<div dir="ltr"><br><div class="gmail_extra"><br><br><div class="gmail_quote">On Fri, Oct 25, 2013 at 9:14 AM, Brad Childs <span dir="ltr"><<a href="mailto:bdc@redhat.com" target="_blank">bdc@redhat.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div style="font-size:12pt;font-family:times new roman,new york,times,serif"><div></div><div><font size="3">If the replica took n+1 day to complete,</font></div>
</div></div></blockquote><div><br></div><div>Gluster's replication is synchronous. So writes are done in parallel. If a server was down and we self-heal it later, we sync both data and mtime.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="font-size:12pt;font-family:times new roman,new york,times,serif"><div><font size="3"> then returning the highest value would show that the file was modified a full day after the user considered it modified. Shouldn't it be the lesser value (if both replicas are consistant)?</font></div>
<div><br></div><div><span>I</span><span style="font-size:medium">n support of James statement </span>f<span style="font-size:medium">rom a user perspective, I may want to know the last time I wrote some data to a file-- timestamp is metadata for the user. Showing the date gluster completed replicating the file to another node is confusing.</span><div>
<br></div></div></div></blockquote><div><br></div><div>As I described above, that is not the case. Delayed replication (healing) happens for both data and mtime.</div><div><br></div><div>Avati</div><div><br></div><div> </div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="font-size:12pt;font-family:times new roman,new york,times,serif"><div><div></div></div><div><br></div><div>
<span name="x"></span>-bc<br><span name="x"></span><br></div><div><br></div><hr><blockquote style="padding-left:5px;font-size:12pt;font-style:normal;margin-left:5px;font-family:Helvetica,Arial,sans-serif;text-decoration:none;font-weight:normal;border-left:2px solid #1010ff">
<b>From: </b>"Anand Avati" <<a href="mailto:avati@gluster.org" target="_blank">avati@gluster.org</a>><br><b>To: </b>"James" <<a href="mailto:purpleidea@gmail.com" target="_blank">purpleidea@gmail.com</a>><br>
<b>Cc: </b>"gluster-users" <<a href="mailto:gluster-users@gluster.org" target="_blank">gluster-users@gluster.org</a>><br><b>Sent: </b>Thursday, October 24, 2013 7:12:56 PM<br><b>Subject: </b>Re: [Gluster-users] metadata for stat : Should it be identical?<div>
<div class="h5"><br><div><br></div><div dir="ltr">Gluster does have logic to always show mtime which is the highest in value. It is probably a bug if you are witnessing different mtimes at different times when no writes have happened in between.<div>
<br></div><div>Avati</div><div><br></div></div><div class="gmail_extra"><br><div><br></div><div class="gmail_quote">On Thu, Oct 24, 2013 at 4:31 PM, James <span dir="ltr"><<a href="mailto:purpleidea@gmail.com" target="_blank">purpleidea@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div>On Thu, 2013-10-24 at 13:00 -0700, Robert Hajime Lanning wrote:<br> ><br> > Design philosophy...<br> ><br>
> There is no metadata server. When you look at timestamps in stat,<br> > you<br> > are seeing the real stat of the file.<br></div>So this raises an interesting point...<br><div><br> ><br> > If you have "replica 2" then you have two files. The stat can come<br>
> from<br> > either one. Mtime will be the modification time of the file<br></div>If the replica N files all have slightly different mtimes (it seems they<br> usually will because they weren't written at exactly the same time),<br>
then isn't this a point of inconsistency for a script running on a fuse<br> mount which expects the same mtime on a file?<br> <br> Shouldn't gluster somehow coordinate to set all the files mtimes to be<br> consistent to say the last mtime in the replica set?<br>
<div><div><br> <br> > referenced<br> > by the time on the server (not the client.)<br> ><br> > One of the strengths of GlusterFS is that it does not have the<br> > bottleneck/single point of failure of a single metadata server.<br>
<br></div></div><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://supercolony.gluster.org/mailman/listinfo/gluster-users" target="_blank">http://supercolony.gluster.org/mailman/listinfo/gluster-users</a><br>
</blockquote></div><br></div><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://supercolony.gluster.org/mailman/listinfo/gluster-users" target="_blank">http://supercolony.gluster.org/mailman/listinfo/gluster-users</a></div>
</div></blockquote><div><br></div></div></blockquote></div><br></div></div>