On Tue, Mar 21, 2017 at 1:22 AM Sterdnot Shaken
wrote:
> Wow! Thanks for answering both of my questions!
>
> So, I did some things you suggested, including setting the MSS in iperf to
> something small (1000 bytes) and tested with no improvement. I then changed
> the VM running on Openstack to ha
Hi Vivek,
On Fri, Mar 17, 2017 at 11:27 AM Vivek Nirala
wrote:
> Alternatively can we use a bridge to attach the nic interfaces(of n number
> of virtual machines) ?
>
Yes, this is exactly what Neutron does by default.
> If yes then is there any specific neutron command to add a bridge and
> a
On Tue, Mar 14, 2017 at 12:06 PM David Gabriel
wrote:
> Thanks so much Neil for your support.
> I fixed this issue.
> It is due to the mask. I forgot to set it.
>
Good to hear!
>
> Please do you have any idea about my earlier question entitled
> 'communication between two ovs bridges' ?
>
I'm
Hi David,
The interface name "eth1" implies - or at least strongly suggests - that
each of your VMs has two interfaces (eth0 and eth1). Is that the case?
It's also a bit odd that you have to do 'ifconfig eth1 up' by hand.
Normally the VM OS should do that; and if it isn't, there are probably
oth
I notice that at the end of the survey, when I click on the "SUBMIT YOUR
SURVEY >" button, I just get the same page again, and so could click
"SUBMIT YOUR SURVEY >" again, and so on...
You might want to look at fixing that; and in the meantime, I hope that
would not result in multiple copies of th
Hi Atif,
There is incorrect Python indentation in the local_settings file,
/usr/share/openstack-dashboard/openstack_dashboard/local/local_settings.py.
>From the perspective of the vanilla Horizon project, I believe
local_settings.py is a file that the user can create and/or modify in order
to inf
In principle... In the fully general case I think there are issues lurking, so it would indeed be interesting to hear about real experience from people who have done this. FWIW here's what I think I understand abo
Yes, in broad terms you would do this by implementing an alternative DHCP driver class and configuring the DHCP agent to use that class instead of its default (which is Dnsmasq). The DHCP driver gets a 'Network m
I would guess that:1. You need to be on Xenial. 2. You need to upgrade whichever package it is that provides add-apt-repository. (A general 'apt-get update && apt-get upgrade' should cover that.)Regards, Neil
Well what I do, on a fresh Xenial box, is:
sudo add-apt-repository ppa:juju/devel
sudo apt-get update
sudo apt-get install -y juju-2.0 charm
But more generally I suggest that you refer to
https://jujucharms.com/docs/stable/getting-started.
Neil
On Tue, Oct 4, 2016 at 9:46 AM R P PANDI wr
I think that can happen when you don't have enough disk space on the relevant machines to construct and store the snapshot. Unfortunately I don't remember which the relevant machines are - could be both the comput
Thanks, John, that's a really useful approach.
On Mon, Aug 22, 2016 at 4:27 PM John Belamaric
wrote:
>
> On Aug 22, 2016, at 11:12 AM, John Belamaric
> wrote:
>
>
> On Aug 19, 2016, at 5:35 PM, Neil Jerram wrote:
>
> On Fri, Aug 19, 2016 at 9:25 PM John Belama
On Fri, Aug 19, 2016 at 9:25 PM John Belamaric
wrote:
> >
> > Yes, and that was what I was alluding to in order to make packets go out
> the correct interface based on the source IP in the packets.
> >
>
> You can configure it to send the packets out the interface the original
> packet came in on
On Wed, Aug 17, 2016 at 7:35 PM Brian Haley wrote:
>
> Is a cloud datacenter going to have multiple link types? I doubt it. And
> having two VMs would be easier anyways as there are probably other things
> you'reha
> going to tune.
>
>
Yes, I think so. I'm currently working on an OpenStack cus
Hi Daniel,
It's because /etc/network/interfaces on CirrOS only starts eth0.
To make a customized CirrOS that does DHCP on eth1 as well as eth0, you can:
- add these lines to /etc/network/interfaces:
auto eth1
iface eth1 inet dhcp
- snapshot the instance after making that change
- thereafter,
Not sure what the problem would be with 'Quay' or 'Street' - they both
sound like good options to me.
On Thu, Jul 14, 2016 at 11:29 AM Eoghan Glynn wrote:
>
>
> > >> Hey all!
> > >>
> > >> The poll emails for the P and Q naming have started to go out - and
> > >> we're experiencing some difficu
Mon, Jun 20, 2016 at 9:45 AM Ahmed Medhat wrote:
> i add it but it is not working also.
> May be This iptables rule for traffic reaching the dhcp port but there
> should be one in the other direction and be dport 67 ?
>
> Best regards,
> Ahmed
>
> On Monday, June 20, 2016,
o the VM.
> I want to add tgat if I adjusted the ip manually inside the VM instance ,
> it works fine and I can ping the dhcp ip 11.0.0.2 in my case.
>
> I hope this info shows you where could be the problem I have.
>
> Best regards,
> Ahmed
>
>
> On Monda
Hi Ahmed,
Unfortunately there are many possible reasons - and different ones
depending on your setup.
I will guess, though, that you're using a default OVS setup, without DVR.
In that case DHCP should be provided by Dnsmasq running on your network
node, and I would start by checking for that proc
Hi Remo,
I've recently been using Juju to deploy a small OpenStack cluster, for
development and testing. I don't think I've been using a dual Juju server,
but I believe that's a common thing to do for larger and longer-lived
deployments. MAAS is just one of the types of 'cloud' that Juju can dep
des only attach to management and tenant
> network and access internet via NAT?
>
>
>
> Best regards,
>
> Rui Mao
>
>
>
> *发件人:* Neil Jerram [mailto:n...@tigera.io]
> *发送时间:* 2016年5月19日 16:20
> *收件人:* Rui Mao ; openstack@lists.openstack.org
> *主题:* Re: [Opens
If you attach a VM to a provider network - instead of to a tenant network -
and give it an Internet-routable fixed IP address, it can access the
Internet without any NAT. This has been the case for several OpenStack
releases.
Now that Neutron has address scopes, I believe it will soon be the case
networking-calico has a script at [1] that can be used to set up a multi-node
stable/liberty DevStack cluster, and extensive comments inside the script
explain that.
If you use this script, it should be on fresh Ubuntu Trusty VMs with no data
that you'd mind losing, and please note that the cl
On 11/03/16 11:52, Butts, Darren (EXT - GB) wrote:
> Neil
>
> Thanks, I understand that 2015.1.2 is a cumulative set of fixes for neutron
> but it would be rather pointless taking the single change only to find I
> needed to also apply additional changes ? It would make more sense to apply
> the
Hi Darren,
On 11/03/16 11:28, Butts, Darren (EXT - GB) wrote:
> All
>
> In order to resolve some issues with the default drop rules being
> applied before user defined rules in the iptables of the compute nodes I
> need to apply the neutron patch 2015.1.2 however there are no
Terminology issue he
I think the Neutron server log is indicating invalid credentials. Suggest
checking the auth settings in your neutron.conf.
Neil
From: Alexandr Porunov
Sent: Friday, 26 February 2016 20:39
To: Openstack
Subject: [Openstack] [neutron] 503 Service Unavailable
Hello everyone,
I am trying to
On 25/02/16 19:08, John van Ommen wrote:
> Is it possible to have one interface configured with VLAN and another
> with VxLAN?
Do you mean for tenant networks, or for provider networks?
I believe it's possible for provider networks, but not for tenant. [1]
[1]
http://lists.openstack.org/piperma
On 02/02/16 09:54, Pavel Fedin wrote:
> Hello!
>
>> I'd say the most important detail is that SERVICE_HOST, in the
>> local.conf for the second compute node, needs to be the name or IP
>> address of the controller. Did you do that?
> Of course i did. This is my local.conf.
> By this time i disc
Hi Pavel,
On 02/02/16 08:47, Pavel Fedin wrote:
> Hello!
>
> I've got DevStack running on one node, and now i am adding a second compute
> node. local.conf for this node is taken from here
> (http://docs.openstack.org/developer/devstack/guides/neutron.html). stack.sh
> completes and i see q-ag
On 19/01/16 07:36, Andreas Scheuring wrote:
> Hi everybody,
>
> I stumbled over a definition that explains the difference between a
> Provider network and a self service network. [1]
I've also spent time trying to understand this, so am happy to offer
that understanding here (for checking?)...
I'm pretty sure there was another thread on this within the last week, on
either openstack-dev@, openstack@ or openstack-operators@, but right now (with
just my phone) I can't find that.
Regards - Neil
From: Thales
Sent: Sunday, 1 November 2015 00:07
To: Openstack
Reply To: Thales
Subject: [Ope
Is there any ETA for permanent RDO Liberty packages?
Thanks,
Neil
From: Matt Kassawara [mailto:mkassaw...@gmail.com]
Sent: 16 October 2015 20:04
To: Martinx - ジェームズ
Cc: openstack@lists.openstack.org
Subject: Re: [Openstack] OpenStack "Liberty" is officially released !
Ubuntu has
Hi Menyi,
I've offered to be an Outreachy mentor for Neutron, so would be happy to help
you. I won't be much on IRC until Monday, though. Would an IRC chat at some
time on Monday work for you?
Regards,
Neil
From: menyi sona
Sent: Friday, 9 October 2015 17:17
To: openstack@lists.openstac
ember 2015 04:50
To: Neil Jerram
Cc: Openstack@lists.openstack.org
Subject: Re: [Openstack] Neutron
Hi Neli,
The DHCP agent is running good.
In controller node:
[Inline image 1]
Dashboard :
[Inline image 2]
Login to the instance and run the command is ifconfig
[Inline image 3]
Than
Hi Venkatesh,
No IP probably means that the DHCP mechanism failed somehow. Are you running a
DHCP agent?
Neil
From: venkatesh kotipalli [mailto:openstackvenkat...@gmail.com]
Sent: 28 September 2015 10:01
To: Openstack@lists.openstack.org
Subject: [Openstack] Neutron
Hi All,
I have
On 21/09/15 20:31, Victoria Mart?nez de la Cruz wrote:
Hi all,
I'm glad to announce that Mahati Chamarthy (mahatic) will join the current
coordination efforts for the Outreachy internships. Thanks Mahati!
Also, I wanted to share this Etherpad [0]with you the current list of
applicants and mento
36 matches
Mail list logo
