<p>nope, at most dozens of sec as I observed</p>
<div class="gmail_quote">在 2012-7-20 PM10:04,"Dietmar Maurer" <<a href="mailto:dietmar@proxmox.com">dietmar@proxmox.com</a>>写道:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
> > Let's assume a complete recovery takes about 2 hours. Does that mean<br>
> > my VMs are blocked for 2 hours (instead of continue operation on other<br>
> nodes)?<br>
> ><br>
><br>
> This is actually why we spend lots of lines in recovery and IO patch, there are<br>
> some mechanism, such as request retry, oid scheduling that tries to complete<br>
> any request in a very short period because IOs from VM are timeouted by<br>
> guest kernel, for e.g,<br>
> 120 seconds for Linux kernel.<br>
<br>
So if a VM access such object, it would block for 2 hours (confused)?<br>
<br>
</blockquote></div>