Ask Your Question

happy75's profile - activity

2020-02-25 14:03:08 -0600 received badge  Notable Question (source)
2020-02-21 02:13:43 -0600 received badge  Famous Question (source)
2020-02-04 19:00:06 -0600 commented answer partition size guidelines for different nodes

thanks :-) I will read the architecture guide.

2020-01-27 09:13:07 -0600 received badge  Notable Question (source)
2020-01-26 23:00:50 -0600 received badge  Popular Question (source)
2020-01-23 21:06:18 -0600 received badge  Notable Question (source)
2020-01-23 20:18:26 -0600 commented answer additional controller node(s)

thanks again. If I want to build an experimental cloud with 11 physical servers, are their any documents out there which gives me examples/ideas on what services to put on the 11 physical servers (e.g. 1 controller node, 9 compute nodes (each with nova and neutron), 1 storage node(ceph,cinder,swift)

2020-01-23 20:00:26 -0600 asked a question partition size guidelines for different nodes

Just wondering if there are any partition size guidelines for the different file systems for different nodes?

I have a 300GB disk for each of the node.

  1. for controller node, would giving 50Gb for / file system and 50Gb /var be enough?

I presume it's better to have different filesystems for / and /var because all the openstack logs go under /var/log/....

  1. for compute node (where all the vm will reside), which filesystem do all the vm (i.e. the vm created by the users) go?

  2. partition size for network and storage node?

I couldn't seem to find any guideline doco on the openstack website (i know it depends very much on the environment but there should be some guideline)....

2020-01-23 17:49:17 -0600 received badge  Popular Question (source)
2020-01-23 14:25:58 -0600 commented answer cinder and swift?

okay I will take a look at Ceph. thanks

2020-01-23 14:25:08 -0600 commented answer additional controller node(s)

thanks again for your responses. Much appreciated :-)

2020-01-23 14:24:35 -0600 commented answer additional controller node(s)

If I only start with one controller, will it affect the performance of the cloud if my plan is to have a cloud of 11 physical servers (the goal is to create 100 vm)? What is the purpose of having multiple controllers? Is that ONLY for high availability and load balancing and no other purpose?

2020-01-23 01:55:18 -0600 commented answer additional controller node(s)

actually as a start I won't need high-availability and load balancing so if that's the case does that mean all I need is just ONE controller node and lots of compute nodes? I mean if I need high-availability and load balancing later, I can then install HA-Proxy (which I will read up more on)

2020-01-23 01:34:51 -0600 received badge  Popular Question (source)
2020-01-23 00:14:08 -0600 commented answer additional controller node(s)

also will you need to install horizon on the additional controller nodes? I believe no, is that right?

2020-01-23 00:09:21 -0600 commented answer additional controller node(s)

will i be able to set up additional controller nodes WITHOUT using HA-proxy?

2020-01-22 23:49:29 -0600 commented answer additional controller node(s)

Thanks. i found something on the internet. it is like this? https://wiki.infn.it/progetti/cloud-a...

2020-01-22 21:35:33 -0600 commented answer additional controller node(s)

and thanks for the links to those documents. I will take a look at them to get some ideas.

2020-01-22 21:35:02 -0600 commented answer additional controller node(s)

thanks. So to clarify, for the configuration of the additional controller, do i just use the same set of instructions as the existing node (the ones provided on the openstack website) - and the only difference is that the additional node has a different ip address?

2020-01-22 21:33:00 -0600 commented answer additional compute node(s)

thanks for the tips

2020-01-22 19:31:50 -0600 commented answer additional compute node(s)

okay so is that why cinder (which provides persistent volumes) should be installed ?

2020-01-22 19:31:50 -0600 received badge  Commentator
2020-01-22 18:57:45 -0600 commented answer cinder and swift?

okay thanks I will install both then - i mean installing both means that the cloud has extra functionality (since you said they don't conflict with each other), right?

2020-01-22 18:25:44 -0600 asked a question additional compute node(s)

I have questions with regards to adding additional compute nodes:

when you manually install and configure an additional compute node (with a different ip address of course) in the openstack environment, does the additional compute node automatically get connected to the same openstack environment and be able to talk to the old compute nodes and all the other controller nodes which already existed in the openstack environment? If one of the compute nodes go down, does that mean all the vms that were created would have been lost?

2020-01-22 18:22:40 -0600 asked a question additional controller node(s)

Hi Openstack experts,

I have some questions with regards to adding additional controller nodes:

(1) when you install and configure an additional controller node (with a different ip address of course) in the openstack environment, does the additional controller node automatically get connected to the same openstack environment and be able to talk to the old controller nodes which already existed in the openstack environment?

(2) if you have 3 controller nodes in the same openstack environment, do these 3 controller nodes take on different roles or do they basically perform the same tasks? do the load automatically get distributed among the 3 controller nodes? If one of the controller nodes go down, does that mean the other 2 nodes will take on what that one was doing so no operation would be disrupted?

2020-01-22 18:15:35 -0600 asked a question cinder and swift?

Hi Openstack experts? Can you create containers/dockers inside vm within the cloud with Cinder and swift? Can you install both the cinder and swift services within the cloud? Would it cause any conflict? Do people normally just install one and not both (depending on they need of course)?

thanks

2020-01-21 16:41:01 -0600 received badge  Famous Question (source)
2020-01-21 16:27:17 -0600 received badge  Enthusiast
2020-01-21 11:42:28 -0600 received badge  Popular Question (source)
2020-01-20 20:21:04 -0600 asked a question Heat - creation of instances

Hi Openstack experts, Can you create lots of virtual instances (say more than 20) using the HOT template (as part of the HEAT service in openstack)? If so, is it just a matter of specifying it in the template?

thanks

2020-01-17 13:33:08 -0600 received badge  Notable Question (source)
2020-01-13 02:21:42 -0600 received badge  Notable Question (source)
2020-01-10 06:17:31 -0600 received badge  Notable Question (source)
2020-01-09 20:50:29 -0600 commented answer Ansible questions

thanks for the links

2020-01-09 16:31:18 -0600 received badge  Popular Question (source)
2020-01-09 08:30:09 -0600 received badge  Popular Question (source)
2020-01-09 01:59:24 -0600 received badge  Notable Question (source)
2020-01-08 20:51:00 -0600 received badge  Popular Question (source)
2020-01-08 16:13:42 -0600 commented answer troubleshooting

thanks. I also found this video as well and it is great!

2020-01-08 07:04:48 -0600 received badge  Popular Question (source)
2020-01-08 06:52:20 -0600 received badge  Popular Question (source)
2020-01-07 23:31:38 -0600 asked a question Different IP address for vm

HI, I am using the provider network option for my openstack test environment. I only have one provider network and it's in the subnet 10.0.0.x I created two vm. They were both created successfully and were given an ip address in the 10.0.0.x range. However, one strange thing is that one of the vm has a different ip address shown on the dashboard than on the machine (i.e when I run "ip a", it shows a different ip address (in the same subnet though) than the one shows in dashboard). As a result, I am not able to ping the vm from the other vm. has anybody seen this problem before ? if so, how do you fix it? Actually where does the vm get the ip address from? Note that I created both vm the same way so don't understand why one is working and the other is not!

thanks!

2020-01-07 22:15:15 -0600 commented question Ansible questions

okay thanks

2020-01-07 17:33:32 -0600 commented answer Openstack provider network

thanks for your answers! I presume what you are saying is that with provider network, you can only create network based on real physical or VLAN IP addresses whereas with the tenant network, the users can make up their own IP addresses and create their own networks, is that right?

2020-01-07 17:30:01 -0600 commented answer do i just add on more servers/services after the minimal deployment

thanks ! actually I have already installed Openstack manually so will not go with Packstack but thanks anyway.. I may try to follow the ansible instructions on the Openstack website to automate the installations.

2020-01-07 07:14:46 -0600 received badge  Student (source)
2020-01-06 23:47:55 -0600 asked a question Openstack provider network

HI Openstack Networking experts,

I have 2 questions: I understand that there are 2 types of network in Openstack: (a) Provider network and (b) Tenant network (1) i have created my private cloud using the provider network option. It seems that I can't even create any network/subnet using the dashboard and this is logging in as the admin user. When I tried to create the network, I got an error message saying that "Unable to create network. No tenant network is available for allocation...". Is this normal? I mean as the admin user, shouldn't I be able to create the network even from the dashboard? Because of this error, I went to create the network and subnet as per the following link and it works, does anybody know why it is okay to create network from the CLI but not okay to do it from dashboard? Am I doing something wrong? https://docs.openstack.org/newton/install-guide-ubuntu/launch-instance-networks-provider.html (https://docs.openstack.org/newton/ins...)

(2) I have allocated the default IP address to my controller and compute node (they are 10.0.0.11 and 10.0.0.31 respectively). Is it okay to use the provider network option if my private cloud is NOT connected to the outside world (internet)? Sorry i am a bit confused as to when to use the provider network and when to tenant network option. I understand the tenant network option is used if you want your users to be able create their own network etc but how does that affect the IP addresses you select for the VMs? For provider network option, does that mean that i can create virtual network inside Openstack using any IP addresses (i.e. any IP addresses that are not used on my network) and then the vm will be able to communicate to each other? What is the recommendation for the sort of IP addresses I should be allocated to the virtual instances, i.e. should it be in the same subnet as the controller and compute node (so 10.0.0.x in my case)? It says on the official Openstack doco that "For provider network option, an instance uses a provider (external) network that connects to the physical network infrastructure via layer-2 (bridging/switching). This network includes a DHCP server that provides IP addresses to instances".

Many thanks in advance!

2020-01-06 17:27:11 -0600 asked a question Ansible questions

Hi Openstack experts,

I have two questions: (1) What do people use Ansible for in the context of Openstack? Is it just for automating the installation of Openstack services?

(2) I would like to use ansible to automate the installation of the openstack services (which i believe would save a lot of time especially if the openstack environment contains many compute nodes). Is the following link a good website to try as a starting point? https://docs.openstack.org/project-deploy-guide/openstack-ansible/ocata/ (https://docs.openstack.org/project-de...) Does anybody know if there are any great resources out there which provides great instructions on installing/deploying multi-node Openstack environment using Ansible?

Thanks in advance!

2020-01-06 17:27:11 -0600 asked a question troubleshooting

Hi Openstack experts, in an openstack private cloud where there are multiple controller and compute nodes, if there are problems with some service(s), how do you know which controller/compute to look for the error messages? e.g. if the neutron is not working, how do you know which controller/compute node to look for the error message for the neutron log file? Sorry I know this is a general question but want to get some sort of guidelines on what people do to troubleshoot. Many thanks!