<html><head></head><body style="font-size:10pt;font-family:Verdana,Arial,Helvetica,sans-serif;"><div>Thanks!</div><div><br></div><div>We see a lot of errors of the type</div><div>"E [rdma.c:4417:tcp_connect_finish] 0-glusterStore2-client-2: tcp connect to failed (Connection refused)"</div><div>These errors do appear on the two servers with infiniband, not on the ones without.</div><div>Also, despite this flood of error messages in nfs.log, things appear to work. Is it safe to ignore this?</div><div><br></div><div>Thanks again,</div><div><br></div><div>/jon</div><div><br></div><div><br></div><div><br></div>On Sep 17, 2012 08:05 "Vijay Bellur" <a href="mailto:vbellur@redhat.com"><vbellur@redhat.com></a> wrote:<br><blockquote type="cite">On 09/17/2012 12:39 AM, Jon Tegner wrote:<blockquote type="cite">Have a volume consisting of 4 bricks. It was set up using infiniband with<br><br>"Transport-type: tcp,rdma"<br><br>Since then the infiniband adapters have been removed from two of the<br>servers, and I would like to remove the "rdma-transport-type" from the<br>setup.<br><br>Please excuse me if this is a stupid question, but I haven't been able<br>to figure out how to achieve this (is it just by removing "rdma" from<br>the "volume.brick.mount.vol-files" on the 4 bricks)?<br></blockquote><br>Right now, there is no way to alter the transport type after a volume <br>has been created.<br><br>If you are interested in preventing mounts to happen over Infiniband, <br>blocking port 24008 on all servers might be an option.<br><br>-Vijay<br><br></blockquote></body></html>