount.glusterfs script isn't
> > working. I'm still r
W dniu 2015-01-27 o 09:20, Franco Broi pisze:
> Well I'm stumped, just seems like the mount.glusterfs script isn't
> working. I'm still running 3.5.1 and the getinode bit of my script looks
> like this:
>
> ...
> Linux)
> getinode="stat -c %i $i"
>
> ...
> inode=$( ${getinode} $moun
Could this be a case of Oracle Linux being evil?
On Tue, 2015-01-27 at 16:20 +0800, Franco Broi wrote:
> Well I'm stumped, just seems like the mount.glusterfs script isn't
> working. I'm still running 3.5.1 and the getinode bit of my script looks
> like this:
>
> ...
> Linux)
> getin
Yes ！
Have enough space, Tks!
I use xfs。 Dose with this.
At 2015-01-27 16:13:46, "Anatoly Pugachev" wrote:
Does host system (kvm host) has enough space for this glusterfs guest ?
On Tue, Jan 27, 2015 at 8:51 AM, 肖力 wrote:
Hi
I test GlusterFS on CentOS 7.0 and use for kvm vm.
My vm c
Well I'm stumped, just seems like the mount.glusterfs script isn't
working. I'm still running 3.5.1 and the getinode bit of my script looks
like this:
...
Linux)
getinode="stat -c %i $i"
...
inode=$( ${getinode} $mount_point 2>/dev/null);
# this is required if the stat retur
Does host system (kvm host) has enough space for this glusterfs guest ?
On Tue, Jan 27, 2015 at 8:51 AM, 肖力 wrote:
> Hi
> I test GlusterFS on CentOS 7.0 and use for kvm vm.
>
> My vm can start ,but system report err:
> I/O error,dev vda ,sector 3670
> I/O error,dev vda ,sector 367034357
W dniu 2015-01-27 o 09:08, Franco Broi pisze:
> So what is the inode of your mounted gluster filesystem? And does
> running 'mount' show it as being fuse.glusterfs?
---
[root@apache2 ~]# stat -c %i /mnt/gluster
101729533
[root@apache2 ~]# /usr/sbin/glusterfs --volfile-server=apache1
--volfile-id=/
So what is the inode of your mounted gluster filesystem? And does
running 'mount' show it as being fuse.glusterfs?
On Tue, 2015-01-27 at 09:05 +0100, Bartłomiej Syryjczyk wrote:
> W dniu 2015-01-27 o 09:00, Franco Broi pisze:
> > Your getinode isn't working...
> >
> > + '[' 0 -ne 0 ']'
> > ++ sta
W dniu 2015-01-27 o 09:00, Franco Broi pisze:
> Your getinode isn't working...
>
> + '[' 0 -ne 0 ']'
> ++ stat -c %i /mnt/gluster
> + inode=
> + '[' 1 -ne 0 ']'
>
> How old is your mount.glusterfs script?
It's fresh (I think so). It's from official repo:
---
[root@apache2 ~]# which mount.glusterfs
Your getinode isn't working...
+ '[' 0 -ne 0 ']'
++ stat -c %i /mnt/gluster
+ inode=
+ '[' 1 -ne 0 ']'
How old is your mount.glusterfs script?
On Tue, 2015-01-27 at 08:52 +0100, Bartłomiej Syryjczyk wrote:
> W dniu 2015-01-27 o 08:46, Franco Broi pisze:
> > [franco@charlie4 ~]$ stat -c %i /data
47 matches
Mail list logo
