Re: master/CiaB: Wait for juju services to have open ports - Timeout when waiting for services

2017-01-03 Thread Mac Lin 
Really appreciate for the help. Have been stuck here for weeks.

AFAIK, no. I attached two result of "juju status" on CloudLab(good) and my
server (bad). It's supposed to work without expose the ports. And I just
noticed that on my server, most of the services are in blocked status. But
it seems the missing parts are fulfilled. For example, the ceilometer is
complaining missing messaging, identity, and database relations, but the
relation does exist.

Please let me know if any info needed.

  ceilometer:
charm: cs:trusty/ceilometer-17
exposed: false
service-status:
  current: blocked
  message: 'Missing relations: messaging, identity, database'
  since: 03 Jan 2017 19:39:55Z
relations:
  amqp:
  - rabbitmq-server
  ceilometer-service:
  - ceilometer-agent
  cluster:
  - ceilometer
  identity-service:
  - keystone
  juju-info:
  - nagios
  nrpe-external-master:
  - nrpe
  shared-db:
  - mongodb
units:
  ceilometer/0:
workload-status:
  current: blocked
  message: 'Missing relations: messaging, identity, database'
  since: 03 Jan 2017 19:39:55Z
agent-status:
  current: executing
  message: running install hook
  since: 03 Jan 2017 14:53:31Z
  version: 1.25.9
agent-state: started
agent-version: 1.25.9
machine: "9"
open-ports:
- 8777/tcp
public-address: ceilometer.cord.lab


On Tue, Jan 3, 2017 at 8:23 PM, Rick Harding 
wrote:

> Has juju expose been run on the applications? The charm can declare that
> these ports are the ones that are opened if exposed, it's not actually set
> until the operator runs the juju expose command on the application.
>
> On Sun, Jan 1, 2017 at 2:58 AM Mac Lin  wrote:
>
>>
>> Hi,
>>
>> I'm running CORD master/cord-in-a-box.sh on an x86_64 server. The same
>> script on CloudLab has no problem.
>>
>> If I log into each failed service(lxc), I could found the port is not
>> listened because service is not up, which is due to not configured
>> properly. The configuration files are mostly at its initial state.
>>
>> How could I let juju to generate the configuration for the services
>> manually?
>>
>> TASK [juju-finish : Wait for juju services to have open ports]
>> *
>> Saturday 31 December 2016  12:04:22 + (0:00:04.738)   0:09:13.981
>> *
>> failed: [10.100.198.201] (item={u'service': u'ceilometer',
>> u'ipv4_last_octet': 20, u'name': u'ceilometer-1', u'forwarded_ports':
>> [{u'int': 8777, u'ext': 8777}], u'aliase
>> s': [u'ceilometer']}) => {"elapsed": 1800, "failed": true, "item":
>> {"aliases": ["ceilometer"], "forwarded_ports": [{"ext": 8777, "int":
>> 8777}], "ipv4_last_octet": 20, "n
>> ame": "ceilometer-1", "service": "ceilometer"}, "msg": "Timeout when
>> waiting for ceilometer-1:8777"}
>> failed: [10.100.198.201] (item={u'service': u'glance',
>> u'ipv4_last_octet': 30, u'name': u'glance-1', u'forwarded_ports': [{u'int':
>> 9292, u'ext': 9292}], u'aliases': [u'g
>> lance']}) => {"elapsed": 1800, "failed": true, "item": {"aliases":
>> ["glance"], "forwarded_ports": [{"ext": 9292, "int": 9292}],
>> "ipv4_last_octet": 30, "name": "glance-1"
>> , "service": "glance"}, "msg": "Timeout when waiting for glance-1:9292"}
>> failed: [10.100.198.201] (item={u'service': u'keystone',
>> u'ipv4_last_octet': 40, u'name': u'keystone-1', u'forwarded_ports':
>> [{u'int': 35357, u'ext': 35357}, {u'int': 49
>> 90, u'ext': 4990}, {u'int': 5000, u'ext': 5000}], u'aliases':
>> [u'keystone']}) => {"elapsed": 1800, "failed": true, "item": {"aliases":
>> ["keystone"], "forwarded_ports": [
>> {"ext": 35357, "int": 35357}, {"ext": 4990, "int": 4990}, {"ext": 5000,
>> "int": 5000}], "ipv4_last_octet": 40, "name": "keystone-1", "service":
>> "keystone"}, "msg": "Timeo
>> ut when waiting for keystone-1:35357"}
>> ok: [10.100.198.201] => (item={u'service': u'nagios', u'ipv4_last_octet':
>> 60, u'name': u'nagios-1', u'forwarded_ports': [{u'int': 80, u'ext': 3128}],
>> u'aliases': [u'nagi
>> os']})
>> failed: [10.100.198.201] (item={u'service': u'neutron-api',
>> u'ipv4_last_octet': 70, u'name': u'neutron-api-1', u'forwarded_ports':
>> [{u'int': 9696, u'ext': 9696}], u'alia
>> ses': [u'neutron-api']}) => {"elapsed": 1800, "failed": true, "item":
>> {"aliases": ["neutron-api"], "forwarded_ports": [{"ext": 9696, "int":
>> 9696}], "ipv4_last_octet": 70
>> , "name": "neutron-api-1", "service": "neutron-api"}, "msg": "Timeout
>> when waiting for neutron-api-1:9696"}
>> failed: [10.100.198.201] (item={u'service': u'nova-cloud-controller',
>> u'ipv4_last_octet': 80, u'name': u'nova-cloud-controller-1',
>> u'forwarded_ports': [{u'int': 8774, u'
>> ext': 8774}], u'aliases': [u'nova-cloud-controller']}) => {"elapsed":
>> 1800, "failed": true, "item": {"aliases": ["nova-cloud-controller"],
>> "forwarded_ports": [{"ext": 87
>> 74, "int":

Re: juju bootstrap complaining about "no registered provider for"

2017-01-03 Thread John Meinel 
That sounds like the juju client knows about the provider but the agent it
is installing (jujud) doesn't.  You may need to change the version of the
client so it guarantees it tries to install the developer build of the
agent rather than an official released agent.

John
=:->

On Jan 4, 2017 02:23, "Narinder Gupta"  wrote:

> Hi juju Team,
> Lenovo team is trying to write a new provider with their xclarity product.
> Currently they are experimenting it mock-provider and did some coding and
> trying to bootstrap the nodes  and facing few issues.
>
> Do you think we can give Lenovo team some guidance on how to write a new
> provider? Also can rectify current issues.
>
> Hi Fang,
> Happy new year.
> juju-dev@lists.ubuntu.com is the correct mailing list you can pose your
> questions about juju providers. Also you can always connect to irc channel
> #juju.
>
> Thanks and Regards,
> Narinder Gupta (PMP)   narinder.gu...@canonical.com
> Canonical, Ltd.narindergupta 
> [irc.freenode.net]+1.281.736.5150 <(281)%20736-5150>  
>   narindergupta2007[skype]
>
> Ubuntu- Linux for human beings | www.ubuntu.com | www.canonical.com
>
>
> On Tue, Jan 3, 2017 at 3:50 PM, Feng Xia1  wrote:
>
>>
>>
>> Hi Narinder,
>>
>>
>>
>> Happy new year!
>>
>>
>>
>> I’m on the path of getting deeper into the juju providers. I put together
>> a mock-up provider, issued “juju bootstrap”, manually started a VM with the
>> IP that juju bootstrap is expecting, and manually put correct nonce.txt on
>> that VM.
>>
>>
>>
>> By this point, juju deploy command happily ssh-ed into this VM, started
>> cloud-init scripts. Everything is good until at the end I get this error:
>>
>>
>> “no registered provider for…”
>>
>>
>>
>> The error message came from juju/juju/environs/config.go, function “func
>> (r *globalProviderRegistry) Provider(providerType string) (EnvironProvider,
>> error) {…”. However, the global registry does have an entry for each of the
>> cloud types.
>>
>>
>>
>> What did I miss?
>>
>>
>>
>> [image:
>> http://lenovocentral.lenovo.com/marketing/branding/email_signature/images/gradient.gif]
>>
>> *Feng xia*
>> Engineer
>> Lenovo USA
>>
>> [image: Phone]5088011794 <(508)%20801-1794>
>> [image: Email]fx...@lenovo.com 
>>
>>
>>
>> Lenovo.com  
>> Twitter  | Facebook
>>  | Instagram
>>  | Blogs  | Forums
>> 
>>
>> [image: ImageLogo-Holiday-Lights]
>>
>>
>>
>>
>>
>
>
> --
> Juju mailing list
> j...@lists.ubuntu.com
> Modify settings or unsubscribe at: https://lists.ubuntu.com/
> mailman/listinfo/juju
>
>
-- 
Juju-dev mailing list
Juju-dev@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju-dev

Re: juju bootstrap complaining about "no registered provider for"

2017-01-03 Thread John Meinel 
That sounds like the juju client knows about the provider but the agent it
is installing (jujud) doesn't.  You may need to change the version of the
client so it guarantees it tries to install the developer build of the
agent rather than an official released agent.

John
=:->

On Jan 4, 2017 02:23, "Narinder Gupta"  wrote:

> Hi juju Team,
> Lenovo team is trying to write a new provider with their xclarity product.
> Currently they are experimenting it mock-provider and did some coding and
> trying to bootstrap the nodes  and facing few issues.
>
> Do you think we can give Lenovo team some guidance on how to write a new
> provider? Also can rectify current issues.
>
> Hi Fang,
> Happy new year.
> juju-...@lists.ubuntu.com is the correct mailing list you can pose your
> questions about juju providers. Also you can always connect to irc channel
> #juju.
>
> Thanks and Regards,
> Narinder Gupta (PMP)   narinder.gu...@canonical.com
> Canonical, Ltd.narindergupta 
> [irc.freenode.net]+1.281.736.5150 <(281)%20736-5150>  
>   narindergupta2007[skype]
>
> Ubuntu- Linux for human beings | www.ubuntu.com | www.canonical.com
>
>
> On Tue, Jan 3, 2017 at 3:50 PM, Feng Xia1  wrote:
>
>>
>>
>> Hi Narinder,
>>
>>
>>
>> Happy new year!
>>
>>
>>
>> I’m on the path of getting deeper into the juju providers. I put together
>> a mock-up provider, issued “juju bootstrap”, manually started a VM with the
>> IP that juju bootstrap is expecting, and manually put correct nonce.txt on
>> that VM.
>>
>>
>>
>> By this point, juju deploy command happily ssh-ed into this VM, started
>> cloud-init scripts. Everything is good until at the end I get this error:
>>
>>
>> “no registered provider for…”
>>
>>
>>
>> The error message came from juju/juju/environs/config.go, function “func
>> (r *globalProviderRegistry) Provider(providerType string) (EnvironProvider,
>> error) {…”. However, the global registry does have an entry for each of the
>> cloud types.
>>
>>
>>
>> What did I miss?
>>
>>
>>
>> [image:
>> http://lenovocentral.lenovo.com/marketing/branding/email_signature/images/gradient.gif]
>>
>> *Feng xia*
>> Engineer
>> Lenovo USA
>>
>> [image: Phone]5088011794 <(508)%20801-1794>
>> [image: Email]fx...@lenovo.com 
>>
>>
>>
>> Lenovo.com  
>> Twitter  | Facebook
>>  | Instagram
>>  | Blogs  | Forums
>> 
>>
>> [image: ImageLogo-Holiday-Lights]
>>
>>
>>
>>
>>
>
>
> --
> Juju mailing list
> Juju@lists.ubuntu.com
> Modify settings or unsubscribe at: https://lists.ubuntu.com/
> mailman/listinfo/juju
>
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju

Re: juju bootstrap complaining about "no registered provider for"

2017-01-03 Thread Narinder Gupta 
Hi juju Team,
Lenovo team is trying to write a new provider with their xclarity product.
Currently they are experimenting it mock-provider and did some coding and
trying to bootstrap the nodes  and facing few issues.

Do you think we can give Lenovo team some guidance on how to write a new
provider? Also can rectify current issues.

Hi Fang,
Happy new year.
juju-...@lists.ubuntu.com is the correct mailing list you can pose your
questions about juju providers. Also you can always connect to irc channel
#juju.

Thanks and Regards,
Narinder Gupta (PMP)   narinder.gu...@canonical.com
Canonical, Ltd.narindergupta
[irc.freenode.net]+1.281.736.5150 <(281)%20736-5150>
 narindergupta2007[skype]

Ubuntu- Linux for human beings | www.ubuntu.com | www.canonical.com


On Tue, Jan 3, 2017 at 3:50 PM, Feng Xia1  wrote:

>
>
> Hi Narinder,
>
>
>
> Happy new year!
>
>
>
> I’m on the path of getting deeper into the juju providers. I put together
> a mock-up provider, issued “juju bootstrap”, manually started a VM with the
> IP that juju bootstrap is expecting, and manually put correct nonce.txt on
> that VM.
>
>
>
> By this point, juju deploy command happily ssh-ed into this VM, started
> cloud-init scripts. Everything is good until at the end I get this error:
>
>
> “no registered provider for…”
>
>
>
> The error message came from juju/juju/environs/config.go, function “func
> (r *globalProviderRegistry) Provider(providerType string) (EnvironProvider,
> error) {…”. However, the global registry does have an entry for each of the
> cloud types.
>
>
>
> What did I miss?
>
>
>
> [image:
> http://lenovocentral.lenovo.com/marketing/branding/email_signature/images/gradient.gif]
>
> *Feng xia*
> Engineer
> Lenovo USA
>
> [image: Phone]5088011794 <(508)%20801-1794>
> [image: Email]fx...@lenovo.com 
>
>
>
> Lenovo.com  
> Twitter  | Facebook
>  | Instagram
>  | Blogs  | Forums
> 
>
> [image: ImageLogo-Holiday-Lights]
>
>
>
>
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju

Re: using a bundle with manually added machines

2017-01-03 Thread Nate Finch 
FWIW, I think this *could* work, but it may be that we just haven't coded
it up.   Just as *juju deploy mysql -n 4* will use existing clean machines
in the model where appropriate, deploying a bundle should as well
(presuming the existing machines match constraints etc as usual).

On Tue, Jan 3, 2017 at 9:08 AM Rick Harding 
wrote:

> Hi Vance, you can deploy a bundle that uses existing machines in the model
> with the juju-deployer [1] tool.
>
> The built in juju deploy method is the first stage in a generic tool and
> does not allow pointing at existing machines because it means the bundles
> are not sharable and that you need your model in a specific state before
> you can use the bundle. Since Juju doesn't allow you to rename machines it
> means that you have to keep the bundle in sync with the model.
>
> 1: https://pypi.python.org/pypi/juju-deployer/
>
> On Thu, Dec 29, 2016 at 9:09 AM Vance Morris  wrote:
>
>
> Hi all,
>
> Is it possible to use a bundle.yaml in conjunction with manually added
> machines?
>
> $ juju version
> 2.0.2-xenial-s390x
>
> $ juju status
> << snip >>
> Machine  StateDNS  Inst id  Series  AZ
> 0started  REDACTED  manual:REDACTED   xenial
> 1started  REDACTED  manual:REDACTED   xenial
> 2started  REDACTED  manual:REDACTED   xenial
> << snip >>
>
> When I go to deploy a bundle that defines machines 0, 1, and 2 and
> describes deployment of services to these machines, the charms are
> deployed, but no units are created.
>
> ERROR cannot deploy bundle: cannot create machine for holding aodh,
> ceilometer, ceph-mon, ceph-osd, cinder, glance, keystone, mongodb, mysql,
> neutron-api, neutron-gateway, nova-cloud-controller, nova-compute,
> openstack-dashboard, rabbitmq-server, swift-proxy and swift-storage-z1
> units: cannot add a new machine: use "juju add-machine ssh:[user@]"
> to provision machines
>
> If I manually deploy charms to the manually added machines, it's happy to
> oblige.
>
>
> VANCE MORRIS
> --
> Phone:  1-720-349-9450 <(720)%20349-9450>
> E-mail: vmor...@us.ibm.com
> [image: IBM]
> 
> 
> 
> 1177 S Belt Line Rd
> Coppell, TX 75019-4642
> United States
>
> --
> Juju mailing list
> Juju@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/juju
>
> --
> Juju mailing list
> Juju@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/juju
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju

Re: Juju Wikipedia Needs Update

2017-01-03 Thread Rick Harding 
Thanks for the heads up. I'll take a look at it. I've wanted to write for
wikipedia sometime.

On Tue, Jan 3, 2017 at 12:55 PM James Beedy  wrote:

> Wikipedia entry for Juju needs update
> https://en.m.wikipedia.org/wiki/Juju_(software)
>
> --
> Juju-dev mailing list
> Juju-dev@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/juju-dev
>
-- 
Juju-dev mailing list
Juju-dev@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju-dev

Juju Wikipedia Needs Update

2017-01-03 Thread James Beedy 
Wikipedia entry for Juju needs update 
https://en.m.wikipedia.org/wiki/Juju_(software)

-- 
Juju-dev mailing list
Juju-dev@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju-dev

Re: Kubernetes Provider

2017-01-03 Thread Tom Barber 
Same question got asked yesterday(lxd on kub). Marks answer was it's not
roadmapped but make it great and someone will make it happen.

On 3 Jan 2017 17:18, "James Beedy"  wrote:

> Is there any work currently being done out there for a kub provider? Has
> anyone looked into this yet?
> --
> Juju mailing list
> Juju@lists.ubuntu.com
> Modify settings or unsubscribe at: https://lists.ubuntu.com/
> mailman/listinfo/juju
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju

Re: Charm release fails with "cannot update base entity for..."

2017-01-03 Thread Merlijn Sebrechts 
This is it:

https://jujucharms.com/u/tengu-team/jupiter-notebook-spark/0

Strangely, only the jup*i*ter-notebook-spark charm is visible, the
jup*y*ter-notebook
charm isn't even though it has the right permissions.

perm:
  Read:
  - tengu-team
  - everyone
  Write:
  - tengu-team


2017-01-03 15:03 GMT+01:00 Marco Ceppi :

> Is this charm or bundle somewhere we can look at?
>
> On Tue, Jan 3, 2017 at 8:06 AM Merlijn Sebrechts <
> merlijn.sebrec...@gmail.com> wrote:
>
>> I have the same issue in the staging charm store.
>>
>>
>> charm release  cs:~tengu-team/jupyter-notebook-spark-0
>> ERROR cannot release charm or bundle: cannot publish charm or bundle:
>> cannot update base entity for "cs:~tengu-team/jupyter-notebook-spark-0":
>> Field name duplication not allowed with modifiers
>>
>>
>> Maybe this has something to do with it; the first time I pushed this
>> charm I pushed to a wrong name (jupIter instead of jupYter):
>>
>> charm push jup*y*ter-notebook-spark/ ~tengu-team/jup*i*ter-notebook-spark
>>
>> Although I did it right the first time in the staging charm store, and it
>> still doesn't work.
>>
>> Also, I'm user `merlijn-sebrechts` and I'm pushing to `tengu-team`. This
>> has worked before.
>>
>> 2017-01-03 13:45 GMT+01:00 roger peppe :
>>
>> Interesting. Thanks for the report.
>>
>> That's an error I don't think we've seen before, and I can't reproduce
>> the problem immediately
>> on one of my own charms.
>>
>> Do you see the same issue if you try and publish and release the same
>> charm in the
>> staging charmstore (set
>> JUJU_CHARMSTORE=https://api.staging.jujucharms.com/charmstore
>> before running the charm command) ?
>>
>>   cheers
>> rog.
>>
>> On 3 January 2017 at 12:17, Merlijn Sebrechts
>>  wrote:
>> > Hi all
>> >
>> >
>> > When releasing my charm I get the following error:
>> >
>> > charm release cs:~tengu-team/jupiter-notebook-spark-0
>> > ERROR cannot release charm or bundle: cannot publish charm or bundle:
>> cannot
>> > update base entity for "cs:~tengu-team/jupiter-notebook-spark-0":
>> Field name
>> > duplication not allowed with modifiers
>> >
>> > A quick google shows that this is a MongoDB error. Is something wrong
>> with
>> > the cs backend?
>> >
>> >
>> >
>> > Kind regards
>> > Merlijn
>> >
>> > --
>> > Juju mailing list
>> > Juju@lists.ubuntu.com
>> > Modify settings or unsubscribe at:
>> > https://lists.ubuntu.com/mailman/listinfo/juju
>> >
>>
>>
>> --
>> Juju mailing list
>> Juju@lists.ubuntu.com
>> Modify settings or unsubscribe at: https://lists.ubuntu.com/
>> mailman/listinfo/juju
>>
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju

Re: using a bundle with manually added machines

2017-01-03 Thread Rick Harding 
Hi Vance, you can deploy a bundle that uses existing machines in the model
with the juju-deployer [1] tool.

The built in juju deploy method is the first stage in a generic tool and
does not allow pointing at existing machines because it means the bundles
are not sharable and that you need your model in a specific state before
you can use the bundle. Since Juju doesn't allow you to rename machines it
means that you have to keep the bundle in sync with the model.

1: https://pypi.python.org/pypi/juju-deployer/

On Thu, Dec 29, 2016 at 9:09 AM Vance Morris  wrote:

>
> Hi all,
>
> Is it possible to use a bundle.yaml in conjunction with manually added
> machines?
>
> $ juju version
> 2.0.2-xenial-s390x
>
> $ juju status
> << snip >>
> Machine  StateDNS  Inst id  Series  AZ
> 0started  REDACTED  manual:REDACTED   xenial
> 1started  REDACTED  manual:REDACTED   xenial
> 2started  REDACTED  manual:REDACTED   xenial
> << snip >>
>
> When I go to deploy a bundle that defines machines 0, 1, and 2 and
> describes deployment of services to these machines, the charms are
> deployed, but no units are created.
>
> ERROR cannot deploy bundle: cannot create machine for holding aodh,
> ceilometer, ceph-mon, ceph-osd, cinder, glance, keystone, mongodb, mysql,
> neutron-api, neutron-gateway, nova-cloud-controller, nova-compute,
> openstack-dashboard, rabbitmq-server, swift-proxy and swift-storage-z1
> units: cannot add a new machine: use "juju add-machine ssh:[user@]"
> to provision machines
>
> If I manually deploy charms to the manually added machines, it's happy to
> oblige.
>
>
> VANCE MORRIS
> --
> Phone:  1-720-349-9450 <(720)%20349-9450>
> E-mail: vmor...@us.ibm.com
> [image: IBM]
> 
> 
> 
> 1177 S Belt Line Rd
> Coppell, TX 75019-4642
> United States
>
> --
> Juju mailing list
> Juju@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/juju
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju

Re: Charm release fails with "cannot update base entity for..."

2017-01-03 Thread Marco Ceppi 
Is this charm or bundle somewhere we can look at?

On Tue, Jan 3, 2017 at 8:06 AM Merlijn Sebrechts <
merlijn.sebrec...@gmail.com> wrote:

> I have the same issue in the staging charm store.
>
>
> charm release  cs:~tengu-team/jupyter-notebook-spark-0
> ERROR cannot release charm or bundle: cannot publish charm or bundle:
> cannot update base entity for "cs:~tengu-team/jupyter-notebook-spark-0":
> Field name duplication not allowed with modifiers
>
>
> Maybe this has something to do with it; the first time I pushed this charm
> I pushed to a wrong name (jupIter instead of jupYter):
>
> charm push jup*y*ter-notebook-spark/ ~tengu-team/jup*i*ter-notebook-spark
>
> Although I did it right the first time in the staging charm store, and it
> still doesn't work.
>
> Also, I'm user `merlijn-sebrechts` and I'm pushing to `tengu-team`. This
> has worked before.
>
> 2017-01-03 13:45 GMT+01:00 roger peppe :
>
> Interesting. Thanks for the report.
>
> That's an error I don't think we've seen before, and I can't reproduce
> the problem immediately
> on one of my own charms.
>
> Do you see the same issue if you try and publish and release the same
> charm in the
> staging charmstore (set
> JUJU_CHARMSTORE=https://api.staging.jujucharms.com/charmstore
> before running the charm command) ?
>
>   cheers
> rog.
>
> On 3 January 2017 at 12:17, Merlijn Sebrechts
>  wrote:
> > Hi all
> >
> >
> > When releasing my charm I get the following error:
> >
> > charm release cs:~tengu-team/jupiter-notebook-spark-0
> > ERROR cannot release charm or bundle: cannot publish charm or bundle:
> cannot
> > update base entity for "cs:~tengu-team/jupiter-notebook-spark-0": Field
> name
> > duplication not allowed with modifiers
> >
> > A quick google shows that this is a MongoDB error. Is something wrong
> with
> > the cs backend?
> >
> >
> >
> > Kind regards
> > Merlijn
> >
> > --
> > Juju mailing list
> > Juju@lists.ubuntu.com
> > Modify settings or unsubscribe at:
> > https://lists.ubuntu.com/mailman/listinfo/juju
> >
>
>
> --
> Juju mailing list
> Juju@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/juju
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju

Re: Charm release fails with "cannot update base entity for..."

2017-01-03 Thread Merlijn Sebrechts 
I have the same issue in the staging charm store.


charm release  cs:~tengu-team/jupyter-notebook-spark-0
ERROR cannot release charm or bundle: cannot publish charm or bundle:
cannot update base entity for "cs:~tengu-team/jupyter-notebook-spark-0":
Field name duplication not allowed with modifiers


Maybe this has something to do with it; the first time I pushed this charm
I pushed to a wrong name (jupIter instead of jupYter):

charm push jup*y*ter-notebook-spark/ ~tengu-team/jup*i*ter-notebook-spark

Although I did it right the first time in the staging charm store, and it
still doesn't work.

Also, I'm user `merlijn-sebrechts` and I'm pushing to `tengu-team`. This
has worked before.

2017-01-03 13:45 GMT+01:00 roger peppe :

> Interesting. Thanks for the report.
>
> That's an error I don't think we've seen before, and I can't reproduce
> the problem immediately
> on one of my own charms.
>
> Do you see the same issue if you try and publish and release the same
> charm in the
> staging charmstore (set
> JUJU_CHARMSTORE=https://api.staging.jujucharms.com/charmstore
> before running the charm command) ?
>
>   cheers
> rog.
>
> On 3 January 2017 at 12:17, Merlijn Sebrechts
>  wrote:
> > Hi all
> >
> >
> > When releasing my charm I get the following error:
> >
> > charm release cs:~tengu-team/jupiter-notebook-spark-0
> > ERROR cannot release charm or bundle: cannot publish charm or bundle:
> cannot
> > update base entity for "cs:~tengu-team/jupiter-notebook-spark-0": Field
> name
> > duplication not allowed with modifiers
> >
> > A quick google shows that this is a MongoDB error. Is something wrong
> with
> > the cs backend?
> >
> >
> >
> > Kind regards
> > Merlijn
> >
> > --
> > Juju mailing list
> > Juju@lists.ubuntu.com
> > Modify settings or unsubscribe at:
> > https://lists.ubuntu.com/mailman/listinfo/juju
> >
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju

Re: using a bundle with manually added machines (redux)

2017-01-03 Thread Rick Harding 
I'm looking into this. The bundle deploy feature in Juju 2.0 does not allow
referring to existing machines because it breaks the reusability of the
bundle.

However, the manual provider is a bit unique in that it's how you get
machines into the system. The bundle deployment should work to pick up
machines that are not in use. However, since they're 'existing' it's
failing for you.

The reason Merlijn is able to do things is that in 1.25 you used the
juju-deployer tool to deploy bundles which supports a wide range of
customized, non-generic functions like pointing at existing machines.

On Tue, Jan 3, 2017 at 4:31 AM Merlijn Sebrechts <
merlijn.sebrec...@gmail.com> wrote:

> I've used the manual provider in Juju 1.x, and that functionality was
> working at the moment. This is one of the bundles that was working:
>
>
> https://raw.githubusercontent.com/IBCNServices/tengu-charms/master/bundles/streaming/bundle.yaml
>
> (ignore the "annotations" of the machines)
>
> 2016-12-29 20:16 GMT+01:00 Vance Morris :
>
> I'm going to assume that it's not working for me because I'm not using a
> cloud provider and instead manually adding the machines to Juju prior to
> deploying the bundle.
>
> I would think that this is a supported approach, but maybe I'm outside the
> bounds here...
>
> Thanks,
> Vance
>
> -Merlijn Sebrechts  wrote: -
> To: Vance Morris/Dallas/IBM@IBMUS
> From: Merlijn Sebrechts 
> Date: 12/29/2016 11:09AM
> Cc: "juju@lists.ubuntu.com" 
> Subject: Re: using a bundle with manually added machines (redux)
>
> Sorry, didn't see that, I was on my phone.
>
> The bundle looks correct to me, and everything seems in order when I
> import it into demo.jujucharms.com so I'm not sure why it doesn't work
> for you..
>
> 2016-12-29 17:13 GMT+01:00 Vance Morris :
> Greets Merlijn,
>
>  The bundle was attached to the original message, sorry I didn't call it
> out.
>
>  See here:
> https://lists.ubuntu.com/archives/juju/attachments/20161229/8b65108b/attachment.obj
>
>  Thanks,
>  Vance
>
>
>  -Merlijn Sebrechts  wrote: -
>  To: Vance Morris/Dallas/IBM@IBMUS
>  From: Merlijn Sebrechts 
>  Date: 12/29/2016 08:33AM
>  Cc: "juju@lists.ubuntu.com" 
>  Subject: Re: using a bundle with manually added machines (redux)
>
>
>  Defining the machines and defining the correct machines for each
> application using `to` should work.
>
>  The error message looks like some applications don't define machines so
> juju tries to create new machines which obviously fails. Is it possible to
> share the bundle that produces this error message and show the command you
> use to deploy that bundle?
>
>  Note that if an application has multiple units, you must specify multiple
> machines to deploy to.
>
>  Op donderdag 29 december 2016 heeft Vance Morris 
> het volgende geschreven:
>  > Whoops, sorry for that last message -- let's try this again!
>  >
>  > Hi all,
>  >
>  > Is it possible to use a bundle.yaml in conjunction with manually added
> machines?
>  >
>  > $ juju version
>  > 2.0.2-xenial-s390x
>  >
>  > $ juju status
>  > << snip >>
>  > Machine  StateDNS   Inst id   Series  AZ
>  > 0started  REDACTED  manual:REDACTED   xenial
>  > 1started  REDACTED  manual:REDACTED   xenial
>  > 2started  REDACTED  manual:REDACTED   xenial
>  > << snip >>
>  >
>  > When I go to deploy a bundle that defines machines 0, 1, and 2 and
> describes deployment of services to these machines, the charms are
> deployed, but no units are created.
>  >
>  > ERROR cannot deploy bundle: cannot create machine for holding aodh,
> ceilometer, ceph-mon, ceph-osd, cinder, glance, keystone, mongodb, mysql,
> neutron-api, neutron-gateway, nova-cloud-controller, nova-compute,
> openstack-dashboard, rabbitmq-server, swift-proxy and swift-storage-z1
> units: cannot add a new machine: use "juju add-machine ssh:[user@]"
> to provision machines
>  >
>  > If I manually deploy charms to the manually added machines, it's happy
> to oblige. Any suggestions?
>  >
>  >
>  > Sincerely,
>  >
>  > Vance Morris
>  > 1-720-349-9450 <(720)%20349-9450>
>  > vmor...@us.ibm.com
>  >
>
>
>
>
>
> --
> Juju mailing list
> Juju@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/juju
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju

Re: errors in my juju deploy of canonical distribution of kubernetes

2017-01-03 Thread Merlijn Sebrechts 
It might be a good idea to state in the README that you need to use
conjure-up for local clouds. These kind of things can indeed be confusing
for people new to this.

2017-01-02 15:36 GMT+01:00 brian mullan :

> Thx Marco... my concern was more for folks new to Canonical & following
> the online instructions. .. only to see it error out..?   Could sour them
> on the juju approach.
>
> On Jan 2, 2017 9:32 AM, "Marco Ceppi"  wrote:
>
>> Conjure-up does some things things in setting up pre-requisites for LXD
>> so that Kubernetes/Docker work as expected. In the next release we'll be
>> detecting these features and instead of having the applications error out,
>> be providing a more concise message as to why it's failed and how to
>> rectify.
>>
>> Marco
>>
>> On Sun, Jan 1, 2017 at 3:45 PM brian mullan 
>> wrote:
>>
>>> I've already used conjure-up for canonical-kubernetes and it DID install
>>> everything without the errors of the non conjure-up method.
>>>
>>> thanks Adam btw  Happy New Year !
>>>
>>> brian
>>>
>>>
>>> On Sun, Jan 1, 2017 at 2:32 PM, Adam Stokes 
>>> wrote:
>>>
>>> Hey Brian
>>>
>>> You'll want to use conjure-up for the localhost (LXD) as we provide some
>>> alterations to the lxc profile that houses those charms.
>>>
>>> On Sun, Jan 1, 2017, 2:26 PM brian mullan 
>>> wrote:
>>>
>>> Environment:
>>>
>>> ubuntu 16.04 server VM (in kvm)
>>>
>>> 50GB disk
>>> 8 GB ram
>>> 4 cpu core
>>>
>>> I followed the instructions at:  https://jujucharms.com/canonic
>>> al-kubernetes/
>>> but I will retry with the conjure-up kubernetes method as well...
>>>
>>> which resulted in the following "juju status (note errors w/Flannel):
>>>
>>> ModelController  Cloud/Region Version
>>> default  lxd-testlocalhost/localhost  2.0.2
>>>
>>> AppVersion  Status   Scale  Charm
>>> Store   Rev  OS  Notes
>>> easyrsa3.0.1active   1  easyrsa
>>> jujucharms5  ubuntu
>>> etcd   2.2.5active   3  etcd
>>> jujucharms   21  ubuntu
>>> *flannel0.6.1error  *  4
>>> flanneljujucharms7  ubuntu
>>> kubeapi-load-balancer  1.10.0   active   1  kubeapi-load-balancer
>>> jujucharms5  ubuntu  exposed
>>> kubernetes-master  1.5.1active   1  kubernetes-master
>>> jujucharms   10  ubuntu
>>> kubernetes-worker  1.5.1waiting  3  kubernetes-worker
>>> jujucharms   12  ubuntu  exposed
>>>
>>> Unit  Workload  Agent  Machine  Public address
>>> Ports Message
>>> easyrsa/0*activeidle   0
>>> 10.158.189.254Certificate Authority connected.
>>> etcd/0*   activeidle   110.158.189.107
>>> 2379/tcp  Healthy with 3 known peers.
>>> etcd/1activeidle   210.158.189.42
>>> 2379/tcp  Healthy with 3 known peers.
>>> etcd/2activeidle   310.158.189.223
>>> 2379/tcp  Healthy with 3 known peers.
>>> kubeapi-load-balancer/0*  activeidle   410.158.189.99
>>> 443/tcp   Loadbalancer ready.
>>> kubernetes-master/0*  activeidle   510.158.189.105
>>> 6443/tcp  Kubernetes master services ready.
>>>  * flannel/0*  error idle
>>> 10.158.189.105hook failed: "cni-relation-changed" for
>>> flannel:cni*
>>> kubernetes-worker/0   waiting   idle   6
>>> 10.158.189.132Waiting for kubelet to start.
>>>  * flannel/2   error idle
>>> 10.158.189.132hook failed: "etcd-relation-joined" for
>>> flannel:etcd*
>>> kubernetes-worker/1*  waiting   idle   7
>>> 10.158.189.37 Waiting for kubelet to start.
>>>  * flannel/1   error idle
>>> 10.158.189.37 hook failed: "etcd-relation-joined" for
>>> flannel:etcd*
>>> kubernetes-worker/2   waiting   idle   8
>>> 10.158.189.176Waiting for kubelet to start.
>>>  * flannel/3   error idle
>>> 10.158.189.176hook failed: "cni-relation-changed" for
>>> flannel:cni*
>>>
>>>
>>>
>>> --
>>> Juju mailing list
>>> Juju@lists.ubuntu.com
>>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm
>>> an/listinfo/juju
>>>
>>>
>>> --
>>> Juju mailing list
>>> Juju@lists.ubuntu.com
>>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailm
>>> an/listinfo/juju
>>>
>>
> --
> Juju mailing list
> Juju@lists.ubuntu.com
> Modify settings or unsubscribe at: https://lists.ubuntu.com/
> mailman/listinfo/juju
>
>
-- 
Juju mailing list
Juju@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju