Hi everyone, New to the list, and hope I can get some help.
I have configured Gluster in a AWS environment using 2 nodes. (servers/instances) I have a website that is running 20 plus apache instances. They are all connection to a NFS server for their code base. This past weekend we did a migration from NFS to Gluster. We are using the Gluster client on each apache instance to connect to the cluster. Everything worked fine, and on each apache server I could browse the mount to Gluster no problem with fine speed. It ran fine for about 7 hours, then apache started to fail. After logging into apache, the mounts to the cluster were still working, but we a bit slow while trying to do a "ls" on the dir I then shut down apache, umounted gluster, remounted gluster, and started apache.... It ran fine for another 10 to 20 minutes, then apache started to fail again. I assume the reason for the failure "could" be load, but Gluster and apache were not taxed at all.... My gut is telling me network, and also I am seeing this in the logs during the time of the issue: [socket.c:1494:__socket_proto_state_machine] 0-socket.management: reading from socket failed. Error (Transport endpoint is not connected), peer (76.226.144.165:1023) [2013-08-18 15:43:07.479124] W [socket.c:1494:__socket_proto_state_machine] 0-socket.management: reading from socket failed. Error (Transport endpoint is not connected), peer (109.22.29.1281023) [2013-08-18 15:43:07.506516] W [socket.c:1494:__socket_proto_state_machine] 0-socket.management: reading from socket failed. Error (Transport endpoint is not connected), peer (168.129.133.122:1023) [2013-08-18 15:43:07.531118] W [socket.c:1494:__socket_proto_state_machine] 0-socket.management: reading from socket failed. Error (Transport endpoint is not connected), peer (76.16.48.127:1023) [2013-08-18 15:43:07.564645] W [socket.c:1494:__socket_proto_state_machine] 0-socket.management: reading from socket failed. Error (Transport endpoint is not connected), peer (67.226.67.69:1023) [2013-08-18 15:43:07.569733] W [socket.c:1494:__socket_proto_state_machine] 0-socket.management: reading from socket failed. Error (Transport endpoint is not connected), peer (175.129.46.148:1023) [2013-08-18 15:43:07.586239] W [socket.c:1494:__socket_proto_state_machine] 0-socket.management: reading from socket failed. Error (Transport endpoint is not connected), peer (53.235.83.220:1023) /var/log/glusterfs/etc-glusterfs-glusterd.vol.log Thanks all in advance! Dan -- Dan Belkie Shelter Six Technologies Inc. 403.397.4491 http://www.sheltersix.com<http://www.sheltersix.com/> [email protected]<mailto:[email protected]> Skype: dbelkie [Logo CONVERT TO CURVES]
<<inline: image001.png>>
_______________________________________________ Gluster-users mailing list [email protected] http://supercolony.gluster.org/mailman/listinfo/gluster-users
