Ask Your Question

Vivek's profile - activity

2017-04-10 18:12:28 -0600 received badge  Famous Question (source)
2017-04-10 18:12:28 -0600 received badge  Notable Question (source)
2016-08-25 21:39:37 -0600 received badge  Famous Question (source)
2016-08-24 12:54:45 -0600 received badge  Notable Question (source)
2016-08-24 12:54:45 -0600 received badge  Popular Question (source)
2016-08-15 06:37:05 -0600 received badge  Popular Question (source)
2016-08-12 02:44:55 -0600 asked a question Mitaka: Nova CPU Overcommit is not taking effect in

Hi All,

Added the following lines to my nova.conf. However i dont see the overcommit is taking effect. Am I missing something here ?

cpu_allocation_ratio=16.0 scheduler_default_filters=CoreFilter

2016-05-18 02:02:23 -0600 asked a question Kilo: How to get the scheduling of VMs spread across multiple ESXI hypervisros in the case with vcenter as compute plugin

Hi,

I have my setup with Vcenter as compute plugin and 2 ESXI servers added to the vcenter dc/cluster managed by Openstack. I am using Contrail as my SDN. Everything is without issues.

However i created some 10 cirros VMs and i am seeing that all the VMs are spawned on single ESXI hypervisor. I understand that from openstacks point of view all it sees is the proxy nova-compute host which is managing the vcenter server, but is there a way or some setting I can make on the vcenter so that there is even distribution of VMs across multiple compute nodes ?

Thanks in advace, Regards, Vivek

2016-05-10 23:05:32 -0600 asked a question Openstack KILO : Scheduling of VMs based on availability zone is not working properly

Hello friends,

I have openstack KILO cluster configured with 3 availability zones 1. KVM 2.KVM-DPDK & 3. vmware1-az.

Tried booting VM from both horizon web-ui as well as from the nova command-line client. In both cases i am observing that VM scheduling is not happening properly i.e when i specified AZ as KVM the VM actually started booting on vmware1-az AZ (on ESXI) and vice versa. Same with the case is KVM-DPDK AZ. This doesn't happen all the time, sometime things work as expected but the behavior is not predictable.

Any known bugs in KILO or am i missing any configuration option ? Any specific log message i need to look for ?

Thanks in advance for the help.

Regards, Vivek

My availability-zone-list as displayed by nova command:

root@b7s16:/var/log/nova# nova availability-zone-list
+-----------------------+----------------------------------------+
| Name                  | Status                                 |
+-----------------------+----------------------------------------+
| internal              | available                              |
| |- b7s16              |                                        |
| | |- nova-conductor   | enabled :-) 2016-05-11T04:03:48.000000 |
| | |- nova-scheduler   | enabled :-) 2016-05-11T04:04:02.000000 |
| | |- nova-consoleauth | enabled :-) 2016-05-11T04:03:55.000000 |
| | |- nova-console     | enabled :-) 2016-05-11T04:04:02.000000 |
| |- b7s17              |                                        |
| | |- nova-conductor   | enabled :-) 2016-05-11T04:03:53.000000 |
| | |- nova-scheduler   | enabled :-) 2016-05-11T04:03:55.000000 |
| | |- nova-consoleauth | enabled :-) 2016-05-11T04:03:53.000000 |
| | |- nova-console     | enabled :-) 2016-05-11T04:03:54.000000 |
| |- b7s18              |                                        |
| | |- nova-conductor   | enabled :-) 2016-05-11T04:03:53.000000 |
| | |- nova-scheduler   | enabled :-) 2016-05-11T04:03:54.000000 |
| | |- nova-consoleauth | enabled :-) 2016-05-11T04:03:53.000000 |
| | |- nova-console     | enabled :-) 2016-05-11T04:03:53.000000 |
| KVM                   | available                              |
| |- b7s12              |                                        |
| | |- nova-compute     | enabled :-) 2016-05-11T04:03:48.000000 |
| |- b7s13              |                                        |
| | |- nova-compute     | enabled :-) 2016-05-11T04:03:58.000000 |
| KVM-DPDK              | available                              |
| |- b7s14              |                                        |
| | |- nova-compute     | enabled :-) 2016-05-11T04:03:56.000000 |
| |- b7s6               |                                        |
| | |- nova-compute     | enabled :-) 2016-05-11T04:04:00.000000 |
| vmware1-az            | available                              |
| |- b7s7               |                                        |
| | |- nova-compute     | enabled :-) 2016-05-11T04:03:59.000000 |
+-----------------------+----------------------------------------+
root@b7s16:/var/log/nova#
2016-03-15 22:39:37 -0600 received badge  Notable Question (source)
2016-01-18 22:46:54 -0600 received badge  Taxonomist
2016-01-01 10:12:17 -0600 received badge  Popular Question (source)
2015-12-12 01:56:54 -0600 asked a question "No valid host was found." error during booting of a instance

I am seeing the following error while trying to boot an nova instance. Is there something wrong with my setup ?

Thanks in advance, Vivek

{"message": "No valid host was found. ", "code": 500, "details": "  File \"/usr/lib/python2.7/dist-packages/nova/conductor/manager.py\", line 616, in build_instanc
    request_spec, filter_properties)                                                                                                                               
  File \"/usr/lib/python2.7/dist-packages/nova/scheduler/utils.py\", line 256, in wrapped                                                                          
    return func(*args, **kwargs)                                                                                                                                   
  File \"/usr/lib/python2.7/dist-packages/nova/scheduler/client/__init__.py\", line 50, in select_destinations                                                     
    context, request_spec, filter_properties)                                                                                                                      
  File \"/usr/lib/python2.7/dist-packages/nova/scheduler/client/__init__.py\", line 35, in __run_method                                                            
    return getattr(self.instance, __name)(*args, **kwargs)                                                                                                         
  File \"/usr/lib/python2.7/dist-packages/nova/scheduler/client/query.py\", line 34, in select_destinations                                                        
    context, request_spec, filter_properties)                                                                                                                      
  File \"/usr/lib/python2.7/dist-packages/nova/scheduler/rpcapi.py\", line 108, in select_destinations                                                             
    request_spec=request_spec, filter_properties=filter_properties)                                                                                                
  File \"/usr/lib/python2.7/dist-packages/oslo/messaging/rpc/client.py\", line 152, in call                                                                        
    retry=self.retry)                                                                                                                                              
  File \"/usr/lib/python2.7/dist-packages/oslo/messaging/transport.py\", line 90, in _send                                                                         
    timeout=timeout, retry=retry)                                                                                                                                  
  File \"/usr/lib/python2.7/dist-packages/oslo/messaging/_drivers/amqpdriver.py\", line 417, in send                                                               
    retry=retry)                                                                                                                                                   
  File \"/usr/lib/python2.7/dist-packages/oslo/messaging/_drivers/amqpdriver.py\", line 408, in _send                                                              
    raise result                                                                                                                                                   
", "created": "2015-12-12T07:47:39Z"}                                                                                                                              
m1.tiny (1)
2015-06-17 16:21:18 -0600 received badge  Famous Question (source)
2015-06-05 23:41:22 -0600 received badge  Notable Question (source)
2015-06-05 01:07:42 -0600 received badge  Popular Question (source)
2015-06-04 16:58:28 -0600 asked a question Incoming traffic from the VM is blocked by iptables altough the security group is configured to permit the traffic

Hello ,

The openstack instance i am running is a network function (virtual router to be precise) and hence this instance supposed to forward traffic received from one neutron network to the other network neutron network.

I have configured a security group to allow all the tcp & udp traffic in both ingress & egress directions and have applied to the instance. However when i check the iptables chain it appears that the the chain neutron-openvswi-s<xxxxxx>-<y> in the FORWARD table which actually controls the traffic coming from the instance to the linux bridge is only allowing the traffic from the specific interface IP and MAC assigned by neutron and not allowing all the tcp or udp traffic as I was expecting it to do.

Am i missing something here or is this a bug/limitation. Also is there some kind of work around to make this work ?

Regards, Vivek

[root@rbu-rpd-ns1 ~(keystone_admin)]# neutron security-group-rule-list | grep -v default
+--------------------------------------+----------------+-----------+----------+------------------+--------------+
| id                                   | security_group | direction | protocol | remote_ip_prefix | remote_group |
+--------------------------------------+----------------+-----------+----------+------------------+--------------+
| 51810933-2091-41f1-ad93-b6f97238cda2 | test           | egress    | tcp      | 0.0.0.0/0        |              |
| 69e5cf3d-0b3e-42e6-b9cc-75c68f2085df | test           | ingress   | tcp      | 0.0.0.0/0        |              |
| c67db669-6f6c-4733-a120-58eb43e95d1e | test           | egress    |          | 0.0.0.0/0        |              |
| d89a0568-f951-40ea-98b6-328f3f85826f | test           | ingress   | udp      | 0.0.0.0/0        |              |
| dbde6471-8411-4a9a-ad6a-3538aee0aafd | test           | ingress   |          | 0.0.0.0/0        |              |
| fbb993c2-9ca0-45e8-be28-0614f7b419e3 | test           | egress    | udp      | 0.0.0.0/0        |              |
+--------------------------------------+----------------+-----------+----------+------------------+--------------+
[root@rbu-rpd-ns1 ~(keystone_admin)]# nova show rtr1 | grep security
| security_groups                      | test                                                     |

[root@rbu-rpd-ns1 ~(keystone_admin)]#

iptable chain :

Chain neutron-openvswi-s744fb19c-1 (1 references)
num      pkts      bytes target     prot opt in     out     source               destination         
1           0        0 RETURN     all  --  *      *       10.10.10.6           0.0.0.0/0            MAC FA:16:3E:89:90:E2
2           0        0 DROP       all  --  *      *       0.0.0.0/0            0.0.0.0/0
2015-05-28 05:20:08 -0600 received badge  Notable Question (source)
2015-05-28 00:15:21 -0600 received badge  Popular Question (source)
2015-05-27 19:48:10 -0600 answered a question Change the neutron tunnel mode of a operational openstack system

2nd try:

Just to be bit more clear on the question I have a working multinode openstack cluster using vxlan as the overlay. Now I want to add a second provider network for vlan networking. How can add the vlan tenant networking without re-installing the openstack all over again ?

Regards, Vivek

2015-05-27 01:50:34 -0600 asked a question Change the neutron tunnel mode of a operational openstack system

Hello all,

I already have a openstack cluser UP & running with VXLAN as the networking overlay for VM-VM overlays. I want to change the overlay from VXLAN to VLAN (or add a second overlay which is VLAN using provider networks). Is there a way to make these changes without building the openstack system again ? What are the steps to change the neutron configuration without affecting the other services or bringing down the system ? Thanks in advance, Regards, Vivek

2015-05-20 01:20:30 -0600 received badge  Famous Question (source)
2015-05-15 05:44:12 -0600 received badge  Famous Question (source)
2015-05-15 05:44:12 -0600 received badge  Notable Question (source)
2015-05-15 05:44:12 -0600 received badge  Popular Question (source)
2015-05-10 22:33:15 -0600 received badge  Famous Question (source)
2015-05-10 06:23:06 -0600 answered a question Accessing the VM from internet

First of all thanks for both the replies.

@ sunnyarora : I have already enabled forwarding in linux kernel and everything else (other traffic flows like inter-vm, vm to internet ) are working fine.

@ YanivZadka2: Does adding a nova sec group rule gonna make any difference here ? Since I am using JUNO with neutron does having a nova network (this sec rule for instance) configuration has any effect ?

2015-05-09 14:32:49 -0600 received badge  Notable Question (source)
2015-05-09 04:50:54 -0600 received badge  Popular Question (source)
2015-05-08 19:06:33 -0600 asked a question Accessing the VM from internet

Hello All,

In my single node setup (Openstack JUNO on CENTOS 6.5 running with neutron-ovs for networking) which is functional, I am able to reach the internet from the virtual-machine using floating-ip and everything seems to be working fine.

But I want to access the VM from the outside network and I tried testing that by pinging the floating-ip and I see that the reachability is not there. So what I would like to know is

  1. Is this a expected behaviour and do I need to turn on something else ?
  2. Am I missing something , in which case what are the common things I can look for ?

Regards, Vivek

2015-05-08 19:01:58 -0600 asked a question ERROR (HTTPNotImplemented): Network driver does not support this function. (HTTP 501) message seen on running the "nova secgroup-list-default-rules" command

I have a single server installation of Openstack JUNO with neutron-ovs networking plugin running on CENTOS 6.5

Getting this error whenever I try to list the security groups. Am I missing something here ?

[root@rbu-rpd-ns4 ~(keystone_admin)]# nova secgroup-list-default-rules ERROR (HTTPNotImplemented): Network driver does not support this function. (HTTP 501) (Request-ID: req-145488dc-a193-42df-bc64-5aac33bee8b1) [root@rbu-rpd-ns4 ~(keystone_admin)]#

2015-04-01 15:16:56 -0600 received badge  Notable Question (source)
2015-04-01 15:16:56 -0600 received badge  Famous Question (source)
2015-02-24 05:16:22 -0600 received badge  Popular Question (source)
2015-02-11 14:17:27 -0600 received badge  Famous Question (source)
2015-02-09 11:13:39 -0600 received badge  Nice Question (source)
2015-02-09 02:18:00 -0600 asked a question Not able to create the heat stack with resource type : OS::Neutron::ProviderNet

Hi,

Can some one please tell me what is wrong with the following stack ? The stack creation is failing !! However if I don't include one of "type: OS::Neutron::ProviderNet" or "type: OS::Neutron::Subnet" sections things seems to be working.

Thanks in advance, Vivek

description: ' HOT template to create networks that can be attached to vPFE ports.'
heat_template_version: '2013-05-23'
parameters:
  mgmt_net_name:
    default: mgmt_net
    description: Network name for the external management network
    label: Network Name
    type: string
  mgmt_vlan_id:
    default: '301'
    description: Management Network Vlan Id extended to external Switch
    label: Vlan Id
    type: string
resources:
  mgmt_net:
    type: OS::Neutron::Net
    properties:
      name: { get_param: mgmt_net_name }
      admin_state_up: true
    type: OS::Neutron::ProviderNet
    properties:
      admin_state_up: 'True'
      name: { get_param: mgmt_net_name}
      network_type: vlan
      physical_network: mynetwork
      segmentation_id: { get_param : mgmt_net_name }
  private_subnet_0:
    type: OS::Neutron::Subnet
    properties:
      name: { get_param: mgmt_net_name}
      network_id: { get_resource : mgmt_net }
      cidr: 192.168.64.0/24
      enable_dhcp: False
      gateway_ip: 192.168.64.254
2015-01-20 13:30:43 -0600 received badge  Famous Question (source)
2015-01-19 20:48:44 -0600 received badge  Nice Question (source)
2015-01-19 02:21:40 -0600 received badge  Notable Question (source)
2015-01-19 02:04:30 -0600 received badge  Enthusiast
2015-01-19 00:33:00 -0600 received badge  Popular Question (source)
2015-01-18 23:30:10 -0600 received badge  Notable Question (source)
2015-01-18 15:31:07 -0600 received badge  Student (source)
2015-01-18 03:08:22 -0600 received badge  Popular Question (source)
2015-01-17 19:10:29 -0600 asked a question To create an L2 only neutron, network

Hi,

Is it possible to create an an L2 only neutron network ?

Thanks

2015-01-17 19:08:38 -0600 asked a question Does neutron (br-int) accepts nova instances sending vlan tagged traffic ?

My use case involves sending vlan tagged traffic between the instances on the same hypervisor and across hypervisor. How does neutron handles vlan tagged traffic from instances ? Do I need to have a special setting for this to work ?

2015-01-15 19:21:52 -0600 received badge  Notable Question (source)
2015-01-09 17:39:59 -0600 received badge  Famous Question (source)