Ask Your Question

calm07's profile - activity

2019-06-08 07:41:14 -0500 received badge  Popular Question (source)
2019-06-03 13:57:43 -0500 asked a question Flavor with name already exists

I am having a strange issue with creating flavors. Note that I have 1000+ flavors. When I try to create a flavor with a specific name, it throws an error saying "Flavor with name already exists". However if I try to show the flavor, I get an error saying - "No flavor with a name exists". If I modify the flavor name slightly by adding an underscore, I am able to create the flavor. It looks like there is some error in matching the name. Any suggestions on how to debug this issue will be much appreciated.

2018-12-13 18:40:18 -0500 asked a question Program is slower on VM than on physical node

I am observing noticeably faster speeds on a physical node when compared to a VM. More details are below.

I have a complicated C++ program that uses OpenMP. It runs at an average of 4-5 mins on a physical node. When I use the same physical node as a compute node and launch a VM on it, it runs at an average of 10-11 minutes. I understand that virtualization will add some overhead but currently the physical node is 2x faster than the VM. Is that the best I can get with virtualization? If I run "top" on the VM, I notice that it uses less CPU% than compared to the physical node.

If I run the program on an Amazon AWS instance, it runs at an average of 4-5 minutes. So it seems to me that some optimization can be done to make my VM faster. I have tried CPU pinning with no noticeable improvement.

There is only one related question in the OpenStack forums - (https://ask.openstack.org/en/question...). But I have already implemented the solution provided there.

I would be very grateful if someone could point me in the right direction. Right now, I am not sure how to debug the issue.

2015-04-26 06:17:28 -0500 received badge  Famous Question (source)
2015-03-03 00:46:39 -0500 received badge  Notable Question (source)
2015-02-27 00:12:10 -0500 commented answer vm can not ping outer world with domain name

This works on neutron ! Thanks a lot

2015-02-27 00:11:33 -0500 received badge  Supporter (source)
2015-02-26 18:04:29 -0500 commented answer Should the tenant router gateway be accessible from the network node itself ?

The following link was very useful - https://github.com/ChaimaGhribi/OpenStack-Icehouse-Installation/blob/master/OpenStack-Icehouse-Installation.rst#configure-compute-node (https://github.com/ChaimaGhribi/OpenS...)

2015-02-26 18:03:30 -0500 answered a question Should the tenant router gateway be accessible from the network node itself ?

Edit 2 - I found the solution. I had initially set a static ip for the external network on the network node because I needed internet access to install packages. So the bridge was not correctly configured. I deleted br-ex, br-tun and then added them as described in the manual. I then edited my interfaces file as shown below and added a static ip for br-ex. Restart the eth0 and br-ex interfaces, neutron services and possibly the openvswicth agent and service on the compute node as well. Delete old network and subnets and create new ones. I am now able to ssh into my instance and ping the external network from my instance. I do have one remaining issue, my instance can ping external ip addresses, but it cannot resolve names say "google.com" for example. Any suggestions ?

/etc/network/interfaces

# interfaces(5) file used by ifup(8) and ifdown(8)
auto lo
iface lo inet loopback


# External network

# The public network interface
auto eth0
iface eth0 inet manual
up ifconfig $IFACE 0.0.0.0 up
up ip link set $IFACE promisc on
down ip link set $IFACE promisc off
down ifconfig $IFACE down

auto br-ex
iface br-ex inet static
address 192.168.16.6
netmask 255.255.255.0
gateway 192.168.16.1
dns-nameservers 128.46.154.76 128.210.11.5 128.210.11.57

#Management Inetrface
auto eth1
iface eth1 inet static
address 10.0.0.21
netmask 255.255.255.0

# VM tunneling interface
auto eth2
iface eth2 inet static
address 10.0.1.21
netmask 255.255.255.0
2015-02-26 18:02:15 -0500 received badge  Editor (source)
2015-02-26 17:52:22 -0500 received badge  Popular Question (source)
2015-02-25 18:25:44 -0500 commented answer Should the tenant router gateway be accessible from the network node itself ?

Thanks a lot for the quick reply. I have added edits to the question according to your comment. Any pointers as to where to start debugging would be highly appreciated.

2015-02-25 01:05:35 -0500 asked a question Should the tenant router gateway be accessible from the network node itself ?

I am trying to setup the 3-node (controller, neutron and compute) openstack configuration on 3 Virtual machines running on Virtualbox. I have enabled promiscous mode on the external network as suggested in the manual. In the verify connectivity step - here ( http://docs.openstack.org/juno/install-guide/install/apt/content/neutron_initial-networks-verify.html (http://docs.openstack.org/juno/instal...) ), it says that "you should be able to ping the IP address of the tenant router gateway from any host on your external physical network." I am able to do that from the controller and compute nodes which are on the physical network. But when I try to do that from the network node (which has the tenant router on it) , it cannot ping it. Does this mean that there is an error in my setup ? If so, any pointers on how to troubleshoot it would be very welcome.

Edit 1 in reply to response - After configuring the network node as mentioned in the manual, I am able to launch an instance, but the network node no longer connects to the external network and so the instance is not able to as well. I have read through similar discussions on this forum and elsewhere but could not find a clear solution. I think I am not bridging the interfaces as they should be bridged. I am including my /etc/network/interface file of the network node as well as the outputs of some of the network-related commands that are often asked in such situations.

Configuration Detail - 3 VM's running Ubuntu 14.04 on VirtualBox. The management and tunnel networks are on eth1 and eth2 and are configured as internal networks. Static ip's are set on 10.0.0/24 and 10.0.1/24 respectively. The external network is on eth0 and is configured using the NatNetwork adapter that Virtualbox provides. ( https://www.virtualbox.org/manual/ch06.html#network_nat_service (https://www.virtualbox.org/manual/ch0...) ) . Basically the host and guests can see each other and the guest machines can access internet using the host (similar to NAT adapter in that aspect.). This works on the compute and controller nodes and used to work on the network node before I configured neutron setup on the network node. I set a static ip on this in the 192.168.16/24 subnet. A part of this is used for the virtual external network on the neutron node.