Hi Francis,<br>what is the version of glusterfs you are using?<br><br>regards,<br><br><div class="gmail_quote">On Thu, Dec 11, 2008 at 5:38 PM, Francis GASCHET <span dir="ltr"><<a href="mailto:fg@numlog.fr">fg@numlog.fr</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;">
<div bgcolor="#ffffff" text="#000000">
Hi all,<br>
<br>
I installed glusterFS on 2 computers under Mandriva 2008.<br>
Connection type : Ethernet 100 mbits/S (mii-tool result: negotiated
100baseTx-FD flow-control)<br>
<br>
Hereafter is my configuration (same on both PCs):<br>
<br>
<b><big>glusterfs-server.vol:<br>
<br>
</big></b><tt>volume dir_main<br>
type storage/posix # POSIX FS
translator<br>
option directory /main # Export
this directory<br>
end-volume<br>
<br>
volume locks_main<br>
type features/posix-locks<br>
subvolumes dir_main<br>
end-volume<br>
<br>
volume main<br>
type protocol/server<br>
option transport-type tcp/server # For
TCP/IP transport<br>
subvolumes locks_main<br>
option auth.ip.main.allow <a href="http://127.0.0.1" target="_blank">127.0.0.1</a>,<a href="http://172.16.1." target="_blank">172.16.1.</a>* # Allow
access to "brick" volume<br>
end-volume</tt><br>
<br>
<br>
<b><big>client.vol:<br>
<br>
</big></b><tt>volume main_loc<br>
type protocol/client<br>
option transport-type tcp/client<br>
option remote-host localhost<br>
option remote-subvolume main<br>
end-volume<br>
<br>
volume main_dist<br>
type protocol/client<br>
option transport-type tcp/client<br>
option remote-host other<br>
option remote-subvolume main<br>
end-volume<br>
<br>
volume raid_main_afr<br>
type cluster/afr<br>
subvolumes main_loc main_dist<br>
option read-subvolume main_loc<br>
end-volume<br>
<br>
volume raid_main_ra<br>
type performance/read-ahead<br>
option page-size 128kB<br>
option page-count 4<br>
option force-atime-update off<br>
subvolumes raid_main_afr<br>
end-volume<br>
<br>
volume raid_main_wb<br>
type performance/write-behind<br>
option aggregate-size 1MB<br>
option flush-behind on<br>
subvolumes raid_main_ra<br>
end-volume<br>
<br>
volume raid_main<br>
type performance/io-cache<br>
option cache-size 512MB<br>
option page-size 1MB<br>
option priority *:0 # *.html:2,*:1<br>
option force-revalidate-timeout 2 # default is 1<br>
subvolumes raid_main_wb<br>
end-volume<br>
</tt><br>
<br>
It works fine, but slowly !<br>
I'm a newbie in glusterFS, so may be some option isn't adequate. Please
advise.<br>
<br>
Due to the option "<tt>read-subvolume main_loc" I didn't expect network
traffic when I just list files or read them, but actually, even with a
simple ls, I see a lot of network traffic.<br>
A "ls -R" takes 7 to 8 seconds for less than 8000 files. If I do it
locally, I get the result instantly.<br>
</tt>Question 1 : Is this traffic normal on read only operation ?<br>
<br>
Question 2 : in the documentation, I read that there is 2 protocols :
ASCII protocol and binary protocol. Currently, according to what I see
with tcpdump, my glusterFS uses the ASCII protocol. I guess it's not
the best for performance ! What is the way to enforce it using the
binary protocol ?<br>
<br>
Thank's for any help.<br>
<br>
Best regards,<br>
<pre cols="72">--
Francis GASCHET / NUMLOG
<a href="http://www.numlog.fr" target="_blank">http://www.numlog.fr</a>
Tel.: +33 (0) 130 791 616
Fax.: +33 (0) 130 819 286
NUMLOG recrute sur LOLIX :
<a href="http://fr.lolix.org/" target="_blank">http://fr.lolix.org/</a>
</pre>
</div>
<br>_______________________________________________<br>
Gluster-users mailing list<br>
<a href="mailto:Gluster-users@gluster.org">Gluster-users@gluster.org</a><br>
<a href="http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users" target="_blank">http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users</a><br>
<br></blockquote></div><br><br clear="all"><br>-- <br>Raghavendra G<br><br>