Re: "There are no more endpoints available from the endpoint mapper"???

From: John Toner [MVP] (jtoner_at_mvps.DIE.SPAM.DIE.org)
Date: 07/06/04


Date: Tue, 6 Jul 2004 09:23:51 -0400

Peter,

After running the first command (cluster node /force) that removes all of
the cluster settings from this node. If this is the last node in the
cluster, your cluster has just been totally wiped out.

The error you're getting in the second command is an RPC error which
basically means that this node cannot communicate with the cluster. For some
reason, the cluster was already on its way down...the evict command
shouldn't have had anything to do with the "crashing" of your cluster. In a
prior post, you were mentioning MNS, so it's possible that you lost the
majority when you issued the cluster node /force command.

Hope this makes sense.

Regards,
John

"Peter Steele" <psteele@z-force.com> wrote in message
news:%23KpuvfSYEHA.3972@TK2MSFTNGP12.phx.gbl...
> I used the cluster command to attempt to remove the first node of a
cluster,
> using this:
>
> cluster node mynode1 /force
>
> That much worked okay and the Cluster Administrator showed that the node
had
> been taken offline. It's resources had automatically been assumed but node
> 2. I then tried
>
> cluster node mynode1 /evict
>
> and got the error below:
>
> >cluster node clu1 /evict
> System error 1753 has occurred (0x000006d9).
> There are no more endpoints available from the endpoint mapper.
>
> Shortly after this the cluster went offline completely and it doesn't look
> like it will recover by itself.
>
> Is it incorrect to remove the first node of a cluster? I assume not since
> the node could of course crash and another node would have to assume its
> role. So what happened here?
>
>