<html><head></head><body>You can set quorum for volumes and/or servers, trading consistency for availability if that's more appropriate for your use case.<br><br><div class="gmail_quote">On February 14, 2014 11:26:26 PM PST, "Wangshen (Peter)" <peter.w@huawei.com> wrote:<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<pre class="k9mail">Hi, All<br />I'm working on distributed storage system analysis, including glusterfs.<br />With some information on internet and done some test, I found that when brain split or network partition occurs, both side of the system can accept new write requests. It seems gluster can not guarantee consistency in this situation.<br />According to CAP theorem(<a href="http://en.wikipedia.org/wiki/CAP_theorem">http://en.wikipedia.org/wiki/CAP_theorem</a>), it seems gluster is an AP system.<br /><br />I can't find any value information about my question, so could you explain these below, or give me some valuable link.<br />1. Is gluster an AP system? Whether there have some way in gluster to guarantee data consistency when the two situation occurs?<br />2. How does gluster handle brain split and network partition?<br />3. Does gluster treat brain split and network partition as the same situation?<br /><br /><br />Thanks!<br /><br />Peter<br /><hr /><br />Gluster-users
mailing list<br />Gluster-users@gluster.org<br /><a href="http://supercolony.gluster.org/mailman/listinfo/gluster-users">http://supercolony.gluster.org/mailman/listinfo/gluster-users</a><br /></pre></blockquote></div><br>
-- <br>
Sent from my Android device with K-9 Mail. Please excuse my brevity.</body></html>