Ask Your Question

PradeepNaik's profile - activity

2019-05-06 23:44:51 -0500 received badge  Nice Answer (source)
2017-05-26 07:01:11 -0500 received badge  Good Answer (source)
2016-07-06 05:49:09 -0500 received badge  Nice Answer (source)
2015-12-06 09:05:50 -0500 received badge  Nice Answer (source)
2015-11-12 10:56:47 -0500 received badge  Nice Answer (source)
2015-06-29 07:01:36 -0500 answered a question keystone: Unable to establish connetion in ubuntu 14.04

your keystone endpoint may have some issues, which you might be incorrectly provided.

post your logs from /var/logs/keystone/***.log

2015-06-29 06:51:49 -0500 edited question keystone: Unable to establish connetion in ubuntu 14.04

Hi All,

While installing openstack juno keystone on ubuntu 14.04 I am getting an error "unable to establish connection to http://localhost: port/....".

In a fraction of seconds keystone service gets stop/waiting,it shows a different PID and when rechecking afterwards, it's dead.

root@hyper5:~# service keystone restart  
stop: Unknown instance:  
keystone start/running, process 37746  
root@hyper5:~# service keystone status  
keystone start/running, process 37750  
root@hyper5:~# service keystone status  
keystone stop/waiting

What all the parameters I need to check before exporting keystone tokens?? In My Keystone.conf file following parrameters are set.

[DEFAULT]
admin_token=0b4a93b7d3ebac2f5666
verbose = True ... 

[database]
connection = mysql://keystone:secret@localhost/keystone
2015-06-23 02:28:52 -0500 answered a question How to build OpenStack from scratch

OpenStack breaks the boundaries of management of compute , storage and networking etc that are the essential building blocks of IT Infrastructure.As an OpenStack expert you need familiar with visualization, linux networking, Storage technologies and more. This might be a long journey, be patient.:-)

In my personal view, this journey might be covered in 5 phases:

Phase1:

Try OpenStack first. you should be able to deploy openstack single node mode(devstack recommended) to get a little sense of what it is and how to use it and learn how to troubleshooting during installation.

Phase2:

Familiar with the architecture.Read official documents in http://docs.openstack.org/ and ask and helping stackers resolve issues in ask.openstack.org

Phase3:

Learn the core technologies that OpenStack leveraged. OpenStack is an orchestration framework which integrating the existing and newest technologies via a uniform framework(manager -> driver) and exposes the features via restful api.You need to master various technologies like virtualization(libvirt, kvm, xen, docker etc) , networking(openvswitch,tcp/ip, etc), storage(iscisi,lvm,etc) etc.

Phase4:

Get yourself comfortable with Python. OpenStack mainly written in Python,if you want to be an contributor or extend the existing features of OpenStack, you must know Python and the python libraries that OpenStack utilizes.

Phase5:

Push OpenStack in production.At this phase, you have loaded enough experiences and knowledge in OpenStack, time to leverage the power of OpenStack in production.

One more thing, the OpenStack provides a great training guide http://docs.openstack.org/training-gu... for different roles this might be the first article you need to take a close look.

Hope that helps!

2015-06-23 02:23:37 -0500 commented answer start openstack after restarting system

running ./stack.sh after every reboot not recommended because it will clears all the previous data. so if you want to retain the previous state of your openstack then run ./rejoin-stack.sh

2015-06-23 02:18:40 -0500 answered a question start openstack after restarting system

hi,

If you are using devstack setup, and after restarting your system if you want to retain the previous state of openstack then run ./rejoin-stack.sh in your terminal.

if you run ./stack.sh then it will clear all the previous instances, database entries,etc.

for more information about devstack after rebooting

2015-06-13 05:12:42 -0500 answered a question Horizon log in fails by "Something went wrong!" error

the physical ethernet connection to controller from neutron may be disconnected or your neutron-server service may be down.

first check wheather you are able to ping between neutron & controller.

then check wheather messaging service is running or not.

2015-06-12 00:15:49 -0500 commented answer neutron qrouter-<ID> is not created in network namespace

hello dpaxps what you said is the routine procedure available in installation guide. plz check the image where status remained DOWN soon after creating it.

2015-06-11 08:19:11 -0500 answered a question How to SSH a vm

you mean to say that you want syntax or any troubleshooting assistance while doing it.

anyway following is the command. first of all launch vm with your choice of image which you done already. add rules to it. like,

nova secgroup-add-rule default icmp -1 -1 0.0.0.0/0
nova secgroup-add-rule default tcp 22 22 0.0.0.0/0

for cirros instance.

ssh cirros@INSTANCE_FIXED_IP/INSTANCE_FLOATING_IP

for ubuntu instance..

1. nova keypair-add OSKEY > OSKEY.priv (where OSKEY is the one you generated from ssh-keygen command)
2. chmod 600 OSKEY
3. now launch the instance with the OSKEY as the key
4. ssh -i OSKEY.priv ubuntu@floatingIP

hope that helps

2015-06-05 04:41:16 -0500 answered a question nova list is giving empty list but nova list --all-tenants is not empty, why?

this is because your VMs are in demo-tenant or may be in other one you created but not admin.

from admin tenant you see all the tenants. But you created instances in demo tenant.

now source demo-openrc.sh and check then do nova list.

2015-06-05 00:58:17 -0500 commented answer neutron qrouter-<ID> is not created in network namespace

Thank you for reply. neutron router-list is working fine, floating IPs assigned, VM also ACTIVE & RUNNING. only issue is ip netns show it must display a qrouter-*******. if a virtual router exists. that is not happening why?

2015-06-05 00:43:12 -0500 received badge  Scholar (source)
2015-06-05 00:43:03 -0500 commented answer neutron qrouter-<ID> is not created in network namespace

i done this setup earlier but this error i didnt faced. My pointer to issue is with L3-agent but logs are not helpful over here.

2015-06-05 00:41:18 -0500 commented answer neutron qrouter-<ID> is not created in network namespace

It is according installation guide. two tenant networks created both have few vms. And i even created the router, status is also active and attached these two tenant networks to it.The interface attached to the router status is down. Also in ip netns a qrouter-**must be there which is also missi

2015-06-04 09:10:20 -0500 answered a question configuring demo-net for tenant network

you are not supposed to use 192.168.1.1 to your tenant network gateway because it is already assigned to your physical router. But this is not true always sometimes giving 192.168.1.1 also works for your tenant network gateway.

but i prefer to use 192.168.1.2 as gateway and things will work fine

2015-06-04 09:01:14 -0500 answered a question cirros instance cannot get ip address

hello there,

technically no ip assigned to vm here problem is with your tunnel interface. if you type nova list your vm will be be ACTIVE & RUNNING also an IP will be displayed but internally it is not, This is technically a bug according to me.

if you type nova console-log <instance-id> here there will be no IP.

for every network creation a namespace will be created from that also you can check

ip netns exec qdhcp-****** ifconfig

openstack installation guide clearly states that you need to use two different physical interfaces for management and tunnel network. if this step is correct then IP will be assigned to your VM

2015-06-04 08:28:36 -0500 answered a question Nova boot : Error spawning a VM

from your log file we can see that Address already assigned it is telling, so do check the address pool you created.

2015-06-04 08:25:32 -0500 answered a question how to ping the instance in other network segment

hello,

i believe you created a router and attached interfaces of 20.20.20.* and 30.30.30.* network. if these two are pinging then each other then your router is fine.

check from host machine i.e. from 10.10.10.5 type ip netns show from a network node

this will return few values like qdhcp-* for your network and qrouter- for your router.

you can ping like ip netns exec qrouter-********** ping 20.20.20.*

if this all works fine then create br-ex and attach public network gateway to it. you can check the details in network topology from dashboard. after creating br-ex you can directly ping to you neutron instances without using namespace.

2015-06-04 08:06:53 -0500 answered a question Neutron: br-tun not working on compute node

hello there,

i gone through with this issue, what worked for me is this.

as installation guide clearly tells we need to maintain two interface like eth0 & eth1.

eth0 for management network and eth1 for tunnel network. i hope you you did same thing. if i type

root@controller:/etc/modprobe.d# ovs-vsctl show
599ec640-7eda-4d4c-a387-75e97bf135c5
    Bridge br-int
        Port "qvo4ba01e0a-34"
            tag: 2
            Interface "qvo4ba01e0a-34"
        Port patch-tun
            Interface patch-tun
                type: patch
                options: {peer=patch-int}
        Port "qvo21f7f45e-51"
            tag: 1
            Interface "qvo21f7f45e-51"
        Port br-int
            Interface br-int
                type: internal
        Port "qvo1c630b2f-c6"
            tag: 1
            Interface "qvo1c630b2f-c6"
    Bridge br-tun
        Port patch-int
            Interface patch-int
                type: patch
                options: {peer=patch-tun}
        Port br-tun
            Interface br-tun
                type: internal
        Port "gre-14140003"
            Interface "gre-14140003"
                type: gre
                options: {in_key=flow, local_ip="20.20.0.2", out_key=flow, remote_ip="20.20.0.3"}
    ovs_version: "2.0.1"

clearly tunneling path is a pyhsical path using physical switch i used. so instead of veth pair use physical NIC for br-tun.

2015-06-04 07:43:27 -0500 answered a question No valid host was found

no valid hosts error comes for many reasons, kindlly follow the link and go through the troubleshooting steps.

No Valid Hosts

2015-06-04 07:12:11 -0500 received badge  Famous Question (source)
2015-06-03 03:57:54 -0500 received badge  Notable Question (source)
2015-06-02 08:46:54 -0500 received badge  Popular Question (source)
2015-06-02 02:15:40 -0500 asked a question neutron qrouter-<ID> is not created in network namespace

image description hello stackers,

i created a virtual router and connected the router interface to two networks.

router is created and i can see it in neutron router-list but when i type ip netns list i cannot see a qrouter-<ID>

Can somebody help me?

2015-05-31 02:11:10 -0500 answered a question Number of GRE tunnels in DVR

hi,

please do check that have you given management IP and Tunnel IP as in a single network. ovs-vsctl show command can put some better light on the OVS_Bridge.

And tunneling is done only during initial configuration. dont bother about tunneling during instance creation as it is an internal process.

check this link, hope this may help you

https://blogs.oracle.com/ronen/entry/...

2015-05-27 07:28:47 -0500 edited question glance images folder permission denied

Hello there,

I encountered a weird error while upgrading glance to kilo. Whenever I start up glance service, it encountered an error (in the log file) saying the follow: "

2015-05-26 20:43:40.261 4744 ERROR glance_store._drivers.filesystem [-] Permission to write in /var/lib/glance/images/ denied
2015-05-26 20:43:40.261 4744 TRACE glance_store._drivers.filesystem None
2015-05-26 20:43:40.261 4744 TRACE glance_store._drivers.filesystem 
2015-05-26 20:43:40.262 4744 WARNING glance_store.driver [-] Failed to configure store correctly: Store filesystem could not be configured correctly. Reason: Permission to write in /var/lib/glance/images/ denied Disabling add method.
"

However, I do not know why this is happening because it seems like "glance-api" is run by "glance" and "/var/lib/glance/images/" is also owned by "glance". It appears to me that "glance" should have sufficient privilege of accessing (writing) to the target folder because :

drwxrwxr-x 2 glance glance 4096 May  5 22:36 images

glance    4754  4744  0 20:43 ?        00:00:00 /usr/bin/python2 /usr/bin/glance-api

I tried to use chmod to change images' access privilege but I failed by using either "root" or "glance".

Could anyone please give me some suggestions regarding how to get this issue resolved?

Thanks,

2015-05-25 02:33:39 -0500 received badge  Good Answer (source)
2015-05-25 00:39:16 -0500 answered a question ERROR: openstack The request you have made requires authentication. (HTTP 401)

hello,

please check these values are available in environment env

export OS_TOKEN=ADMIN_TOKEN
export OS_TOKEN=294a4c8a8a475f9b9836
export OS_URL=http://controller:35357/v2.0

and these values are local to the terminal.

2015-05-25 00:33:49 -0500 answered a question Unexpected vif_type=binding_failed

hi,

even i encountered this problem after all the debugging what solved problem for me is the following.

ovs-vsctl show   this command returns some value like `br-int`, `br-tun`

bridge br-int is not able to assign appropriate port to VM, so i removed br-int and once again added and it worked for me.

hope that helps you.

2015-05-25 00:29:44 -0500 answered a question what is the status shown in nova list

hello,

status here means the current status of your VM, after launching it properly the status turn to ACTIVE which means it is ready to use.

if you turnoff the VM the status goes to SHUTOFF. if some issues occur then status turn to ERROR.

is it what you required? Or you pointing towards something els.

2015-05-18 07:46:50 -0500 answered a question horizon Associate Floating IP to an instance in error state

hello,

first of all there is no point in allocating floating IP to an instance in error state. go through with the logs in /var/log/nova/* and fix the issue & the instance must be in Active & Running state and it should get a static IP first.

Floating IP's are used to access the public network from your instance (e.g Internet)

Static or private IP's are required for instance to access other instance present in the same network, Or internal communication you can say.
2015-05-11 04:52:34 -0500 answered a question can't ping or ssh cirros vm floating ip

hi,

for ping & ssh to your instance few rules are there to be added. i hope you done with that step.

Next one is Internet connectivity to your VM is done by IP-forwarding. first check what interface you provided for VM to communicate Internet (WIRELESS/LAN) based on that execute the following commands.

sudo iptables -t nat -A POSTROUTING -o eth0 -j MASQUERADE   (where eth0 is my interface for VM)

and one more thing make some changes in sysctl.conf

net.ipv4.ip_forward=1
2015-05-11 04:43:04 -0500 answered a question Which are the basic minimum services (mandatory) which are installed as part of any openstack installation ?

hello,

keystone is the authentication mechanism provided by OPENSTACK,

but not a mandatory thing to use keystone only,

if your organization has a better mechanism then you can use that, provided it must be able to communicate the rest of the components.

2015-03-10 06:24:37 -0500 answered a question deleted .ssh/known_host file deleted

hello.

.ssh/known_host file is auto generated.

if it is deleted then don't worry, wen you try to ssh again with any of your instance the known_host file will be automatically generated and the respective id's are saved.

i just tried it by moving the known_hosts file from .ssh folder to some other folder making .ssh empty and then tried ssh command to some other system the known_host file is automatically generated and ID's are saved in it.

give a try.

2015-03-09 06:29:37 -0500 answered a question '$ sudo keystone-manage db_sync' hangs

hey there,

its a common host resolution issue with all. and yes sometimes the host name may not be resolved so better make a practise of providing IP itself.

2015-03-09 06:16:31 -0500 answered a question ERROR 2003 (HY000): Can't connect to MySQL server on '127.0.0.1'

hey there,

mysql must be binded to an IP address if you want to bind it for localhost 127.0.0.1 then its is fine.

provided in localrc.conf the same IP must reflect.

my usual practise is will make one IP static in /etc/network/interfaces and provide that IP to localrc.conf

2015-03-07 08:00:26 -0500 edited question DHCP request does not reach tapXXX (qdhcp-XXX) network interface

To learn a little about OpenStack, a have installed 3 VMs using libvirt/KVM. I followed the http://docs.openstack.org/juno/install-guide/install/yum/content/

I am using CentOS 7, OPenStack Juno and neutron with openvswitch. One VM is the controller, the other is network node and the third is compute node. I am using a cirros image (cirros-0.3.3-x86_64).

Right now I can start an instance but DHCP does not work.
I can configure manually the IP DHCP allocates and it works, I can access the instance using ssh from outside and also can access external network from the instance.

I found that the instance DHCP request is reaching network node using the wrong interface:

[root@network neutron]# ip netns<br>
qrouter-89fe919a-0659-4e18-a609-aa698a110c9c<br>
qdhcp-2a920ddf-60ac-4e9f-8731-ca64b02f37df

[root@network neutron]# ip netns exec **qrouter-89fe919a-0659-4e18-a609-aa698a110c9c** ip link<br>
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT<br> 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00<br>
11: **qr-27520527-5c**: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN mode DEFAULT<br>
    link/ether fa:16:3e:b0:3d:2c brd ff:ff:ff:ff:ff:ff<br>
12: qg-79330b81-77: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN mode DEFAULT <br>
    link/ether fa:16:3e:56:e3:d2 brd ff:ff:ff:ff:ff:ff<br>

[root@network neutron]# ip netns exec qrouter-89fe919a-0659-4e18-a609-aa698a110c9c tcpdump -n -i qr-27520527-5c<br>
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode<br>
listening on qr-27520527-5c, link-type EN10MB (Ethernet), capture size 65535 bytes<br>
**(at this point I ran /etc/init.d/S40network start at instance console)**<br>
18:38:27.335127 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from fa:16:3e:63:f5:86, length 290<br>
18:39:27.414937 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from fa:16:3e:63:f5:86, length 290<br>

"fa:16:3e:63:f5:86" is the instance MAC address.<br>
So the DHCP request reaches the network node, but using the wrong interrface.

Dnsmasq is listening on tap8dc583ac-23 interface that is on qdhcp-XXX namespace.

root@network neutron]# ip netns exec **qdhcp-2a920ddf-60ac-4e9f-8731-ca64b02f37df** ip link<br>
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT <br>
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00<br>
9: **tap8dc583ac-23**: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN mode DEFAULT <br>
    link/ether fa:16:3e:69:91:f2 brd ff:ff:ff:ff:ff:ff<br>

As far as I know, the correct interface is tap8dc583ac-23, but there are something very strange maybe on openvswitch configuration.

Thanks for any help