ithub.com/apache/mesos/commit/23d717741df7ca291270b762b7b93a37b4a144ef>
> and
> no one has ever thought about setting the field before. Your use case seems
> reasonable and I see no reason why Aurora cannot set this value and modify
> the executor appropriately.
>
>
>> On Thu, Ma
One pain point that currently exists with Aurora/Mesos/Docker integration
is that it requires making a choice between two bad options:
1) require that the aurora role exist in the docker image as a unix user
2) run everything as root by setting "USER root" in the Dockerfile and pass
--nosetuid to
Congrats everyone!!
The only negative to this is all my aurora mail rules and git remotes are
going to break!
Jay
On Thu, Mar 19, 2015 at 2:29 AM, Larry Weya wrote:
> Congratulations, definitely well deserved. Great product and excellent
> support from the team.
> On 19 Mar 2015 04:22, "Toby W
ibute to it, if necessary - definitely for testing it out
>> :)
>> >
>> > Let us know what we could do.
>> >
>> > Dobromir
>> >
>> > On Thu, Oct 16, 2014 at 1:50 PM, Jay Buffington wrote:
>> >
>> > > I'm
I'm hoping that AURORA-633 will be up for code review in November.
I believe the docker containerizer is the best way to use docker with
mesos (compared to other options like a docker executor or using the
external containerizer).
Right now you can write an Aurora job that has processes which exe
t; > received a broadcasted recover request
>> > I1016 16:12:30.623085 26071 recover.cpp:188] Received a recover response
>> > from a replica in EMPTY status
>> >
>> > /***
>> > Joe Stein
>> > Founder, Principal
This was fixed about three months ago:
https://issues.apache.org/jira/browse/AURORA-584 Perhaps you're
running an older version that doesn't have that commit?
The tl;dr of that jira is that when you go to a scheduler that isn't
the leader it should do a HTTP redirect (302?) to the leader.
Jay
I'm not sure it should be removed, but I've also found this very
confusing. Also, it's only applied for jobs that have service=True
production=True. At a minimum we need to get it documented.
Jay
On Thu, Sep 4, 2014 at 5:01 PM, Bill Farner wrote:
> FWIW the same goes for the 'host' constraint
To reiterate Bill's point about preemption: if you add a second
framework then when you submit jobs that have production=True Aurora
will kill jobs that have production=False (the default) unnecessarily.
Since Aurora will not have a complete view of all of the available
resources, it will think the
On Tue, Aug 12, 2014 at 1:18 PM, Bill Farner wrote:
> - read/write, persistent data
>
> There's not a great story here, at least not in the traditional
> mesos/aurora "run anywhere" model. The best approach we have so far is to
> mount to a directory owned by the user that will access the data, c
On Thu, Jun 5, 2014 at 10:02 AM, Henry Saputra
wrote:
>
> But seemed like no TCP bound to http://192.168.33.7:8081
>
>
Is the aurora-scheduler process running? You can check like this:
$ ps aux |grep [a]urora
root 19711 0.0 0.0 4116 488 ?S16:47 0:00 cronolog
--link=/var/
On Tue, May 20, 2014 at 9:14 AM, Mark Chu-Carroll wrote:
> I agree. At the moment, the update process is completely driven by the
> client, which creates some odd usability issues.
>
If we ran updates primarily on the server, they could be handled
> asynchronously or synchronously in a way con
>
>
> > ## Standard Shortcuts
> >
> > To save typing, we'll add _automatic_ shorthand generation to
> > the command-line framework. For nouns and verbs, any unambiguous
> > prefix of the appropriate word will be automatically expanded to
> > the full word.
> >
> >* `aurora j c` will be expanded
developers to engage. Meetings are every Monday at http://www.timeanddate.com/worldclock/fixedtime.html?msg=Aurora+Weekly+Meeting&iso=20140421T11&p1=224&ah=1";>11am
Pacific Time. See below for IRC channel details.
On Wed, Apr 16, 2014 at 3:37 PM, Jay Buffington wrot
+1 I like the date and time.
ASFBot has meeting features that will, among other things, automatically
generate meeting logs:
http://wilderness.apache.org/manual.html#meetings
On Wed, Apr 16, 2014 at 3:23 PM, Dave Lester wrote:
> +1!
>
> I have ideas regarding the meeting structure and observa
Hi,
I need to provide isolation between my mesos tasks running via Aurora
because they each expect a unique view of mounts. I've settled on a simple
approach of just using the unshare syscall to create a new mount namespace,
calling mount to create the mounts the app expects, then exec'ing the
ap
On Wed, Apr 9, 2014 at 1:55 PM, Kevin Sweeney wrote:
> My thought is that if I want to build the source distributions that got
> published for release 0.5.0 I should be able to clone the repo, run git
> checkout 0.5.0, ./gradlew ...; ./pants ... and produce the same things we
> publish to the apa
On Mon, Apr 7, 2014 at 8:05 PM, Kevin Sweeney wrote:
>
> If I create a tag when .auroraversion is updated to a new -snapshot on
> master like so (we haven't had a release yet but assume I did this during
> the release of 0.4.0):
>
> % git tag -a -m 'Create 0.5.0-snapshot' 0.5.0-snapshot
> 19e0875
On Mon, Apr 7, 2014 at 8:35 PM, Bill Farner wrote:
> Plucking from the man page:
>
> NAME
>git-describe - Show the most recent tag that is reachable from a
> commit
>
> Does this proposal fall apart if/when we start using tags for other
> purposes?
>
Not as long as you use annotated tags
19 matches
Mail list logo
