cloudstack-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kyle Flavin <>
Subject RE: Removing a host from a cluster and adding it to another
Date Wed, 28 Jan 2015 04:19:10 GMT
Ah, yes I can confirm that there is a host tag with the following query on the CS database.
 Here is the sanitized output from the host_tags table:

select * from host_tags where host_id=<host_id>;

| id  | host_id | tag            |
| <id> |      <hosted> | <host_name> |

Is it safe to delete this entry, or should I add the host back into its old cluster and then
delete the tag through the UI?

-----Original Message-----
From: Somesh Naidu [] 
Sent: Tuesday, January 27, 2015 5:20 PM
Subject: RE: Removing a host from a cluster and adding it to another

Yes, check what's in the "tags" parameter of that host. Cleaning it should resolve the issue.


-----Original Message-----
From: Kyle Flavin [] 
Sent: Tuesday, January 27, 2015 7:50 PM
Subject: Removing a host from a cluster and adding it to another

I've removed a host from one Cloudstack cluster, and I'm attempting to add it to a new cluster.
 It will be the only host in the cluster, and used for testing purposes.  I put the host into
maintenance mode, removed it from its original cluster within the Cloudstack UI, then deleted
it from the Xen pool using xe pool-eject.

Now when I create a new cluster, and try to add the host, I get an error message that just
says "Unable to add host" from within the Cloudstack web UI.  I see messages in management-server.log
that say "Trying to add a new host at http://<myhost<http://%3cmyhost>>", but
no indication of an error:

2015-01-27 16:28:31,142 INFO  [c.c.r.ResourceManagerImpl] (catalina-exec-13:ctx-4ab1a388 ctx-fa2957ea)
Trying to add a new host at http://<myhost> in data center 1

On the Xen host itself, I see the following in the xensource.log:

Jan 28 00:28:31 <host> t1 /opt/xensource/bin/xapi: [debug|<host>|748 INET|pool_patch.get_by_uuid
D:5fb0d4d944bb|dispatcher] Server_helpers.exec exception_handler: Got exception UUID_INVALID:
[ pool_patch; 996dd2e7-ad95-49cc-a0be-2c9adc4dfb0b ]

It references an invalid ID, but I'm not sure what that UUID belongs to; it isn't the UUID
of the pool.

Is there something that needs to be done to "clear" the configuration of a Xenserver host
prior to adding it to a new cluster?  Or is Cloudstack "remembering" the old host name and
not allowing me to put it into the new cluster?


View raw message