[ 
https://issues.apache.org/jira/browse/AMBARI-16013?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15252589#comment-15252589
 ] 

Sebastian Toader commented on AMBARI-16013:
-------------------------------------------

{code:title=branch-2.2.2}
commit 95bc6b6056f7c87890cb2b1473b2fc72e6be89c4
Author: Toader, Sebastian <[email protected]>
Date:   Thu Apr 21 19:07:36 2016 +0200

    AMBARI-16013. Host_status stuck in UNKNOWN status after blueprint deploy 
with host in heartbeat-lost. (stoader)

{code}

{code:title=branch-2.2}
commit ff25b1d7f3142d18114b351180c1fff85caa8799
Author: Toader, Sebastian <[email protected]>
Date:   Thu Apr 21 19:07:36 2016 +0200

    AMBARI-16013. Host_status stuck in UNKNOWN status after blueprint deploy 
with host in heartbeat-lost. (stoader)
{code}

{code:title=trunk}
commit 5b5bf1a34b1b060202421fcdb91a73f47376c273
Author: Toader, Sebastian <[email protected]>
Date:   Thu Apr 21 19:09:56 2016 +0200

    AMBARI-16013. Host_status stuck in UNKNOWN status after blueprint deploy 
with host in heartbeat-lost. (stoader)
{code}

> Host_status stuck in UNKNOWN status after blueprint deploy with host in 
> heartbeat-lost
> --------------------------------------------------------------------------------------
>
>                 Key: AMBARI-16013
>                 URL: https://issues.apache.org/jira/browse/AMBARI-16013
>             Project: Ambari
>          Issue Type: Bug
>          Components: ambari-server
>    Affects Versions: 2.2.2
>            Reporter: Sebastian Toader
>            Assignee: Sebastian Toader
>            Priority: Blocker
>             Fix For: 2.2.2
>
>         Attachments: AMBARI-16013.branch-2.2.v2.patch, 
> AMBARI-16013.trunk.v2.patch
>
>
> Deploy a cluster using blueprint when all nodes are in HEARTBEAT_LOST state 
> (e.g. nodes already registered with Ambari server once but than all were 
> stopped prior posting the blueprint/cluster creation template to the server).
> The blueprint and cluster creation succeeded and UI looked good with all the 
> hosts in heartbeat-lost state. 
> Then start the agents one by one. Expected behaviour is that as soon as all 
> required nodes are up  Ambari to start scheduling tasks on the connected 
> nodes to install the cluster.
> However the hosts were stuck in {{host_status: UNKNOWN}} state and Ambari did 
> not start scheduling any tasks to the connected hosts.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to