<div dir="ltr"><div>Hi, I think this is a bug - but if it's an intended behavior, I cannot get the rationale behind it:<br><br></div><div>I have a cluster with 2 nodes, node01 disc based and node02 ram based.<br></div>
<div>I have clustered them. Then after shutting down both, rebooting node02 it starts in standalone mode.<br></div><div>Booting node01, it fails saying that node02 is no longer in cluster with it: I would expect this behavior only in case I had explicitly reset node02, but that is not the case: booting the RAM node without its peer let it loose its membership - that's seems quite odd to me.<br>
<br></div><div>IMHO a more reasonable behavior would be that node02, when started with node01 being down, not being able to find any reference peer, and being ram based, should just fail.<br></div><div><br></div><div>Can you confirm this is a bug, or if the intended behavior, what is the rationale?<br>
<br></div><div>RabbitMQ version is 3.1.5<br><br></div><div>Thanks,<br></div><div> Sigismondo<br><br></div></div>