The networking setup is based on FlatDHCP manager that is based on nova-network. Though the bridges were deleted, I was able to recover them. Once when I was able to start VMs using the virt-manager, I saw the virtual network interface on the compute node. But, I was not able to perform SSH to the VMs. In the compute and network logs, there is a new error that says AMQP server on the cloud controller is not accessible.
Can anybody help me in this regard to get rid of the problem? From: He, Jiantao H [mailto:jiantao.h...@intel.com] Sent: Donnerstag, 9. Oktober 2014 03:29 To: Narayanan, Krishnaprasad; 'openstack@lists.openstack.org' Subject: RE: Nova compute and network service are not starting after upgrading to Ubuntu 14.04 May be your bridges were deleted, what about 'ovs-vsctl show' command said? From: Narayanan, Krishnaprasad [mailto:naray...@uni-mainz.de] Sent: Thursday, October 9, 2014 07:23 To: 'openstack@lists.openstack.org' Subject: Re: [Openstack] Nova compute and network service are not starting after upgrading to Ubuntu 14.04 Dear all, In continuation with my email below, can somebody suggest me the steps to recover / setup a VNIC that was deleted from the compute node after upgradation process. When I tried executing ifconfig -a "name of the virtual interface" on the compute node, I am getting an error which displays "No device found". Regards, Krishnaprasad From: Narayanan, Krishnaprasad Sent: Mittwoch, 8. Oktober 2014 18:11 To: 'openstack@lists.openstack.org' Subject: Nova compute and service are not starting after upgrading to Ubuntu 14.04 Hallo all, I upgraded the operating system of one of the compute nodes in our cloud infrastructure from Ubuntu 12.04 to 14.04 and after the upgradation, I found that the nova-compute and network service were not started. The network configuration in our Havana setup is based on FlatDHCP manager. After the reboot process, I saw the nova-compute and network service was started but after a minute, they were stopped. When I executed "ps aux | grep nova" on the terminal, I saw a strange observation where I found more than 10 process related to nova-api service. The message in the nova-compute log is: "WARNING nova.virt.libvirt.driver [-] URI qemu:///system does not support events: Cannot write data: Broken pipe 2014-10-08 16:58:41.249 1968 ERROR nova.openstack.common.threadgroup [-] internal error: client socket is closed". The version of nova and nova-manage in the cloud controller and compute node are different. In the controller node, the nova version is 2.15.0 and nova-manage version is 2013.2.2 whereas on the compute node, their version is 2.17.0 and is 2014.1.2 Can I kindly know whether the problem is due to the following? a) The version mis-match of the nova components in the controller and compute node b) Are there any configurations that are to be updated in the libvirt / qemu / nova / nova-compute? Regards, Krishnaprasad
_______________________________________________ Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack Post to : openstack@lists.openstack.org Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
