On 1 Jun 2015 10:50 am, "Alan Pevec" wrote:
>
> 2015-05-29 18:30 GMT+02:00 Jeremy Stanley :
> > On 2015-05-29 16:30:12 +0100 (+0100), Dave Walker wrote:
> >> This is generally my opinion as-well, I always hoped that *every*
> >> commit would be considered a release rather than an arbitrary
> >> ta
s no worse on either of those problems, and
will supply some infrastructure we might use to tackle them in future.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List
f reference for developers and developer teams considering
new technologies, and we can iterate on this over a few cycles with an
eye to making it more formal.
Please do chime in (here, or in the page, or $however) to provide more
information or context.
Thanks!
-Rob
--
Robert Collins
Distinguished
actually consider the request as failed.
>
> Also, I think that if nova can't schedule "max count" instances, but can
> schedule at least "min count" instances, then it shouldn't put the
> unscheduled ones into an error state--it should just delete them.
I thin
heers,
Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
gh the logs before overriding its vote and merging a patch with
> red HA CI. If the patch in question is not critical with regards to time,
> recheck is an option :)
Awesome.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
_
he test isn't
interested in dep management, its interested in whether the plugins
implement the right config options. There is a stevedore API that
would be even better to use - see the review linked above.)
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
_
...@lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not
On 21 May 2015 at 21:22, YAMAMOTO Takashi wrote:
> hi,
>
> can you explain why you want to use pbr 1.0.1 in the first place?
> what's wrong with requiring pbr<1.0 ?
pbr 1.0 adds important features for python 2.x <> 3.x management and
optional dependencies via extras.
o become fully resolvable. You can force this by pip
install -U pbr after your things are installed.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not fo
ve to always using the uuid?
>
> I ask because instance.name is not guaranteed to be unique depending
> on how weird the nova deployment is.
I don't know. I know that we had issues with nova-baremetal and around
the start of Ironic and IIRC moved to uuids everywhere as a result.
-Ro
On 11 May 2015 at 23:51, Sean Dague wrote:
> It appears that we've basically run out of interest / time in
> realistically keeping pypy working in our system.
>
> With the focus on really getting python 3.4 working, it seems like it
> would just be better to drop pypy entirely in the system. In th
>=0.3.14
-hacking>=0.9.2,<0.10
+hacking>=0.10.0,<0.11
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: ope
There was one more issue - https://bugs.launchpad.net/pbr/+bug/1456663
- which has now been fixed in pbr 1.0.1.
-Rob
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.ope
until we've
worked through our management of dependencies to be less fragile.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage que
n, Sean, and Joe for helping unwind, analyze and fix this.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openst
+++ b/test-requirements.txt
@@ -7,0 +8 @@ sphinx>=1.1.2,<1.2
+six>=1.9.0
@@ -11,0 +13 @@ testtools>=0.9.34
+virtualenv
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development
for optional
dependencies ('extras' in setuptools parlance) - which will allow
this. We're having a session about such things in Vancouver too.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
nt, be difference because it is important in
this specific case.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-req
justing.
>
> http://www.openstack.org/blog/2015/05/technical-committee-highlights-may-13-2015
Cool - thanks very much for leading this!
Uhm, one small errata. My term just started :)
-Rob
--
Robert Collins
Distinguished Technolo
me...
Anyhow, I hope we do find two champions for the client libraries,
because I believe in enabling as many people as possible to use our
clouds; and just like we have php sdk, it would be great to know that
our python sdk's work on pypy too.
-Rob
--
Robert Collins
Distinguished Te
As an aside it might be nice to get this specifier from
global-requirements, so that its managed in the same place as all our
other specifiers.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
Open
s without neutron. Folk can install neutron first, or use
a local requirements file to pick a particular revision to install
from git, and pip will be fine with that (thats what CI does right
now:)).
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
_
not Python per se.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
h
sive with pymysql before embarking on a change to the
> concurrency model. If it doesn't, not much harm done, and if it does,
> but makes us CPU bound, well then we have even more of a reason to set
> out on such a large task.
And yes.
-Rob
--
Robert C
how many
processes will they need?
FWIW, I think moving to an eventlet friendly library should be the
first step because it can be done much more rapidly and with arguably
less risk.
I don't think the discuss ends there though :)
-Rob
--
Robert Collins
Distinguished Technologist
HP Conver
hold the GIL, but not whether thats happening.
> Do you need a DB example, by using the mysql C driver,
> and waiting in an actual I/O primitive ?
waiting in an I/O primitive is fine as long as the GIL has been released.
-Rob
--
Robert Collins
Distinguished Technologist
HP Conver
ative is to add this in the objects layer, on top of
> the rpc dispatch:
> https://github.com/openstack/nova/blob/master/nova/objects/base.py#L188
> But that somehow feels like a layer violation, maybe its not.
No opinion here, sorry :)
-Rob
--
Robert Collins
Distinguished Technologist
H
On 9 May 2015 at 08:33, Sean Dague wrote:
> On 05/08/2015 02:48 PM, Robert Collins wrote:
>> Thats certainly possible too. Upside: if it works we know it works in
>> pip. Downside, we'll be tracking something that is in active
>> development and late-prototype /early a
On 10 May 2015 at 08:03, Jeremy Stanley wrote:
> On 2015-05-10 07:49:25 +1200 (+1200), Robert Collins wrote:
> [...]
>> We should do the following immediately we create a virtualenv anywhere
>> in our infra:
>> get-pip.py
>> pip install -U wheel setuptools
> [...
On 10 May 2015 at 07:19, Jeremy Stanley wrote:
> On 2015-05-10 07:09:38 +1200 (+1200), Robert Collins wrote:
>> This is my understanding: We explicitly install pip latest in our jobs
>> (via get-pip, retrieved over https). Virtualenv shouldn't be version
>> sensitive at
On 9 May 2015 at 23:42, Jeremy Stanley wrote:
> On 2015-05-09 06:46:59 +1200 (+1200), Robert Collins wrote:
>> As I read it, we've got some tooling that isn't PEP-440 compatible
>> (https://www.python.org/dev/peps/pep-0440/#compatible-release defines
>> ~=) and
On 8 May 2015 at 21:36, Andreas Jaeger wrote:
> On 05/08/2015 10:12 AM, Andreas Jaeger wrote:
>>
>> On 05/08/2015 10:02 AM, Robert Collins wrote:
>>>
>>> I don't know if they are *intended* to be, but right now there is no
>>> set of versions th
tion create a bug we can use to track this? I
don't think it has anything to do with the choice of cap-or-not.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mail
On 8 May 2015 at 23:23, Sean Dague wrote:
> On 05/08/2015 07:13 AM, Robert Collins wrote:
>> The resolver I have doesn't preserve the '1b' feature at all at this
>> point, and we're going to need to find a way to separate out 'I want
>> X' fro
p itself will succeed once the resolver is merged.
The resolver I have doesn't preserve the '1b' feature at all at this
point, and we're going to need to find a way to separate out 'I want
X' from 'I want X and I know better than you', which will let folk get
On 8 May 2015 at 20:39, Andreas Jaeger wrote:
> On 05/08/2015 10:02 AM, Robert Collins wrote:
>>
>> I don't know if they are*intended* to be, but right now there is no
>> set of versions that can be co-installed, of everything listed in
>> global requirements.
rising to the gate ;)
I don't have a fully formed view yet on complete co-installability in
our global requirements. But without that it may be hard to calculate
a working set that we can pin (if we go all-pinned), and its certainly
harder to write tooling to do any sort of analysis beca
e right forum for discussing the pressures that lead to this
hack, so we can do something that works better with the underlying
tooling, rather than in such a disruptive fashion?
-Rob
--
Robert Collins
Distinguished Technologist
H
On 5 May 2015 at 18:17, Matthias Runge wrote:
> On 05/05/15 05:29, Robert Collins wrote:
>
>>> Probably, but it's legally wrong (ie: worst case, you can be sued) to
>>> leave
>>> a package which is in direct violation of the license of things it
>>>
u can be sued) to leave
> a package which is in direct violation of the license of things it contains.
So,we shouldn't use angular at all then, because as a js framework its
distributed to users when they use the website, but the license file
isn't included in that distribution.
-Rob
On 7 April 2015 at 10:43, Robert Collins wrote:
>> $ time openstack -h
>>
>> real0m2.491s
>> user0m2.378s
>> sys 0m0.111s
>
>
> pbr should be snappy - taking 100ms to get the version is wrong.
I've now tested this.
With an egg-info prese
On 5 May 2015 at 11:48, Davanum Srinivas wrote:
> Not sure about the history, but +1 to remove oslo-core from that group.
Done.
If its a terrible mistake we can add it back easily enough.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cl
finite future pbr cannot ever ever ever ever break API
compatibility. We cannot issue stable versions or point releases. And
this is all due to setup_requires and how that works, and other bugs
in the ecosystem.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
gular
libraries can release the one special-case library in oslo (pbr,
because setup_requires) - this doesn't make sense to me or Clark :).
Anyone with insight please shout out now, if we don't hear anything
back justifying this, I'll remove oslo-core from pbr-release later
this week.
-Rob
On 5 May 2015 at 08:12, Monty Taylor wrote:
> On 05/04/2015 03:53 PM, Robert Collins wrote:
> I'm fine with that in principle - I tend to release personal libraries
> pretty much as soon as something interesting hits them. I have no
> personal fear of high release counts.
>
On 5 May 2015 at 08:04, Clark Boylan wrote:
>
>
> On Mon, May 4, 2015, at 12:53 PM, Robert Collins wrote:
> I don't understand what you mean by "zero real world testing of master".
> We test that pbr can make sdists and install packages on every change to
> ma
e testing, and we don't get that, and b) to
let -core notice mistakes and back things out. Waiting to release once
an improvement is in master just delays giving the benefits to our
users.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
___
roblem space.
I like the thrust of your proposal.
Any reason not to ask the existing newsletter to be a bit richer? Much
of the same effort is required to do the existing one and the content
you propose IMO.
-Rob
--
Robert Collins
Distinguished Technologist
HP Con
tps://review.openstack.org/#/c/168805/
> Complete failure log @
> http://logs.openstack.org/05/168805/8/check/gate-nova-python27/584359c/console.html#_2015-05-04_07_23_32_995
>
> thanx,
> deepak
>
>
> __
> OpenStack Dev
t explored for far too long. I keep running into URL's that are
> labeled obsolete.
>
http://ci.openstack.org/
Cheers,
Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Develo
On 4 May 2015 at 08:59, Steve Baker wrote:
> On 02/05/15 07:06, Robert Collins wrote:
>>
>> On 2 May 2015 at 04:17, Doug Hellmann wrote:
>>>
>>> Excerpts from Robert Collins's message of 2015-05-01 10:59:49 +1200:
>>>>
>>>> pbr 0.1
ew.openstack.org/#/c/162311/
introduced the versions. https://bugs.launchpad.net/heat/+bug/1450733
is the bug for this.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Deve
explicit: If admin calls are implicit in setting up user
tests, that would be more than sufficient IMO.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not fo
want to let the contributor
make an assessment as to at what point the 'is a coauthor' threshold
has been passed.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Developme
ly, and pbr will generate numbers for you, with
you choosing the actual one when you make a new tag.
We're dealing with the most visible projects that have bad metadata
now in -infra, but projects using pbr elsewhere will probably be
puzzled - thus this email :)
-Rob
On 1 May 2015 at 09:04, Dean Troyer wrote:
> On Thu, Apr 30, 2015 at 3:28 PM, Robert Collins
> wrote:
>>
>> Blaming pip for our problems, when we haven't been willing to go and
>> contribute fixes to pip is massively unfair to the pip developers.
>
>
> So
s going to
have similar symptoms: issues that arise won't get fixed unless we're
willing to work up the dependency graph and fix things at the right
place.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
___
secretary and be responsible for providing some push (vs
pull) insight into the current state of things.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (
On 28 April 2015 at 22:57, Sean Dague wrote:
> On 04/27/2015 05:53 PM, Robert Collins wrote:
> I'm still generally suspicious of the precompute / install model because
> solving that ends up being... interesting some times. I also think there
> is a related issue of "dep
py27)$ openstack help
Actually no, its '--notest' - see tox --help. What you've done is run
the tests with a filter that didn't match any tests.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
___
cost of
a small additional feature in pip.
Thierry says we're going to slot this into
https://libertydesignsummit.sched.org/event/da0f31eddd0def88c6c51fb131fe87bd#.VT6v1N_va00
or the followup after lunch - I'd like to pin it down more than that,
if we can.
-Rob
I realised it was worse than that - see the review.
On 26 April 2015 at 07:11, Robert Collins wrote:
> On 26 April 2015 at 01:49, Jeremy Stanley wrote:
>> On 2015-04-25 12:12:15 +1200 (+1200), Robert Collins wrote:
>> [...]
>>> I'd like to make that a little more o
On 26 April 2015 at 01:49, Jeremy Stanley wrote:
> On 2015-04-25 12:12:15 +1200 (+1200), Robert Collins wrote:
> [...]
>> I'd like to make that a little more official:
>> - put it in our docs
>> - stop testing python setup.py install.
> [...]
>
> And emi
y to track this.
I believe we'll need to switch to the REST JSON API, which gertty uses
and gets comments -> some nontrivial refactoring implied, and I
shudder to think about the performance impact.
I've wanted to rewrite reviewstats for a while... this might be the trigger...
-Rob
have expressed this in the past, and I think it's one of those
> things that requires vigilance and periodic reminders.
We could publish a document of common bad metrics and why we as a
community reject them. That might give folk inside contributing
companies some additional support in convincing
wstats [1] to count
> +0 comments and I think it would be good for stackalytics to consider
> the same.
>
> [1] http://git.openstack.org/cgit/openstack-infra/reviewstats
IIRC the limit was gerrit, not reviewstats, and we've not revisited
since the gerrit up
tup.py
install fails as a reason to say 'use pip install .'.
I'd like to make that a little more official:
- put it in our docs
- stop testing python setup.py install.
This would relegate setup.py to building, not installation, which is
what its intended to be these da
#/c/165798/
>
> --
> Julien Danjou
> -- Free Software hacker
> -- http://julien.danjou.info
>
> __
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
> http://lists.op
itself. I'm working on that now,
working on improvements to pip to enable much better robustness in our
gate.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List
/1446847
Endish of my workday, but https://review.openstack.org/#/c/176113/ has
some promise. It doesn't install the 'right' versions- thats something
we're going to have to circle back on, but it does run the tests
success
o justice to both the PTL and TC
offices. If elected I'll be able to devote the majority of my time to
the TC and related issues - making me much more effective - so I can
help shift the dial on all the challenges we face.
-R
de guidance.
Cheers,
Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
h
used oslo.version, and it vanished. Anyway, pbr is
> actually very small, so I don't think it's an issue.
>
> On 04/20/2015 02:22 PM, Victor Stinner wrote:
>> I read somewhere that pkg_resources may also be used to get the
>> version.
>
> That's correct,
.3 should work with 1.5. Why doesn't it?
> Details of the version numbers will be hashed out at the summit and the
> release management team. (E.g do we move to a 2015.x.x model? Do we
> increment the X per release in semver, or th
aces.
Yes, that was exactly my concern.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ..
bugs, file them on https://launchpad.net/pbr - don't
worry about looking for existing dups, I'll treat each one as a
support request, and dup it if-and-only-if its truely a dupe.
-Rob
--
Robert Collins
Distinguis
resources.
> I read somewhere that pkg_resources may also be used to get the version.
Yes, and pbr does that, when the package is installed.
So - I still don't understand *why* RDO is patching it out.
It needs one of the RDO packagers to file a bug describing what
happens in sufficien
27;t think its a bug in the applications.
Maybe someone from RedHat can file a bug on pbr describing what goes wrong?
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing Li
I'd like to release 1.0.0 just
to get us into 'ok its stable' mentality.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage qu
nstance.
This will be better than that was, but only because its being
explicitly documented as such.
Could you enlarge on what 'coordinating the requirements' means?
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
___
te the configs, start the services -
seems straight forward, good for maintenance, and a pretty low bar.
Maybe even migrate some of it to Python at the same time :).
> You are right, https://github.com/pypa/pip/issues/988 "Pip needs a
> dependency resolver" ends up being a problem fo
some, but it _does_ work. Without that, pypi suddenly is sectioned
> off into islands the moment a popular library narrows its requirements.
There's already --no-deps to give folk an escape clause in the pip
world. The coinstallability problem is not created by the tool
observing it
On 16 April 2015 at 01:50, Sean Dague wrote:
> On 04/12/2015 06:43 PM, Robert Collins wrote:
>> Thoughts? If there's broad apathy-or-agreement I can turn this into a
>> spec for fine coverage of ramifications and corner cases.
>
> I'm definitely happy someone
le branches, and captured output to a
common place for consumption by deployers/testers
- looser gate lists for master branches, captured in the same way to
the same place as stable branches, for deployers/testsers
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
_
On 15 April 2015 at 09:35, Joe Gordon wrote:
>
>
> On Sun, Apr 12, 2015 at 6:09 PM, Robert Collins
> wrote:
>>
>> On 13 April 2015 at 12:53, Monty Taylor wrote:
>>
>> > If we pin the stable branches with hard pins of direct and indirect
>> > depen
On 15 April 2015 at 09:33, Joe Gordon wrote:
>
>
> On Tue, Apr 14, 2015 at 2:36 AM, Thierry Carrez
> wrote:
>>
>> Robert Collins wrote:
>> > On 13 April 2015 at 22:04, Thierry Carrez wrote:
>> >> How does this proposal affect stable branches ? In
On 14 April 2015 at 21:36, Thierry Carrez wrote:
> Robert Collins wrote:
>> On 13 April 2015 at 22:04, Thierry Carrez wrote:
>>> How does this proposal affect stable branches ? In order to keep the
>>> breakage there under control, we now have stable branches for all
On 15 April 2015 at 01:11, Sylvain Bauza wrote:
>
>
> Le 14/04/2015 14:49, Alexis Lee a écrit :
>>
>> Robert Collins said on Thu, Apr 09, 2015 at 11:26:28AM +1200:
>>>
>>> Neutron was using 'unittest', but we depend on the fixes in cPython
>>
> how to get them to a better state.
https://review.openstack.org/#/c/124651/6 should make master releasable.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing Lis
x27;d love it if the conversation focused on what that needs,
rather than 'ripping it out' - since I really loath the copy-paste
hell that I fear that will lead to.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
100ms issue above is either:
- the package isn't installed, so we're falling back to complex code
- a bug.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing Li
date based its
harder to predict what will indicate a breaking version :). And of
course for non-semver, 1.2.3 doesn't tell us whether 1.3 will be
breaking, or even 1.2.4.
- a known good version to base our cap on
If we generated the first item and stored it somewhere, then when we
generate
On 13 April 2015 at 13:09, Robert Collins wrote:
> On 13 April 2015 at 12:53, Monty Taylor wrote:
>
>> What we have in the gate is the thing that produces the artifacts that
>> someone installing using the pip tool would get. Shipping anything with
>> those artif
- someone can propose it. Otherwise, the stable
> release should not be moving.
Can we do that in stable branches? We've still got the problem of
bumping dependencies across multiple packages.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
_
cerned - but about what specifically? What goes wrong
with wheels (not wheels with C code). Whats not robust about the
pattern? The cargo package manager you referred to is entirely
declarative
James: I don't think the binary distribution stuff is relevant t
an turn this into a
spec for fine coverage of ramifications and corner cases.
-Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage questions)
Uns
nittest'
itself anywhere in OpenStack, only 'unittest2'.
testtools has a hard dependency on unittest2 these days (as we port
features from it into cPython's unittest, they come back as backports
in unittest2), so there's no dependency-chain reason to avoid using
uni
On 8 Apr 2015 4:49 pm, "Michael Still" wrote:
>
> It is my sad duty to inform the community that Chris Yeoh passed away
this morning. Chris
Oh crap. Thank you for letting us know in such a caring way. Vale, Chris.
-Rob
tem do you use to do your OpenStack development
Cheers,
Rob
--
Robert Collins
Distinguished Technologist
HP Converged Cloud
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openst
301 - 400 of 1124 matches
Mail list logo
