cfs != storage<div><br></div><div>You need to get a highly available storage that is concurrently accessible from multiple nodes.</div><div><br></div><div>ocfs2 will allow multiple nodes to concurrently access the same storage. With posix semantics.</div>
<div>If a node dies, the remaining nodes will pause to recover and then continue functioning. The</div><div>dead node can then restart and rejoin the cluster.</div><div><br></div><div><div class="gmail_quote">On Thu, Sep 13, 2012 at 5:02 PM, Eric <span dir="ltr"><<a href="mailto:epretorious@yahoo.com" target="_blank">epretorious@yahoo.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:10pt;font-family:Courier New,courier,monaco,monospace,sans-serif"><div><div><div style="font-size:10pt;font-family:Courier New,courier,monaco,monospace,sans-serif">
Is it possible to create a highly-available OCFS2 cluster (i.e., A storage cluster that mitigates the single point of failure [SPoF] created by storing an OCFS2 volume on a single LUN)?<br><div style="font-style:normal;font-size:13.3333px;background-color:transparent;font-family:Courier New,courier,monaco,monospace,sans-serif">
<br></div><div style="font-style:normal;font-size:13.3333px;background-color:transparent;font-family:Courier New,courier,monaco,monospace,sans-serif">The
OCFS2 Project Page makes this claim...</div><div style="font-style:normal;font-size:13.3333px;background-color:transparent;font-family:Courier New,courier,monaco,monospace,sans-serif"><br></div><div style="font-style:normal;font-size:13.3333px;background-color:transparent;font-family:Courier New,courier,monaco,monospace,sans-serif">
> OCFS2 is a general-purpose shared-disk cluster file system for Linux capable of
providing both <i>high performance</i> and <i>high availability</i>.</div><div style="font-style:normal;font-size:13.3333px;background-color:transparent;font-family:Courier New,courier,monaco,monospace,sans-serif"><br></div>
<div style="font-style:normal;font-size:13.3333px;background-color:transparent;font-family:Courier New,courier,monaco,monospace,sans-serif">...but without backing-up the claim of high availability storage (at either the HDD- or the node-level).</div>
<div style="font-style:normal;font-size:13.3333px;background-color:transparent;font-family:Courier New,courier,monaco,monospace,sans-serif"><br></div><div style="font-style:normal;font-size:13.3333px;background-color:transparent;font-family:Courier New,courier,monaco,monospace,sans-serif">
I've found a couple of articles hinting at using Linux Multipathing or DRBD but very little detailed information about either.<br><br>TIA,<br>Eric Pretorious<br>Truckee, CA<br></div></div></div></div></div></div><br>
_______________________________________________<br>
Ocfs2-users mailing list<br>
<a href="mailto:Ocfs2-users@oss.oracle.com">Ocfs2-users@oss.oracle.com</a><br>
<a href="https://oss.oracle.com/mailman/listinfo/ocfs2-users" target="_blank">https://oss.oracle.com/mailman/listinfo/ocfs2-users</a><br></blockquote></div><br></div>