Hi Cedric,<div><br></div><div>Regarding the geo-replication state, the log essentially means that the client-server communication between geo-rep master & slave has gone down,</div><div>this could be for various reasons.</div>
<div><br></div><div>We could narrow down to the exact cause, if you could run the session again with debug log-level and send us the log-files of master & slave.</div><div><br></div><div>run it in debug level by executing the following command:</div>
<div> <span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px">#gluster volume geo-replication test ssh://root@slave.mydomain.com:file:///data/test config log-level DEBUG</span></div>
<div><br></div><div>to locate the master's log-file execute the following command:</div><div><span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px"> #gluster volume geo-replication test ssh://root@slave.mydomain.com:file:///data/test config log-file</span></div>
<div><span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px"><br></span></div><div><font face="arial, sans-serif"><span style="border-collapse:collapse">to locate the slave log-file do the following:</span></font></div>
<div><font face="arial, sans-serif"><span style="border-collapse:collapse"> execute this command on the slave domain:</span></font></div><div><font face="arial, sans-serif"><span style="border-collapse:collapse"> #gluster volume geo-replication </span></font><span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px">ssh://root@slave.mydomain.com:file:///data/test</span><span style="border-collapse:collapse;font-family:arial, sans-serif"> config log-file</span></div>
<div><span style="border-collapse:collapse;font-family:arial, sans-serif"> you would get a template which includes ${session_owner}</span></div><div><span style="border-collapse:collapse;font-family:arial, sans-serif"><br>
</span></div><div><span style="border-collapse:collapse;font-family:arial, sans-serif"> to get the session_owner of the geo-replication session execute the following command in MASTER:</span></div><div><span style="border-collapse:collapse;font-family:arial, sans-serif"> </span><span class="Apple-style-span" style="border-collapse: collapse; font-family: arial, sans-serif; font-size: 13px; "> #gluster volume geo-replication test ssh://root@slave.mydomain.com:file:///data/test config</span><span class="Apple-style-span" style="border-collapse: collapse; font-family: arial, sans-serif; "> session-owner</span></div>
<div><span class="Apple-style-span" style="border-collapse: collapse; font-family: arial, sans-serif; "><br></span></div><div>Regarding nfs.rpc-auth-allow:</div>
<div>It is a bug which would be addressed in the next minor-release, you can follow the status of it at <a href="http://bugs.gluster.com/show_bug.cgi?id=2866" target="_blank">http://bugs.gluster.com/show_bug.cgi?id=2866</a></div>
<div><br></div><div><a href="http://bugs.gluster.com/show_bug.cgi?id=2866" target="_blank"></a>Regards,</div><div>Kaushik BV<br><br>
<div class="gmail_quote">
On Tue, May 3, 2011 at 1:57 PM, Cedric Lagneau <span dir="ltr"><<a href="mailto:cedric.lagneau@openwide.fr" target="_blank">cedric.lagneau@openwide.fr</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">hi,<br>
<br>
I've some issue with geo-replication (since 3.2.0) and nfs auth (since initial release).<br>
<br>
<br>
Geo-replication<br>
---------------<br>
System : Debian 6.0 amd64<br>
Glusterfs: 3.2.0<br>
<br>
MASTER (volume) => SLAVE (directory)<br>
For some volume it works, but for others i can't enable geo-replication and have this error with a faulty status:<br>
2011-05-03 09:57:40.315774] E [syncdutils:131:log_raise_exception] <top>: FAIL:<br>
Traceback (most recent call last):<br>
File "/usr/lib/glusterfs/glusterfs/python/syncdaemon/syncdutils.py", line 152, in twrap<br>
tf(*aa)<br>
File "/usr/lib/glusterfs/glusterfs/python/syncdaemon/repce.py", line 118, in listen<br>
rid, exc, res = recv(self.inf)<br>
File "/usr/lib/glusterfs/glusterfs/python/syncdaemon/repce.py", line 42, in recv<br>
return pickle.load(inf)<br>
EOFError<br>
<br>
Command line :<br>
gluster volume geo-replication test slave.mydomain.com:/data/test/ start<br>
<br>
On /etc/glusterd i don't see any diff between /etc/glusterd files.<br>
<br>
#gluster volume geo-replication status<br>
MASTER SLAVE STATUS<br>
--------------------------------------------------------------------------------<br>
test ssh://root@slave.mydomain.com:file:///data/test faulty<br>
test2 ssh://root@slave.mydomain.com:file:///data/test2 OK<br>
<br>
<br>
<br>
NFS auth allow<br>
--------------<br>
Even if i set nfs.rpc-auth-allow on volume to restrict acces on some ip i can always mount via nfs. How it works ?<br>
<br>
Sample:<br>
<br>
#gluster volume set test nfs.rpc-auth-allow 10.0.0.10<br>
#gluster volume info test<br>
Options Reconfigured:<br>
nfs.rpc-auth-allow: 10.0.0.10<br>
<br>
My client ip 192.168.10.25 can mount the test volume with nfs with:<br>
mount -t nfs -o vers=3 glusterserveur:/test /mnt/test<br>
<br>
<br>
thanks for your help,<br>
<br>
<br>
best regards,<br>
<font color="#888888"><br>
<br>
--<br>
<br>
Cédric Lagneau<br>
_______________________________________________<br>
Gluster-users mailing list<br>
<a href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a><br>
<a href="http://gluster.org/cgi-bin/mailman/listinfo/gluster-users" target="_blank">http://gluster.org/cgi-bin/mailman/listinfo/gluster-users</a><br>
</font></blockquote></div><br></div>