Anand's profile - activity

2014-08-08 00:53:35 -0600 received badge  Famous Question (source)
2014-06-18 04:08:27 -0600 received badge  Notable Question (source)
2014-06-18 04:08:27 -0600 received badge  Popular Question (source)
2014-02-17 04:14:27 -0600 asked a question Writing tests for Keystone ?


I am working on a project where i need to test Keystone for possible vulnerabilities. Currently, I am focussing on the keystone client middleware. I am interested in knowing the possible ways to test keystone components or in general OpenStack components. Any help on this would be appreciated.

On my part I am aware of the following things: 1. I have done the code review of the Keystone middleware to understand its working. 2. I follow Keystone related bugs on launchpad to further get a grip of things. 3. I am aware of Testr though I haven't used it yet.

However, I have still not written any code for testing Keystone, without it my understanding is still on the periphery. Therefore, are there any other tools / mechanisms / methodologies / dummy guides / blogs or miscellaneous resources that any one could enlighten me about . I would eventually like to start developing for OpenStack therefore I want to know as much i can :)

Thank you for your help !!!

2013-12-31 02:15:02 -0600 asked a question Is there a way to send information from a booted instance to neutron server ?


I'd like to know that is there a way for the booted instance to send network related data to the neutron server. For example, if i boot an instance and install a Network Intrusion detection system (NIDS) and will it be possible to send the information such as source IP address, source Port to the neutron server in case of an attack or malicious traffic.

The above case is too specific but what i want to know is that is there any way in which the instance can make contact with the neutron server to send some instance related information. Even if there is no implemented API for this it is alright :). I'd just like to know whether such an operation is possible and if so then how would it be done.

Thank you.

2013-11-25 04:00:36 -0600 received badge  Famous Question (source)
2013-11-13 04:00:11 -0600 received badge  Famous Question (source)
2013-11-06 04:39:16 -0600 received badge  Famous Question (source)
2013-11-04 06:58:33 -0600 received badge  Notable Question (source)
2013-11-01 20:51:15 -0600 received badge  Student (source)
2013-11-01 02:23:47 -0600 received badge  Supporter (source)
2013-11-01 02:23:44 -0600 received badge  Scholar (source)
2013-10-31 15:36:04 -0600 received badge  Popular Question (source)
2013-10-30 08:32:12 -0600 asked a question Create another private network on devstack single node ?


I want to boot a VM with multi-NIC on single node OpenStack. I have installed OpenStack Single using devstack and I am using the OVS plugin.

By default, there are two networks created by devstack: Private and Public.

I created another network called Private2 and when I launched an instance I assigned both the networks, Private and Private2, to the instance. The Private network is allocated to nic1 and Private2 to nic2. The Private2 network assigns an IP address to the virtual machine but when i do an ifconfig inside the virtual machine I am not able to see an interface with IP address allocated from Private2. There is only one interface that is shown and it belongs to the Private network not Private2. Also, ping and ssh does not work with IP address allotted from Private2 network.

The url shows the snapshot of my network topology. The admin status is up for the Interfaces for both the routers. Neutron net-list shell command also lists the Private2 network. Other than that, the ovs-vsctl show command also displays an interface corresponding to the IP allotted to the instance from the Private2 subnet.

The machine that i use to run OpenStack has a single physical NIC. Could that be the issue ?

Any suggestion or help in this regard would be appreciated. Thank you.

2013-10-30 02:42:59 -0600 received badge  Enthusiast
2013-10-29 08:03:53 -0600 commented answer ping: sendmsg: operation not permitted in Neutron when pinging instances

Thank you :)

2013-10-24 05:14:04 -0600 received badge  Notable Question (source)
2013-10-24 00:59:47 -0600 commented answer ping: sendmsg: operation not permitted in Neutron when pinging instances

This is what i have in the nova.conf : security_group_api = neutron firewall_driver = nova.virt.libvirt.firewall.IptablesFirewallDriver Are there any changes that i need to make to nova.conf ? I created a new security group with the above mentioned changes but the problem still persists.

2013-10-23 03:22:41 -0600 received badge  Popular Question (source)
2013-10-22 09:54:07 -0600 asked a question ping: sendmsg: operation not permitted in Neutron when pinging instances


I have installed single node openstack from devstack and booted two cirros machine instances. I am using OVS plugin for my work.

The instances are able to ping the outside world including public IP address as (facebook) and (google). However, when i ping from the host machine to the instances i get an error message ping: sendmsg: operation not permitted

The configuration is as follows:

Configuration of the router: Interface towards external bridge, br-ext, :

Interface towards integration bridge, br-int, :

I am able to ping the ip address and from the host machine.

Configuration of the Instances:

Instance #1 IP address:

Instance #2 IP address:

Default Security Group Rules:

Rule ALL TCMP and ALL ICMP Direction Ingress.

The machines are assigned fixed IP address from network and I am able to ping from the host machine.

According to route-n command The Gateway for an address in the network is

If anyone could help on this, i would really appreciate it. Thank you.

2013-10-18 22:13:43 -0600 received badge  Famous Question (source)
2013-10-16 03:47:35 -0600 answered a question What changed recently in Neutron's server code flow?


As pointed above, devstack has started using the ml2 plugin as the default plugin. However, to revert to openvswitch try the following procedure:

  1. Go to devstack folder and into the directory called lib.
  2. In the lib look for the file called neutron.
  3. In the file, look for the comment #Default Neutron Plugin followed by Q_PLUGIN=${Q_PLUGIN:-ml2}.
  4. Change the ml2 to openvswitch
  5. Run / as you normally do.

This will set up the devstack with OVS plugin. Hope it helps :)

2013-10-16 03:41:11 -0600 received badge  Notable Question (source)
2013-10-12 04:08:30 -0600 received badge  Popular Question (source)
2013-10-10 09:24:12 -0600 asked a question What changed recently in Neutron's server code flow?

Using devstack.

I would like to ask that is there any change in the code flow of the neutron server.

As per my understanding (assuming Neutron is running OVS plugin), when a curl request for example GET /v2.0/networks was made it was received by the file This would then route it to the controller file After that, it would go to the file. The last one being located in neutron/plugins/openvswitch directory.

The neutron services could be restarted with screen -r navigating to the selected services and then restarting them. But I think there has been a change in this flow. I tried to dig inside and found that the flow does not go through the ovs_neutron_plugin file. I commented the functions related to Show Network ,get_network(), and List networks get_networks(), then restarted the service again, using the procedure mentioned in the previous paragraph, however the request still got executed. Stack trace was also not of help as the code flow did not go to the functions get_network() and get_networks() .

If anyone is aware of this could you please update me about it, i would be thankful for your help.

2013-08-22 19:46:35 -0600 received badge  Notable Question (source)
2013-08-02 01:45:38 -0600 received badge  Famous Question (source)
2013-05-29 08:22:05 -0600 received badge  Popular Question (source)
2013-05-28 13:00:49 -0600 asked a question How to install Folsom single node from a script ?


I have searched a bit over the net can could not find a proper way to install folsom on a single node. I did try a couple of script but most of them install install Folsom barring Quantum or some of them are too confusing to understand

I would be thankful if someone could post links to site/ blog much like an idiots guide to programming or something from where i can install Folsom.

Thank you

2013-05-09 11:45:49 -0600 received badge  Notable Question (source)
2013-04-25 01:34:37 -0600 received badge  Popular Question (source)
2013-04-23 04:08:33 -0600 received badge  Famous Question (source)
2013-04-22 10:54:05 -0600 received badge  Famous Question (source)
2013-04-17 10:23:06 -0600 received badge  Notable Question (source)
2013-04-17 08:38:08 -0600 asked a question Handling Api requests in quantum ?


I want to know when a api call such as GET /v2.0/networks Accept: application/json is made how and where does the server interpret it and respond to it.

I have been looking in the source code and this is what i have found. The server file at starts the server.

Then the flow goes to the file (

Here the wsgi server is started at line # 104 and consequently the Quantum Service is started. All this is done with the aid of (

So now the quantum service has started. I checked the log information to verify.

What i want to know is if i make a request such as get network list GET /v2.0/networks Accept: application/json

what would be the flow of that. As asked above how would the Quantum server now read it interpret it and respond to it.

So far after this is what i have found out. In the (Same location as above) the request could go to Class Request (line # 266) As this would determine the most acceptable content-type, Accept* headers in the scenario described above.

Based on the request it would use the Router class (line # 904). As this, helps to maps incoming requests to WSGI app.

Then there is the ( This helps to form the response for resources like network, port and subnet. It has the resource attribute map (line # 475) which composes the response for network, port or subnet

Also there is the which has the class Api Router(line # 68). From its description it should route the api requests to its core functions.

Here is where i am lost. I don't have a clear idea how do these relate and work . Any help on this would be appreciated. Thank you.

2013-04-17 04:19:34 -0600 received badge  Taxonomist
2013-04-10 15:09:36 -0600 received badge  Popular Question (source)
2013-04-08 07:41:42 -0600 asked a question How to restart Networking service in Devstack single node?


I am new to quantum and i wanted to know how do i restart the quantum server in devstack single node installation.

Since restart will first require the server to be stopped i looked in to devstack/lib/quantum file. In this there is a function to stop the quantum server called stop_quantum().

Then to start there is a function in the same file called startquantumserviceandcheck().

basically, I am making some modifications to the extensions and i need to restart the server in this regard. This is the way i have come up with if there is a better alternative please let me know.

2013-04-06 11:44:51 -0600 received badge  Notable Question (source)
2013-04-05 13:52:09 -0600 received badge  Popular Question (source)
2013-04-02 06:34:54 -0600 received badge  Editor (source)
2013-04-02 06:34:54 -0600 edited question Quantum API Call Chain Mechanism

I want to know the exact call chain that occurs when an api call is made on quantum and how it is executed by the quantum server. For example, how is the quantum net-create net1 command executed What is the call chain that follows is something that i am looking for. Also, i would be interested in knowing the exact files that are used in this process. Thank you.