[
https://issues.apache.org/jira/browse/CLOUDSTACK-6036?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tomasz Zieba reopened CLOUDSTACK-6036:
--------------------------------------
Steps to reproduce on ASC 4.2.1 (clean install) and on ASC 4.2.1 (upgrade from
3.0.2)
1. Running instance of the machine
1a) create script which delay stopping by 10 minutes
2. Stop VM
You can see in management-server.log.2014-02-24 any details.
Very strange thing is that:
a) when machine stopping process is delay'ed - HA-Worker take very dangerous
process - Stop was unsuccessful. Rescheduling. This is connected with "Unable
to update VM" status in DB.
b) consequently machine is stopped after 10 minutes but HA-Worker has a
rescheduled process of stopping machine.
c) so when you are starting a machine next time - it starts very well - but for
few minutes machine is stopped by reschduled process.
Details in attachement.
> CloudStack stops the machine for no reason
> -------------------------------------------
>
> Key: CLOUDSTACK-6036
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-6036
> Project: CloudStack
> Issue Type: Bug
> Security Level: Public(Anyone can view this level - this is the
> default.)
> Components: Management Server
> Affects Versions: 4.2.1
> Environment: ACS 4.2.1 after upgrade from 3.0.2
> XCP 1.1
> Reporter: Tomasz Zieba
> Assignee: Koushik Das
> Priority: Critical
> Fix For: 4.4.0
>
> Attachments: management-server.log.2014-02-19.gz,
> management-server.log.2014-02-20.gz, management-server.log.2014-02-24.txt
>
>
> After the upgrade from version 3.0.2 to 4.2.1 appeared very strange error
> associated with self-stopping machine after changing the offering.
> Steps to reproduce:
> 1. Running instance of the machine
> 2. Stop with the operating system
> 3. Change offering of machine
> 4. Start the machine
> 5. Waiting for 10 minutes
> 6. CloudStack stops the machine for no reason
> 7. Restart the machine
> In the logs you can find information:
> 2014-02-05 06:27:00,974 DEBUG [xen.resource.CitrixResourceBase]
> (DirectAgent-316:null) 11. The VM i-41-824-VM is in Running state
> 2014-02-05 06:27:00,974 DEBUG [agent.transport.Request]
> (DirectAgent-316:null) Seq 50-1756626952: Processing: { Ans: , MgmtId:
> 160544475005497, via: 50, Ver: v1, Flags: 10,
> [{"com.cloud.agent.api.ClusterSyncAnswer":{"_clusterId":2,"_newStates":{"i-41-824-VM":{"t":"f32a4fee-b64e-4868-9c52-2a27e32d4c0e","u":"Running","v":"viridian:true;acpi:true;apic:true;pae:true;nx:false;timeoffset:0;cores-per-socket:4"}},"_isExecuted":false,"result":true,"wait":0}}]
> }
> 2014-02-05 06:27:00,981 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (DirectAgent-316:null) VM i-41-824-VM: cs state = Running and realState =
> Running
> 2014-02-05 06:27:00,981 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (DirectAgent-316:null) VM i-41-824-VM: cs state = Running and realState =
> Running
> 2014-02-05 06:36:01,240 DEBUG [agent.transport.Request]
> (HA-Worker-1:work-1511) Seq 51-1374970375: Sending { Cmd , MgmtId:
> 160544475005497, via: 51, Ver: v1, Flags: 100111,
> [{"com.cloud.agent.api.StopCommand":{"isProxy":false,"executeInSequence":true,"vmName":"i-41-824-VM","wait":0}}]
> }
> 2014-02-05 06:36:01,240 DEBUG [agent.transport.Request]
> (HA-Worker-1:work-1511) Seq 51-1374970375: Executing: { Cmd , MgmtId:
> 160544475005497, via: 51, Ver: v1, Flags: 100111,
> [{"com.cloud.agent.api.StopCommand":{"isProxy":false,"executeInSequence":true,"vmName":"i-41-824-VM","wait":0}}]
> }
> 2014-02-05 06:36:01,383 DEBUG [xen.resource.CitrixResourceBase]
> (DirectAgent-150:null) 9. The VM i-41-824-VM is in Stopping state
> 2014-02-05 06:36:27,625 DEBUG [xen.resource.CitrixResourceBase]
> (DirectAgent-150:null) 10. The VM i-41-824-VM is in Stopped state
> You will notice that the stop of the machine corresponds to the component
> HA-Worker.
> Such operation after the upgrade is complicating work of users.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
