Forgot to add that I am using Debian 5.0.1 and glusterfs 2.0.0 final, not using the gluster patched fuse.<br><br><div class="gmail_quote">On Fri, May 1, 2009 at 4:47 PM, Ate Poorthuis <span dir="ltr"><<a href="mailto:atepoorthuis@gmail.com">atepoorthuis@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">I have a fully working gluster setup (2 distributed nodes that are replicated on the server side to 2 other nodes). When exporting the mountpoint as samba share, the gigE link is fully saturated on writing over samba. However, when reading, performance is seriously reduced.<br>
<br>Client bandwidth:<br><br> | iface Rx Tx Total<br> ==============================================================================<br> lo: 0.00 KB/s 0.00 KB/s 0.00 KB/s<br>
eth0: 255.29 KB/s 5546.58 KB/s 5801.86 KB/s<br> eth1: 16890.96 KB/s 389.40 KB/s 17280.36 KB/s<br> ------------------------------------------------------------------------------<br>
total: 17146.25 KB/s 5935.98 KB/s 23082.23 KB/s<br><br>And the following errors shows up in the logs:<br><br>2009-05-01 16:31:25 E [fuse-bridge.c:2332:fuse_getlk_cbk] glusterfs-fuse: 73272: ERR => -1 (Transport endpoint is not connected)<br>
2009-05-01 16:31:25 E [fuse-bridge.c:2332:fuse_getlk_cbk] glusterfs-fuse: 73289: ERR => -1 (Transport endpoint is not connected)<br>2009-05-01 16:31:25 E [fuse-bridge.c:2332:fuse_getlk_cbk] glusterfs-fuse: 73306: ERR => -1 (Transport endpoint is not connected)<br>
2009-05-01 16:31:25 E [fuse-bridge.c:2332:fuse_getlk_cbk] glusterfs-fuse: 73323: ERR => -1 (Transport endpoint is not connected)<br><br>I hope someone can provide any pointers on how to solve this. Vol files can be found below.<br>
<br>Thanks a lot,<br><br>Ate<br><br>Server .vol<br>--------<br>volume posix1<br> type storage/posix # POSIX FS translator<br> option directory /srv/export/gfs1/ # Export this directory<br>end-volume<br>
<br>volume locks1<br> type features/locks<br> subvolumes posix1<br>end-volume<br><br>volume read-ahead1<br> type performance/read-ahead<br> option page-count 4<br> subvolumes locks1<br>end-volume<br><br>volume write-behind1 <br>
type performance/write-behind <br> subvolumes read-ahead1<br>end-volume<br><br>volume afr_52<br> type protocol/client<br> option transport-type tcp/client<br> option remote-host 192.168.5.52 <br> option remote-subvolume write-behind1<br>
end-volume<br><br>volume afr<br> type cluster/replicate<br> subvolumes write-behind1 afr_52<br>end-volume<br><br>### Add network serving capability to above brick.<br>volume server<br> type protocol/server<br> option transport-type tcp<br>
subvolumes write-behind1 afr<br> option auth.addr.write-behind1.allow * # Allow access to "brick" volume<br>option auth.addr.afr.allow * # Allow access to "brick" volume<br>end-volume<br><br><br>Client .vol<br>
-----------<br>volume brick1_51<br> type protocol/client<br> option transport-type tcp<br> option remote-host 192.168.5.51 # IP address of the remote brick<br> option remote-subvolume afr # name of the remote volume<br>
end-volume<br>volume brick1_101<br> type protocol/client<br> option transport-type tcp<br> option remote-host 192.168.5.101 # IP address of the remote brick<br> option remote-subvolume afr # name of the remote volume<br>
end-volume<br><br>volume ha1<br> type /testing/cluster/ha<br> subvolumes brick1_51 brick1_101<br>end-volume<br><br>volume brick1_52<br> type protocol/client<br> option transport-type tcp<br> option remote-host 192.168.5.52 # IP address of the remote brick<br>
option remote-subvolume afr # name of the remote volume<br>end-volume<br><br>volume brick1_102<br> type protocol/client<br> option transport-type tcp<br> option remote-host 192.168.5.102 # IP address of the remote brick<br>
option remote-subvolume afr # name of the remote volume<br>end-volume<br><br>volume ha2<br> type /testing/cluster/ha<br> subvolumes brick1_52 brick1_102<br>end-volume<br><br>volume bricks<br> type cluster/distribute<br>
option min-free-disk 5%<br> subvolumes ha1 ha2<br>end-volume<br><br>### Add readahead feature<br>volume readahead<br> type performance/read-ahead<br> option page-size 1MB # unit in bytes<br> option page-count 2 # cache per file = (page-count x page-size)<br>
subvolumes bricks<br>end-volume<br>### Add writeback feature<br>volume writeback<br> type performance/write-behind<br> option flush-behind off<br> subvolumes readahead <br>end-volume<br><br>
</blockquote></div><br>