<div dir="ltr"><div>Done: <a href="https://github.com/infinispan/infinispan/pull/1900">https://github.com/infinispan/infinispan/pull/1900</a><br></div><br></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Thu, Jun 13, 2013 at 4:15 PM, Mircea Markus <span dir="ltr"><<a href="mailto:mmarkus@redhat.com" target="_blank">mmarkus@redhat.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi Dan,<br>
<br>
<snip><br>
But unfortunately the partial shutdown case does NOT work, if the coordinator node is a member of partial shutdown nodes.<br>
When the coordinator is down while suspending rebalance, one of the rest servers is becoming a new coordinator with rebalanceEnabled=true, and after that, the rebalance process is starting.<br>
Once this happened, it will cause the unexpected cache data accumulation on the nodes that are awaiting shutdown request.<br>
</snip><br>
<br>
Seems like this scenario is not covered. I guess we should broadcast (to all the members) the clean shutdown request so that the new coordinator would know pick it up and don't start the rebalance?<br>
<br>
Cheers,<br>
<span class="HOEnZb"><font color="#888888">--<br>
Mircea Markus<br>
Infinispan lead (<a href="http://www.infinispan.org" target="_blank">www.infinispan.org</a>)<br>
<br>
<br>
<br>
<br>
<br>
_______________________________________________<br>
infinispan-dev mailing list<br>
<a href="mailto:infinispan-dev@lists.jboss.org">infinispan-dev@lists.jboss.org</a><br>
<a href="https://lists.jboss.org/mailman/listinfo/infinispan-dev" target="_blank">https://lists.jboss.org/mailman/listinfo/infinispan-dev</a><br>
</font></span></blockquote></div><br></div>