Duh, I must have been asleep when I wrote that. The meeting is on Tuesdays,
look for Gantt on the OpenStack meetings page:
https://wiki.openstack.org/wiki/Meetings
1) It all depends upon how routable the IP address is for the host running
Horizon. Since 10.x.x.x is not routable and only visible from other nodes on
the 10.x.x.x network then moving from the office to your home you probably
won’t be able to see it without setting up some sort of tunnel so that you can
get to your Horizon machine from the outside. Talk to your network people
about how to get a routable IP address.
2) Your best option would be to look at the code. Check out the main
filter scheduler in the current Nova source tree at:
nova/scheduler/filter_scheduler.py
and then look at one of the filters such as:
nova/scheduler/filters/ram_filter.py
If you look at the tracking page for the priority work for Nova for the Kilo
release at:
https://etherpad.openstack.org/p/kilo-nova-priorities-tracking
There’s a large section on the scheduler, you can follow the blueprints in that
sections to see what work we are doing for the next release.
--
Don Dugger
"Censeo Toto nos in Kansa esse decisse." - D. Gale
Ph: 303/443-3786
From: Silvia Fichera [mailto:[email protected]]
Sent: Tuesday, March 17, 2015 4:27 AM
To: Dugger, Donald D
Cc: [email protected]
Subject: Re: [Openstack] Questions from a newbie
1- I'm asking what happens if I move my laptop from my office network to
another one (e.g. if I need to use the dashboard also from home).
2- Have you got a code flow to better understand the information publishing
process? And how it will be managed in the future?
I started my internship just one week ago, they told my that the purpose is add
a feature that will consider also the network link status togheter with the
host status. So I think that the scheduler should interact also with the
network module, but by now I'm still lost in understanding the architecture and
the OpenSTack scenario.
I think I will join the meeting when I have time, on which day is it?
Thank you again
2015-03-17 5:30 GMT+01:00 Dugger, Donald D
<[email protected]<mailto:[email protected]>>:
Silvia-
1) Not totally sure what you’re asking. When `devstack’ completes there
should be an instance of Horizon running on that machine. You can then connect
to that Horizon from any other machine on that network, assuming your host has
a routable IP address (using 127.0.0.1 as the HOST_IP would make Horizon only
available from that machine). Note, by default Horizon listens on port 9000 so
make sure you are trying to connect to
10.1.115.126:9000<http://10.1.115.126:9000>
2) Yes, currently the host information is stored in the DB but there is a
lot of work going on in that area. We hope to move to an object model where
the compute nodes will be sending state information directly to the scheduler
in the Kilo release.
I’d be curious to know what it is about the scheduler that you are planning on
working on. I should also point out that we have a weekly IRC meeting to
discuss scheduler issues at 1500 UTC at #openstack-meeting (the meeting is
titled Gantt which is the name of the new scheduler project we are creating),
you’re welcome to join if that time zone works for you.
--
Don Dugger
"Censeo Toto nos in Kansa esse decisse." - D. Gale
Ph: 303/443-3786
From: Silvia Fichera
[mailto:[email protected]<mailto:[email protected]>]
Sent: Thursday, March 12, 2015 3:15 AM
To: [email protected]<mailto:[email protected]>
Subject: [Openstack] Questions from a newbie
Hi,
I'm completely new in OpenStack's Universe and I need to understand a lot of
things!
Bust let's start with a couple of questions.
1. I have installed OpenStack using DevStack in a physical machine (I have
already an Ubuntu system). At the end of the installation there was written the
address where I can open Horizon (e.g. 10.1.115.126). What happens if I connect
my laptop from another network? Does it remain the same? Or should I do a
specific configuration?
2. I will work with nova in particular, especially in the scheduling part. I
would like to know where the scheduler take the information about the state of
the host to assign it after the filter & weight.
I think that those parameters about the status of the host are memorised in the
database, but it's only my hypotheses, I haven't read anything about it.
And if it's so, how are collected those information? By who?
Sorry if I did some silly question, but I would like to well understand the
basis.
Let me also know if exist a good guide that explains how to start with
OpenStack from zero!
Thank you
--
Silvia Fichera
--
Silvia Fichera
_______________________________________________
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : [email protected]
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
