Ask Your Question

Kashyap Kopparam's profile - activity

2015-01-28 04:44:15 -0600 received badge  Student (source)
2014-12-09 04:44:16 -0600 answered a question devstack multi-node lab compute node fails

I solved this issue by using devstack-vagrant https://github.com/openstack-dev/devs... . Well I dint solve the issue, but got a workaround. :)

2014-12-08 05:58:59 -0600 received badge  Famous Question (source)
2014-12-08 05:58:03 -0600 commented question devstack multi-node lab compute node fails

Updated local.conf

2014-12-08 01:13:18 -0600 commented answer devstack multi-node lab compute node fails

I often run into this problem as well. But the stack fails much before this problem occurs.

2014-12-05 03:21:37 -0600 received badge  Notable Question (source)
2014-12-04 21:17:08 -0600 received badge  Popular Question (source)
2014-12-04 05:13:19 -0600 asked a question devstack multi-node lab compute node fails

I am trying to setup the mutil-node devstack setup http://docs.openstack.org/developer/d... . I can install devstack on the controller node just fine. But while adding a compute node, devstack fails with

2014-12-04 10:43:31.826 | + cinder type-create lvmdriver-1
2014-12-04 10:43:32.127 | ERROR: Unable to establish connection to http://127.0.0.1:5000/v2.0/tokens
2014-12-04 10:43:32.148 | ++ err_trap
2014-12-04 10:43:32.148 | ++ local r=1
2014-12-04 10:43:32.148 | stack.sh failed: full log in /opt/stack/logs/stack.sh.log.2014-12-04-161234
2014-12-04 10:43:32.148 | Error on exit

I have made sure that the endpoints in the controller have routable IPs, but for some reason the keystone client on the compute node thinks the auth uri is the loopback ip. How can change it to have the routable ip?

I have been reproducing this error multiple times with different approaches. I tried having the two nodes in a private network of vagrant boxes, tried with two physical machines running Ubuntu. But I'm not able to solve this issue. Any inputs are greatly appreciated.

Controller local.conf

[[local|localrc]]
HOST_IP=10.4.1.255
FLAT_INTERFACE=eth0
FIXED_RANGE=192.168.0.0/20
FIXED_NETWORK_SIZE=4096
FLOATING_RANGE=10.4.2.0/28
# Credentials
DATABASE_PASSWORD=secrete
ADMIN_PASSWORD=secrete
SERVICE_PASSWORD=secrete
SERVICE_TOKEN=secrete
RABBIT_PASSWORD=secrete
MULTI_HOST=1

# Enable Logging
LOGFILE=/opt/stack/logs/stack.sh.log
VERBOSE=True
LOG_COLOR=True
SCREEN_LOGDIR=/opt/stack/logs
RECLONE=yes

Compute local.conf

[[local|localrc]]
HOST_IP=10.4.1.208
FLAT_INTERFACE=eth0
FIXED_RANGE=192.168.0.0/20
FIXED_NETWORK_SIZE=4096
FLOATING_RANGE=10.4.2.0/28
MULTI_HOST=1
LOGFILE=/opt/stack/logs/stack.sh.log
ADMIN_PASSWORD=secrete
MYSQL_PASSWORD=secrete
RABBIT_PASSWORD=secrete
SERVICE_PASSWORD=secrete
SERVICE_TOKEN=secrete
DATABASE_TYPE=mysql
SERVICE_HOST=10.4.1.255
MYSQL_HOST=10.4.1.255
RABBIT_HOST=10.4.1.255
GLANCE_HOSTPORT=10.4.1.255:9292
ENABLED_SERVICES=n-cpu,n-net,n-api,c-sch,c-api,c-vol
NOVA_VNC_ENABLED=True
NOVNCPROXY_URL="http://10.4.1.255:6080/vnc_auto.html"
VNCSERVER_LISTEN=$HOST_IP
VNCSERVER_PROXYCLIENT_ADDRESS=$VNCSERVER_LISTEN
2014-12-02 04:23:33 -0600 answered a question What is default timezone of Ceilometer ?

All timestamps are stored in the database in UTC timezone.

2014-12-01 22:57:44 -0600 commented question problem with keystone-manage db_sync

The 'keystonedbadmin' user is being denied access for the admin action db_sync. You can source an admin user's credentials and then try to db_sync.

2014-11-27 08:57:22 -0600 commented question Failed to get metadata for ip :

so what are you using kishan?

2014-11-27 04:11:14 -0600 received badge  Good Answer (source)
2014-11-27 04:10:02 -0600 received badge  Nice Answer (source)
2014-11-27 04:09:33 -0600 received badge  Nice Answer (source)
2014-11-27 04:09:19 -0600 received badge  Nice Answer (source)
2014-11-27 00:46:16 -0600 received badge  Famous Question (source)
2014-11-26 04:12:13 -0600 answered a question sub-process /usr/bin/dpkg returned an error code (1)

Pradeep you have to run the apt-get command with --configure flag. The nova-common and nova-api packages in your machine are not configured to get the dependencies. Please run $ sudo apt-get install nova-api nova-common --configure.

PS: Please copy the text in the terminal and post it here as code text, it is easier to read, less data to download for us and you can post more text than what your screen can display at once :).

2014-11-26 03:58:56 -0600 commented answer OpenStack juno packages for ubuntu 14.04

@Garysday: I'm not sure if adding the trusty-updates and then installing glance with cloud-archive would do any good. And yes, there are a whole lot of gaps in the docs, which calls for people like us to fill em up. Please consider contributing to the docs whenever possible. Thanks.

2014-11-25 16:52:29 -0600 received badge  Notable Question (source)
2014-11-25 16:45:33 -0600 received badge  Popular Question (source)
2014-11-25 03:10:52 -0600 answered a question OpenStack juno packages for ubuntu 14.04
  1. The package configuration is needed for all nodes. This will enable openstack packages to be installed, like keystone, nova etc. All nodes will have at least one package from openstack.
  2. I have used cloud-archive for juno and it works just fine. Is there a particular reason you are using the staging workgroup?
2014-11-25 00:35:03 -0600 commented question Single machine all in one installation-instance internet access problem

You can use any interface for openstack. Using wlan0 for openstack would surely solve your problem of internet access. If you want to continue using eth0, you might want to look into MASQUERADE rules on your firewall which would translate packets on an interface for internet access.

2014-11-25 00:28:17 -0600 answered a question deploy icehouse openstack on single over a LAN

You can use Devstack http://docs.openstack.org/developer/d... to have all of openstack on one machine. You could also use RDO Packstack https://openstack.redhat.com/Quickstart on a single machine.

2014-11-25 00:06:33 -0600 received badge  Scholar (source)
2014-11-24 06:26:39 -0600 received badge  Critic (source)
2014-11-24 06:11:39 -0600 commented question Launching instance. Spawning state

OK. You are using libvirt driver. Try to launch multiple instances and see if any of them get till the active state. The libvirt driver is looking for an instance in the hypervisor(yours can be qemu or kvm), but is not finding it.

2014-11-24 05:44:44 -0600 commented question Launching instance. Spawning state

9lives' command showed you the error you were looking for. You were missing the -i flag in grep command, else you would have seen the error before you even posted here :P. Ignore my requested logs.

2014-11-24 05:08:33 -0600 commented question Launching instance. Spawning state

Pablo, look at the instance log from the dashboard. If you are using a proper OS like Fedora or Ubuntu's image to spawn instances on a QEMU compute-host, it will take like 20 minutes for it to finish booting up. please post the instance logs and nova-compute logs.

2014-11-24 04:28:26 -0600 commented question Instance failed to spawn problem

Please search before posting. The error you are facing is NovaException: Unexpected vif_type=binding_failed. Your situations seems to be a duplicate of https://ask.openstack.org/en/question...

2014-11-24 04:22:23 -0600 asked a question Not able to use OS::Heat::MultipartMime resource type in Heat.

I have this template which uses OS::MultipartMime to combine the CloudConfig and SoftwareConfig resource types. But when I try to run this or any template with heat I run into errors and the stack fails. Here is the the full trace of the error that I see in heat engine. I have tried multiple MultipartMime examples but all of them fail like this. Has anyone seen this before, found a workaround, or just have any clue how I can get around this for now?

2014-11-24 15:49:49.562 INFO heat.engine.stack [-] Stack CREATE FAILED (simple): Resource CREATE failed: TypeError: object of type 'NoneType' has no len()
Traceback (most recent call last):

  File "/usr/local/lib/python2.7/dist-packages/oslo/messaging/rpc/dispatcher.py", line 134, in _dispatch_and_reply
    incoming.message))

  File "/usr/local/lib/python2.7/dist-packages/oslo/messaging/rpc/dispatcher.py", line 177, in _dispatch
    return self._do_dispatch(endpoint, method, ctxt, args)

  File "/usr/local/lib/python2.7/dist-packages/oslo/messaging/rpc/dispatcher.py", line 123, in _do_dispatch
    result = getattr(endpoint, method)(ctxt, **new_args)

  File "/usr/local/lib/python2.7/dist-packages/osprofiler/profiler.py", line 105, in wrapper
    return f(*args, **kwargs)

  File "/opt/stack/heat/heat/engine/service.py", line 72, in wrapped
    return func(self, ctx, *args, **kwargs)

  File "/opt/stack/heat/heat/engine/service.py", line 1409, in show_software_config
    sc = db_api.software_config_get(cnxt, config_id)

  File "/opt/stack/heat/heat/db/api.py", line 258, in software_config_get
    return IMPL.software_config_get(context, config_id)

  File "/opt/stack/heat/heat/db/sqlalchemy/api.py", line 706, in software_config_get
    result = model_query(context, models.SoftwareConfig).get(config_id)

  File "/usr/lib/python2.7/dist-packages/sqlalchemy/orm/query.py", line 799, in get
    if len(ident) != len(mapper.primary_key):

TypeError: object of type 'NoneType' has no len()
2014-11-24 03:14:11 -0600 edited answer Devstack installation fails with a message:ERROR: openstackclient.shell Exception raised: requests>=1.1

You can try to upgrade the requests package using pip and then run stack.sh again.

$ pip install --upgrade requests setuptools

$ ./stack.sh

2014-07-24 07:03:47 -0600 answered a question Is it possible to create user defined stack

Yes, you can write your own template and create a customised stack. Please look at HOT Specification for the available resources.

2014-07-24 03:32:03 -0600 commented question NoVNC special keys

If you are using Linux on the host machine you can use alt codes. just type alt+0123 and alt+0125 (i think) for the curly braces. On a mac you can use the on screen keyboard to insert those characters.

2014-07-24 00:53:15 -0600 commented question nova list shows empty table

The one reason I can think of is: you have created the VMs on Horizon by logging in with one user say "admin". And you are running nova list after sourcing another user say "demo" user's credentials. Please check with which user you have created the VMs and try nova list with the same user.

2014-07-24 00:41:29 -0600 commented question cannot ping external gateway

Can you ping the host on which the virtual router is configured. If so, you might need to add a MASQUERADE rule to the firewall running on the host machine which will allow traffic from the neutron network to the external network. I'm sorry for not providing the exact rule that you need to add. :) It could be something like

 iptables -t nat -I POSTROUTING 1 -s 1 192.48.1.0/24 -o eth0 -j MASQUERADE

Reading Fragmented floating IP pools and multiple AS hack may also help.

2014-07-24 00:36:44 -0600 commented question NoVNC special keys

All special keys work just fine for me. You might be using something like Vimium that is causing those keys to not work on VNC console. Try using a different browser with no plugins.

2014-07-18 07:22:44 -0600 commented answer Invalid user token - rejecting request

Sorry man, beats me. Also, I kindly suggest you follow the official documentation of OpenStack. Please up vote if I helped you (implied).

2014-07-16 05:50:23 -0600 commented answer Invalid user token - rejecting request

You cannot escape that. Swift was designed to serve objects with authentication only.

2014-07-15 07:25:13 -0600 commented answer Invalid user token - rejecting request
  • I thought you said client tool would work, but I see a failure on the last line.
  • As you can see in this line

    INFO:swiftclient:REQ: curl -i http://127.0.0.1:8080/v1/AUTH_918112e... -X GET -H "X-Auth-Token: ADMIN"

the curl request has a header which contains the thing after -H. This will not be present if you simply paste the url on browser.

2014-07-15 04:48:47 -0600 commented question Invalid user token - rejecting request

You can run the command using the client tools with the --debug option. This will output the URL that the command uses to query the object storage. I believe that it has some parameters like the keystone token that it adds to the POST request. Just entering that URL on the browser is expected to give an auth error.

2014-07-15 04:42:12 -0600 received badge  Commentator
2014-07-15 04:42:12 -0600 commented answer stop: Unknown job: keystone

you obviously have missed the part which installs the keystone packages. Please run

$ apt-get install keystone --reinstall
2014-07-15 04:39:36 -0600 commented question Deploy Vagrant's box to OpenStack's Nova

Vagrant uses port forwarding to give you access to it. So I guess whatever you need to put into nova you can just go ahead with it. But you will need to add more ports to forward to access your apps from the host. Also, you need to give more info on what you are trying to do to get better answers.

2014-07-15 01:26:08 -0600 commented answer Instance Error, caused by libvirtError: internal error no supported architecture for os type 'hvm ??

This affect me as well. But I don't seem to understand why this happens even when I'm running it directly on a Linux machine which supports virtualization. I ran this according to the install guide and it returns 4.

$ egrep -c '(vmx|svm)' /proc/cpuinfo

So I still do not know why this happens to me when I try to use KVM.

2014-06-20 00:04:03 -0600 received badge  Good Answer (source)