[Gluster-users] libgfapi using unix domain sockets

Ankireddypalle Reddy areddy at commvault.com
Wed May 25 10:03:41 UTC 2016 

Poornima,
            Thanks for checking this. We are using disperse volumes. Unix domain sockets will be used for communication between libgfapi and the brick daemons on the local server. The communication to brick daemons on the other nodes of the volume would  be through tcp/rdma. Is my assumption correct.

Thanks and Regards,
Ram

From: Poornima Gurusiddaiah [mailto:pgurusid at redhat.com]
Sent: Wednesday, May 25, 2016 2:09 AM
To: Ankireddypalle Reddy
Cc: gluster-users at gluster.org
Subject: Re: [Gluster-users] libgfapi using unix domain sockets

Hi,

Whenever a new fd is created it is allocated from the mem-pool, if the mem pool is full it will be calloc'd. The current limit for fd-mem-pool is 1024, if there are more than 1024 fd's open, then the perf may be affected.
Also, the unix socket used while glfs_set_volfile_server() is only for Vol file, i.e. only for the connection btw Client and glusterd (management deamon). Hence, you may not see the IO performance increase, the patch http://review.gluster.org/#/c/12709/ introduces unix socket domain for IO path. This is what you may be interested in i guess.

Regards,
Poornima

________________________________
From: "Ankireddypalle Reddy" <areddy at commvault.com<mailto:areddy at commvault.com>>
To: gluster-users at gluster.org<mailto:gluster-users at gluster.org>
Sent: Tuesday, May 24, 2016 9:16:31 PM
Subject: Re: [Gluster-users] libgfapi using unix domain sockets

Is there any suggested best practice for the number of glfs_fd_t  that can be associated with a glfs_t. Does having a single glfs_t in an application with large number of glfs_fd_t cause any resource contention issues.

Thanks and Regards,
Ram
From: gluster-users-bounces at gluster.org<mailto:gluster-users-bounces at gluster.org> [mailto:gluster-users-bounces at gluster.org] On Behalf Of Ankireddypalle Reddy
Sent: Tuesday, May 24, 2016 11:34 AM
To: gluster-users at gluster.org<mailto:gluster-users at gluster.org>
Subject: Re: [Gluster-users] libgfapi using unix domain sockets

I figured it out.

Protocol: unix
Hostname: /var/run/glusterd.socket
Port: 0

Thanks and Regards,
Ram
From: gluster-users-bounces at gluster.org<mailto:gluster-users-bounces at gluster.org> [mailto:gluster-users-bounces at gluster.org] On Behalf Of Ankireddypalle Reddy
Sent: Tuesday, May 24, 2016 10:20 AM
To: gluster-users at gluster.org<mailto:gluster-users at gluster.org>
Subject: [Gluster-users] libgfapi using unix domain sockets

Hi,
        I am trying to use libgfapi for connecting to a gluster  volume using unix domain sockets. I am not able to find the socket path that should be provided while making the “glfs_set_volfile_server” function call.

ps -eaf | grep gluster
root     15178 31450  0 09:52 pts/1    00:00:00 grep --color=auto gluster
root     26739 26291  0 May16 ?        00:01:52 /opt/commvault/Base/IndexingService -serviceName IndexingService_cleanup -cn glustervm6 cvshost:glustervm6*glustervm6 cvsport:58600:0 cvsmyplatform:2 cvsremoteplatform:4 -vm Instance001
root     28335     1  0 May12 ?        00:02:15 /usr/sbin/glusterd -p /var/run/glusterd.pid --log-level INFO
root     30047     1  0 May12 ?        00:06:38 /usr/sbin/glusterfsd -s glustervm6sds --volfile-id StoragePool.glustervm6sds.ws-disk1-ws_brick -p /var/lib/glusterd/vols/StoragePool/run/glustervm6sds-ws-disk1-ws_brick.pid -S /var/run/gluster/9ed1d13b4265b95be4ed642578e7f28b.socket --brick-name /ws/disk1/ws_brick -l /var/log/glusterfs/bricks/ws-disk1-ws_brick.log --xlator-option *-posix.glusterd-uuid=3ab81d79-9a99-4822-abb2-62e76a029240 --brick-port 49152 --xlator-option StoragePool-server.listen-port=49152
root     30066     1  0 May12 ?        00:13:58 /usr/sbin/glusterfsd -s glustervm6sds --volfile-id StoragePool.glustervm6sds.ws-disk2-ws_brick -p /var/lib/glusterd/vols/StoragePool/run/glustervm6sds-ws-disk2-ws_brick.pid -S /var/run/gluster/be6fc96032a95d6bf00d41049ca0356a.socket --brick-name /ws/disk2/ws_brick -l /var/log/glusterfs/bricks/ws-disk2-ws_brick.log --xlator-option *-posix.glusterd-uuid=3ab81d79-9a99-4822-abb2-62e76a029240 --brick-port 49153 --xlator-option StoragePool-server.listen-port=49153
root     30088     1  0 May12 ?        00:00:21 /usr/sbin/glusterfs -s localhost --volfile-id gluster/nfs -p /var/lib/glusterd/nfs/run/nfs.pid -l /var/log/glusterfs/nfs.log -S /var/run/gluster/93db4047a97542a6457b2178ce6512d7.socket
root     30093     1  0 May12 ?        00:10:24 /usr/sbin/glusterfs -s localhost --volfile-id gluster/glustershd -p /var/lib/glusterd/glustershd/run/glustershd.pid -l /var/log/glusterfs/glustershd.log -S /var/run/gluster/3d435606821403370720761863000928.socket --xlator-option *replicate*.node-uuid=3ab81d79-9a99-4822-abb2-62e76a029240
root     30186     1  0 May12 ?        00:00:31 /usr/sbin/glusterfs --volfile-server=glustervm6sds --volfile-id=/StoragePool /ws/glus

Thanks and Regards,
Ram





***************************Legal Disclaimer***************************

"This communication may contain confidential and privileged material for the

sole use of the intended recipient. Any unauthorized review, use or distribution

by others is strictly prohibited. If you have received the message by mistake,

please advise the sender by reply email and delete the message. Thank you."

**********************************************************************





***************************Legal Disclaimer***************************

"This communication may contain confidential and privileged material for the

sole use of the intended recipient. Any unauthorized review, use or distribution

by others is strictly prohibited. If you have received the message by mistake,

please advise the sender by reply email and delete the message. Thank you."

**********************************************************************



***************************Legal Disclaimer***************************

"This communication may contain confidential and privileged material for the

sole use of the intended recipient. Any unauthorized review, use or distribution

by others is strictly prohibited. If you have received the message by mistake,

please advise the sender by reply email and delete the message. Thank you."

**********************************************************************

_______________________________________________
Gluster-users mailing list
Gluster-users at gluster.org<mailto:Gluster-users at gluster.org>
http://www.gluster.org/mailman/listinfo/gluster-users




***************************Legal Disclaimer***************************
"This communication may contain confidential and privileged material for the
sole use of the intended recipient. Any unauthorized review, use or distribution
by others is strictly prohibited. If you have received the message by mistake,
please advise the sender by reply email and delete the message. Thank you."
**********************************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160525/53539cab/attachment.html>

More information about the Gluster-users mailing list