<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Il 27/05/2016 13:34, Lindsay Mathieson ha scritto:<br>
<blockquote
cite="mid:2e685448-b689-9748-5246-43b1b37e1617@gmail.com"
type="cite">
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
<p>Major upgrades of any system tend to involve pain - I know the
ceph upgrade from Hammer to Jewel (9.x - 10.x) was a non trivial
procedure that involved taking the cluster down.<br>
</p>
</blockquote>
<br>
A clustered system, aiming to host many petabytes of datas should be<br>
developed by allowing rolling upgrads with backward compatibility.<br>
<br>
Shutting down the whole infrastructure is never an option. Actually,<br>
even with Gluster, we have the same drawback of standard SANs<br>
where a firmware update need to reboot the whole storage.<br>
</body>
</html>