Brian Smith
2012-Feb-01 21:49 UTC
[Gluster-users] Sorry for x-post: RDMA Mounts drop with "transport endpoints not connected"
Having serious issues w/ glusterfs 3.2.5 over rdma. Clients are periodically dropping off with "transport endpoint not connected". Any help would be appreciated. Environment is HPC. GlusterFS is being used as a shared /work|/scratch directory. Standard distributed volume configuration. Nothing fancy. Pastie log snippet is here: http://pastie.org/3291330 Any help would be appreciated! -- Brian Smith Senior Systems Administrator IT Research Computing, University of South Florida 4202 E. Fowler Ave. ENB308 Office Phone: +1 813 974-1467 Organization URL: http://rc.usf.edu
Joe Landman
2012-Feb-01 21:51 UTC
[Gluster-users] Sorry for x-post: RDMA Mounts drop with "transport endpoints not connected"
On 02/01/2012 04:49 PM, Brian Smith wrote:> Having serious issues w/ glusterfs 3.2.5 over rdma. Clients are > periodically dropping off with "transport endpoint not connected". Any > help would be appreciated. Environment is HPC. GlusterFS is being used > as a shared /work|/scratch directory. Standard distributed volume > configuration. Nothing fancy. > > Pastie log snippet is here: http://pastie.org/3291330 > > Any help would be appreciated! >What OS, kernel rev, OFED, etc. What HCAs, switch, etc. What does ibv_devinfo report for nodes experiencing the transport endpoint issue? -- Joseph Landman, Ph.D Founder and CEO Scalable Informatics Inc. email: landman at scalableinformatics.com web : http://scalableinformatics.com http://scalableinformatics.com/sicluster phone: +1 734 786 8423 x121 fax : +1 866 888 3112 cell : +1 734 612 4615