Whit, after testing for a while (after copying several 100000 files) it seems that either glusterfs or glusternfs is crashing under load. The the average load on the machine goes up to 8 or 9, before it was max around 1, but there is no according process I found a post which describes this behaviour: http://serverfault.com/questions/365061/high-load-average-low-cpu-usage-why So it could be that the load goes up after gluster nfs has crashed. There is not much information in the nfs.log [2012-12-05 21:02:18.035575] W [socket.c:1512:__socket_proto_state_machine] 0-gv01-client-1: reading from socket failed. Error (Transport endpoint is not connected), peer (10.12.96.27:24009) [2012-12-05 21:02:18.036141] I [client.c:2090:client_rpc_notify] 0-gv01-client-1: disconnected [2012-12-05 21:02:28.780956] E [socket.c:1715:socket_connect_finish] 0-gv01-client-1: connection to 10.XXX.XXX.XXX:24009 failed (Connection refused) [2012-12-05 21:19:41.217198] I [glusterfsd.c:1666:main] 0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version 3.3.1 [2012-12-05 21:19:42.389196] I [nfs.c:821:init] 0-nfs: NFS service started [2012-12-05 21:19:42.393121] W [graph.c:316:_log_if_unknown_option] 0-nfs-server: option 'rpc-auth.auth-glusterfs' is not recognized [2012-12-05 21:19:42.393180] W [graph.c:316:_log_if_unknown_option] 0-nfs-server: option 'rpc-auth-allow-insecure' is not recognized [2012-12-05 21:19:42.393212] W [graph.c:316:_log_if_unknown_option] 0-nfs-server: option 'transport-type' is not recognized [2012-12-05 21:19:42.393269] I [client.c:2142:notify] 0-gv01-client-0: parent translators are ready, attempting connect on transport [2012-12-05 21:19:42.400269] I [client.c:2142:notify] 0-gv01-client-1: parent translators are ready, attempting connect on transport Given volfile: +------------------------------------------------------------------------------+ Just for anybody else, this is on Centos 6.3, Samba 3.6.10 I still don't know what causes this behaviour. If anybody has an idea how to fix it... Thanks, Gunnar Am 05.12.2012 14:33, schrieb Whit Blauvelt: > Gunnar, > >> Second fastest is #1, nfs mount shared by Samba 4000 files in around 6 min >> Slowest is #2 where I need more than 12 min for 4000 files. > Thanks for running that test. That's a significant difference. > > I wonder in the Samba > Gluster client > Gluster server scenario whether the > slowness is the Gluster client transacting with both servers rather than > just the local one. > > You've at least confirmed my suspicion that Samba > NFS > Gluster is not at > any speed disadvantage. And in many months of running that way, as I said, > there have been no performance complaints - although with this an > unsupported configuration it could turn out we've just been lucky and that > there's something yet that can go wrong. > > Whit
