hi, all
when I`m copying a file (about 1GB) into the volume, I tried to replace
one brick of the volume.
Then the copying was halted with the error msg: cp: writing ./d,
transport endpoint is not connected.
Besides, I use two server bricks without afr.
I`m not sure, is it
Hi Doug,
Try to make the change suggested by Anand and let us know how you get on. I am
interested to hear about the performance on 3.3 because bad performance has
been subject of many emails for a while here.
Regards,
Fernando
-Original Message-
From:
Jules Wang wrote:
hi, all
when I`m copying a file (about 1GB) into the volume, I tried to
replace one brick of the volume.
Then the copying was halted with the error msg: cp: writing
./d, transport endpoint is not connected.
Besides, I use two server bricks *without* afr.
Unfortunately we have had similar problems. RDMA in GlusterFS 3.3.x is
not well (or even at all, maybe?) supported.
Although I was able to mount a volume with transport rdma using
mount -t glusterfs masterib:/volume.rdma /volume
Note the .rdma after volume name.
Due to the lack of rdma support in
I have an existing volume configured to use GbE and just got two
Infiniband cards. How can I reconfigure the peers to use the IPoIB?
On 10/19/12 2:48 PM, Bartek Krawczyk wrote:
Due to the lack of rdma support in 3.3.x we decided to stick with plain IPoIB.
yum groupinstall Infiniband support
yum install opensm ibacm
in /etc/rdma/rcma.conf:
IPOIB_LOAD=yes
service rdma start
service opensm start
service ibacm start
when you'll have ib0 etc. interfaces just configure them with IP
addresses and peer probe those new addreses in glusterfs.
Be careful
On 19 October 2012 15:00, Bartek Krawczyk bbartlomiej.m...@gmail.com wrote:
in /etc/rdma/rcma.conf:
IPOIB_LOAD=yes
/etc/rdma/rdma.conf of course
--
Bartek Krawczyk
network and system administrator
___
Gluster-users mailing list
Hi Fernando, Anand,
Thanks for the suggestion. I removed the enable-direct-io option in
/etc/fstab, remounted, and the performance is the same within
statistical precision.
cheers, Doug
On 10/19/2012 02:13 AM, Fernando Frediani (Qube) wrote:
Hi Doug,
Try to make the change suggested by
Greetings,
If you want to learn more about Unified File and Object, new documentation, the
newly released 3.3.1 or have other burning questions, come on by the community
office hours in #gluster-meeting at 2pm EDT / 11am PDT / 18:00 GMT
This is going to be a weekly thing we try to do to keep
Video of today's session:
http://www.youtube.com/watch?v=ra7ZWh_otb4
Respond here with topics you'd like to see us cover in future sessions.
-JM
- Original Message -
Greetings,
If you want to learn more about Unified File and Object, new
documentation, the newly released 3.3.1
Hi
On Oct 19, 2012, at 4:30 PM, John Mark Walker johnm...@redhat.com wrote:
Video of today's session:
http://www.youtube.com/watch?v=ra7ZWh_otb4
Respond here with topics you'd like to see us cover in future sessions.
I'd like to see you discuss a situation where a cluster will not
On 10/19/2012 02:02 PM, Frank Carmickle wrote:
I'd like to see you discuss a situation where a cluster will not accept more
members. This is with the 3.3 branch.
Thanks
--FC
The problem with a specific support question instead of a topic is that
the conversation would go something like:
Thanks Vijay!
On Fri, Oct 19, 2012 at 10:58 PM, Vijay Bellur vbel...@redhat.com wrote:
On 10/19/2012 11:10 PM, Chandan Kumar wrote:
Hi Vijay,
Do you have a bug report of this issue? Basically I was planning to
deply Gluster replication mode on ext4 in production next month. Not
sure how
13 matches
Mail list logo