Well, as I turn on debug mode and restart services (both nova-compute
and neutron-ovs-agent) on the compute nodes, things get better, so far
all new instances start fine. We will see how long this lasts ...
Thanks,
Xin
On 1/23/2015 3:19 PM, Zhao, Xin wrote:
On 1/23/2015 2:49 PM, Steven Dake wrote:
> On 01/23/2015 09:33 AM, Zhao, Xin wrote:
>>
>> On 1/23/2015 11:26 AM, Lars Kellogg-Stedman wrote:
>>> On Fri, Jan 23, 2015 at 11:10:35AM -0500, Zhao, Xin wrote:
>>>> without an IP assigned. The scheduler log shows the scheduler has
>>>> chosen
>>>> the compute node to start the instance, but the neutron server log
>>>> shows no
>>>> sign of allocating MAC/IP addresses for it, I suspect the neutron
>>>> server
>>>> didn't get the request for some reason.
>>>>
>>>> Any suggestions on how to debug this?
>>> Are there any relevant errors in the nova-compute log on the target
>>> compute hosts?
>>
>> Nothing in the compute.log on the compute host. But, I don't have
>> debug turned on there, I will turn it on and see if there are
>> anything...
>>
>> Thanks,
>> Xin
>>
> Are you using qpid or rabbitmq? Does restarting either allow the
> rest of the compute services to proceed? (OpenStack will reconnect
> when you restart the amqp service).
>
Yes I tried restarting rabbitmq, but that doesn't help.
Thanks,
Xin
> Regards
> -steve
>
>> _______________________________________________
>> Rdo-list mailing list
>> Rdo-list(a)redhat.com
>>
https://www.redhat.com/mailman/listinfo/rdo-list
>>
>> To unsubscribe: rdo-list-unsubscribe(a)redhat.com
>
_______________________________________________
Rdo-list mailing list
Rdo-list(a)redhat.com
https://www.redhat.com/mailman/listinfo/rdo-list
To unsubscribe: rdo-list-unsubscribe(a)redhat.com