Re: [openstack-dev] [devstack] noVNC disabled by default?

2015-02-03 Thread Marty Falatic (mfalatic)
I¹ve opened a bug to track this effort for DevStack (and to help point
others in the right direction when they stumble upon this):
https://bugs.launchpad.net/devstack/+bug/1417735

 - Marty


On 1/12/15, 10:02 AM, "Ben Nemec"  wrote:

>On 01/09/2015 05:24 PM, Sean Dague wrote:
>> On 01/09/2015 06:12 PM, Solly Ross wrote:
>>> Hi,
>>> 
>>> I just noticed that noVNC was disabled by default in devstack (the
>>>relevant 
>>> change was 
>>> 
>>> https://review.openstack.org/#/c/140860/).
>>> 
>>>
>>> 
>>> Now, if I understand correctly (based on the short commit message),
>>>the 
>>> rationale is that we don't want devstack to reply on non-OpenStack Git
>>> 
>>> repos, so that devstack doesn't fail when some external Git hosting
>>> 
>>> service (e.g. GitHub) goes down.
>> 
>> Realistically the policy is more about the fact that we should be using
>> released (and commonly available) versions of dependent software.
>> Ideally from packages, but definitely not from git trees. We don't want
>> to be testing everyone else's bleeding edge, there are lots of edges and
>> pointy parts in OpenStack as it is.
>> 
>>>
>>> 
>>> This is all fine and dandy (and a decent idea, IMO), but this leaves
>>>devstack   
>>> installing a "broken" installation of Horizon by default -- Horizon
>>>still  
>>> attempts to show the noVNC console when you go to the "console" tab
>>>for an  
>>> instance, which is a bit confusing, initially.  Now, it wasn't
>>>particularly
>>> hard to track not particularly hard to track down *why* this happened
>>>(hmm...   
>>> my stackrc seems to be missing "n-novnc" in ENABLED_SERVICES.
>>>Go-go-gadget
>>> `git blame`), but it strikes me as a bit inconsistent and
>>>inconvenient.   
>>>
>>> 
>>> Personally, I would like to see noVNC back as a default service, since
>>>it   
>>> can be useful when trying to see what your VM is actually doing during
>>> 
>>> boot, or if you're having network issues.  Is there anything I can do
>>> 
>>> as a noVNC maintainer to help?
>>> 
>>>
>>> 
>>> We (the noVNC team) do publish releases, and I've been trying to make
>>> 
>>> sure that they happen in a more timely fashion.  In the past, it was
>>>necessary  
>>> to use Git master to ensure that you got the latest version (there was
>>>a
>>> 2-year gap between 0.4 and 0.5!), but I'm trying to change that.
>>>Currently,
>>> it would appear that most of the distros are still using the old
>>>version (0.4), 
>>> but versions 0.5 and 0.5.1 are up on GitHub as release tarballs (0.5
>>>being a 3  
>>> months old and 0.5.1 having been tagged a couple weeks ago).  I will
>>>attempt to 
>>> work with distro maintainers to get the packages updated.  However, in
>>>the mean 
>>> time, is there a place would be acceptable to place the releases so
>>>that devstack
>>> can install them?
>> 
>> If you rewrite the noNVC installation in devstack to work from a release
>> URL that includes the released version on it, I think that would be
>> sufficient to turn it back on. Again, ideally this should be in distros,
>
>FWIW, I looked into installing novnc from distro packages quite a while
>ago and ran into problems because the dependencies were wonky.  Like,
>novnc would pull in Nova which then overwrote a bunch of the devstack
>Nova stuff.  I don't know if that's still an issue, but that's the main
>reason I never pushed ahead with removing the git install of novnc (that
>was during the release drought, so those weren't an option at the time
>either).
>
>> but I think we could work on doing release installs until then,
>> especially if the install process is crisp.
>> 
>> I am looking at the upstream release tarball right now though, and don't
>> see and INSTALL instructions in it. So lets see what the devstack patch
>> would look like to do the install.
>> 
>>  -Sean
>> 
>
>
>__
>OpenStack Development Mailing List (not for usage questions)
>Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev


__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev


Re: [openstack-dev] [all] Key signing at the summit?

2014-10-27 Thread Marty Falatic (mfalatic)
I'm relatively new to the keysigning *event* concept - can someone give a 
little more detail on this and where it comes into play? Does anyone else use a 
service (e.g., keybase.io) for this purpose?

 - Marty Falatic




-Original Message-
From: Clint Byrum [mailto:cl...@fewbar.com] 
Sent: Sunday, October 26, 2014 5:01 PM
To: openstack-dev
Subject: [openstack-dev] [all] Key signing at the summit?

Hi everyone! We have a summit rapidly approaching, and to my knowledge, no key 
signing event planned. That is unfortunate, as the web of trust that we started 
building in Atlanta would be quite stronger as we add more European developers 
who I'm sure will be present in large numbers due to proximity.

I believe we may be too late to do a mass keysigning, though if others feel we 
have time to gather fingerprints and disseminate a list for printing, then I 
will try to devote some time to it in the next 24 hours. However, I want to 
encourage everyone to print out your signature en-masse (hint hint: print it on 
your business cards) so that you can do an impromptu verification of identity 
with somebody. It may also be valuable to arrange a time where everyone can be 
present with their government issued IDs even if there isn't a mass list.

___
OpenStack-dev mailing list
OpenStack-dev@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

___
OpenStack-dev mailing list
OpenStack-dev@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev