Re: juju bootstrap to a particular subnet

as using the wrong subnet > mask. All working now. > > Thanks, > -Chris > > On Mon, Jul 2, 2018 at 12:09 AM, John Meinel > wrote: > >> It should still be available for AWS. I think I saw a case where you >> might get that error just when it can't find the subne

Re: juju bootstrap to a particular subnet

It should still be available for AWS. I think I saw a case where you might get that error just when it can't find the subnet you asked for (so 'subnet=' is known, but the 172.32.* couldn't be found). I'm not positive about it, but I do still see the subnet matching code in place. John =:-> On

Re: How do you clean a juju disk or make it larger?

IIRC older Juju used the default EC2 settings, which gave 8GB hard drives, but newer should default to 32GB disks. I'm not sure how that varies across all providers, though. Note that you should always be able to bootstrap with a custom root-disk constraint. eg "juju bootstrap

Re: bridge activation error: yaml: unmarshal errors:, line 7: field bonds not found in struct netplan

btw, including a copy of your /etc/netplan/*.yaml files once you've bootstrapped would probably be useful to make sure we're addressing the issues that you're seeing. John =:-> On Tue, Apr 17, 2018 at 4:18 PM, John Meinel <j...@arbash-meinel.com> wrote: > We should get a b

Re: bridge activation error: yaml: unmarshal errors:, line 7: field bonds not found in struct netplan

We should get a bug following this. But essentially Bionic changed how networking configuration is being done, from /etc/network/interfaces to /etc/netplan/*.yaml files. When we started implementing support for netplan, they hadn't finished support for bonds, so we didn't implement support for

Re: Implement aws style user-data

I think the actual question is how to create a service that responds to metadata requests. In that case, it is a bit hard to say from the outside, because it is very dependent on what software-defined-networking you're using. 169.254.169.254 is a link-local address, which means it can't be routed

Re: Juju's dependencies, juju/utils and a road forward

I agree with breaking up utils. I've certainly read similar comments elsewhere that calling something 'utils' inherently causes issues like this, and it seems to fit with our experience. I also agree that X stable depending on Y unstable doesn't seem right. Though we're also very bad at ever

Re: juju deploys with a vsphere controller using hardware vm version 10

We've done some digging on this. It turns out that we can fairly easily request VSphere to deploy the best version it knows about (ESXi 5.5 supports 10, 6.0 supports 11, 6.5 supports 13). It is fairly easy to change our descriptor to request any of them, and we saw that 6.0 picked 11 and ignored

Re: juju deploys with a vsphere controller using hardware vm version 10

We've done some digging on this. It turns out that we can fairly easily request VSphere to deploy the best version it knows about (ESXi 5.5 supports 10, 6.0 supports 11, 6.5 supports 13). It is fairly easy to change our descriptor to request any of them, and we saw that 6.0 picked 11 and ignored

Re: juju 2.3.3 failing to bootstrap localhost

That sounds more like the 'ubuntu' user inside the container is having trouble? Juju connects to initially launched machines as the ubuntu user, not as your username. John =:-> On Thu, Feb 22, 2018 at 5:32 AM, fengxia wrote: > Verified by login to the container. I could `apt

Re: juju juju-upgrade -m controller woes

hine-2 until it is finished > with it. > > I suspect that my only option is to build a new container/controller > and redeploy my machines on it before removing the old > container/controller. > > On Tue, 2018-01-30 at 14:40 +0400, John Meinel wrote: > > I'm a bit curious

Re: juju juju-upgrade -m controller woes

hine-2 until it is finished > with it. > > I suspect that my only option is to build a new container/controller > and redeploy my machines on it before removing the old > container/controller. > > On Tue, 2018-01-30 at 14:40 +0400, John Meinel wrote: > > I'm a bit curious

Re: juju juju-upgrade -m controller woes

I'm a bit curious what you mean by "manually provisioned machine". Are you saying that you used the VMWare APIs directly to launch an instance, and then "juju bootstrap IP" to start using that machine, and then "juju add-machine ssh:IP" to add the second machine? You could try doing "juju

Re: juju juju-upgrade -m controller woes

I'm a bit curious what you mean by "manually provisioned machine". Are you saying that you used the VMWare APIs directly to launch an instance, and then "juju bootstrap IP" to start using that machine, and then "juju add-machine ssh:IP" to add the second machine? You could try doing "juju

Re: Getting the exposed ports

gmail.com> wrote: > Yes, so with juju-run you could run 'opened-ports' in the hook context of > each unit on that machine, and this get all the opened+ports on the machine. > > On 2 Dec 2017 04:26, "John Meinel" <j...@arbash-meinel.com> wrote: > >> I'm pretty

Re: Getting the exposed ports

I'm pretty sure that opened-ports only reports the ports that Juju had opened for the charm that is making the request. I don't think we list all ports opened on the machine for all other applications. So you might need to have a relation that can report it's opened ports to the subordinate John

Re: Deploying a charm with t's & c's

Do you know what terms you need to accept? You should be able to 'juju accept termname'. The other thing to check is if your local clock time is fairly accurate. I believe we filed a bug recently that the term server gives out temporary auth tokens that are only valid for 1 minute. (and we end up

Re: ownership and sharing of models via team namespace

So models shared with you are currently possible, but I believe everything is still owned as an individual user, rather than at a team scope. You should be able to add other users as 'admin' on an existing model, but we're still working through a backlog of places that treat "owner" differently

Re: ownership and sharing of models via team namespace

So models shared with you are currently possible, but I believe everything is still owned as an individual user, rather than at a team scope. You should be able to add other users as 'admin' on an existing model, but we're still working through a backlog of places that treat "owner" differently

Re: Endpoints trials + MAAS charms

It does seem like Juju operating the LXD provider, but spanning multiple machines would be an answer. I do believe that LXD itself is introducing clustering into their API in the 18.04 cycle. Which would probably need some updates on the Juju side to handle their targeted provisioning (create a

Re: Is there a universal interface I can use?

heers > Tilman > > On 23.11.2017 11:39, Tilman Baumann wrote: > > Cool. Thanks > > > > The two fields I was interested in, username and password are mising > > though. :D > > But I'm thinking right now if I even want to go that route... > > > >

Re: How to open port with juju when no application is instlaled?

I had thought of that, but it only works if you then 'juju expose postgresql' which would also expose the other ports of postgres John =:-> On Nov 23, 2017 23:03, "Tim Penhey" wrote: > I think you might be able to use: > > juju run postgresql/0 'open-port 80' > >

Re: How to open port with juju when no application is instlaled?

I believe there is an nginx charm, which you could have installed with "juju deploy nginx --to X" (where X is the machine id of the postgres charm), and then used "juju expose nginx". Is there a reason you prefer to install it manually? One other option would be to co-locate the "ubuntu" charm

Re: Is there a universal interface I can use?

I did start working on a Cassandra interface for something I was working on. I don't know that it is complete but https://github.com/jameinel/interface-cassandra Was my attempt at it. John =:-> On Nov 23, 2017 02:26, "Haw Loeung" wrote: > Hi Tilman, > > On Wed, Nov

Re: keystone charm

It seems like you would tend to use: import os os.makedirs(os.path.dirname(config_file)) Which doesn't treat a directory that already exists as a failure. However, I'm guessing that there is some other expectation that the directory should be created at some other time. John =:-> On Tue, Nov

Re: Juju 2.3 beta2 is here!

... > > Perhaps just: > > > > juju deploy --map-machines A=B,C=D > > > > ... or some variant of that? > > > > Let's use the betas to refine and condense and clarify. > > +1 to that. I'm wondering if use-existing-machines is ever appropriate > on its own, as the machine numbers in a model are

Re: Effect of lease tweaks

(sent too soon) Summary: before: 1043 local.oplog.rs 868 juju.txns 470 juju.leases after: 802 local.oplog.rs 625 juju.txns 267 juju.leases So there seems to be a fairly noticeable decrease in load on the system around leases (~70%). Again, not super scientific because I didn't measure over

Re: backward compatibility

So I think its fine for giving feedback from client against a controller (new client, old controller). Though how often we want to warn, have a way to disable the warning (for how long, etc)? The other side seems more difficult, as far as 2.0 client talking to a 2.4 controller. We could start

Re: backward compatibility

So I think its fine for giving feedback from client against a controller (new client, old controller). Though how often we want to warn, have a way to disable the warning (for how long, etc)? The other side seems more difficult, as far as 2.0 client talking to a 2.4 controller. We could start

Re: default network space

So at the moment, I don't think Juju supports what you're looking for, which is cross model relations without public addresses. We've certainly discussed supporting all private for cross model. The main issue is that we often drive parts of the firewalls (security groups) but without understanding

Re: default network space

So at the moment, I don't think Juju supports what you're looking for, which is cross model relations without public addresses. We've certainly discussed supporting all private for cross model. The main issue is that we often drive parts of the firewalls (security groups) but without understanding

Re: Relation scope clarification

e existing "store my data in postgres, 'pgsql' relation" ? John =:-> On Tue, Oct 17, 2017 at 11:16 PM, John Meinel <j...@arbash-meinel.com> wrote: > Why is the subordinate container scoped? The specific scope of container > is that you only see the single instance that yo

Re: Relation scope clarification

Why is the subordinate container scoped? The specific scope of container is that you only see the single instance that you share a machine with. Typically subordinates will use something like the juju-info relation because all they really care about is to be colocated on the same machine. I can't

Re: JAAS confusion

OR cannot destroy model: context deadline exceeded > > Is there an open bug that I can paste error messages and logs to? > > ~ PeteVG > > > > On Mon, Oct 9, 2017 at 4:18 PM John Meinel <j...@arbash-meinel.com> wrote: > >> The 6 accessible models is gone in 2.3 (II

Re: JAAS confusion

The 6 accessible models is gone in 2.3 (IIRC), because it was actually just reflecting some locally cached information about model numbers, but wasn't actually being kept up to date properly. I think the inability to remove a model that is half-dead might be fixed already in 2.3 but has to do

Re: "Attempting to connect to" takes a long time

We could use better progress information but once connected we also update the packages installed on the machine and download and install a few packages. Having the latest Ubuntu image often means we have fewer packages to install. John =:-> On Oct 6, 2017 7:00 AM, "fengxia"

Re: Jujucharms

It appears to be a more general Canonical outage, as irc.canonical.com is also affected. John =:-> On Sat, Sep 9, 2017 at 6:03 PM, Tom Barber wrote: > Okay, own up, who killed Jujucharms.com whilst i'm trying to catch up on > work? :) > > > -- > Juju mailing list >

Re: How to use LXC local image for new machine

If you're seeing "ubuntu-trusty" then you're using a version of Juju that doesn't support the pressed. I believe reusing the cached image first was added in 2.1, when we switched the naming scheme to be more specific. John =:-> On Sep 1, 2017 18:10, "Alex Kavanagh"

Re: Juju-gui only show localhost as the deployment option

The main reasons we don't directly offer cross cloud support is mostly about user experience. There are potentially tricky things around getting routing to work correctly between machines. Also there are several things that are cached on controllers, which only helps if the controller is "local"

Re: Juju-gui only show localhost as the deployment option

Supporting multiple 'projects' in Openstack should already be supported by passing different credentials during "juju add-model". John =:-> On Thu, Aug 31, 2017 at 1:55 AM, Giuseppe Attardi wrote: > I have a slightly different requirement. > Currently in our OpenStack

Re: developer commands

I don't think you intended to send the email as PGP encrypted. John =:-> 2017-08-28 12:06 GMT+04:00 Tim Penhey : > -BEGIN PGP MESSAGE- > Version: GnuPG v2 > > hQIMA233D38ktbXXAQ/+P+Wl6YvVE3PVo1tsN/ynVPSeR5Xu3SfKLoRmXvxM0om/ >

Re: LXD bootstrap inside KVM failed

If you set up a space in MAAS and you declare that the VM is in that space, then when you deploy something into a container, MAAS and Juju will coordinate to get a Container set up onto a bridge connected to the interface in the VM that is connected to the same space, with an IP address from the

Re: Consuming MongoDB from a Snap

I just wanted to note that some of the reason for 128GB was because 2.0 and 2.1 did leak memory over time. And if you have a leak you will always eventually run out. In 2.2 we've fixed all the ones we've found so far and we're actively doing some performance measuring to give better guidelines.

Re: How to set state in @hook code block?

I'm pretty sure hooks execute in the parent directory. So if you have charm/ hooks/ db-relation-changed myfiles/ test.yaml you need to open("hooks/myfiles/test.yaml") John =:-> On Jul 20, 2017 04:58, "fengxia" wrote: > Hi Juju, > > I think I read at some point

Re: call for testing: relations across Juju models

> > ... > > Current known limitations: > Only works in the same model > You need to bootstrap with the feature flag to test it out > Does not currently work with relations to subordinates. Work is in progress > I'm guessing you mean "only works in the same controller". If cross model relations

Re: Organising apiserver facades

I'd really like to see us split apart the facades-by-purpose. So we'd collect the facades for Agents separately from facades for Users (and possibly also facades for Controller). I'm not sure if moving things just into 'facades' just moves the problem around and leaves us with just a *different*

Re: Using Deb Proxy to cache Debs to minimize time starting-up machines and its installation scripts

For what its worth, I use "apt-cacher-ng" to do this very thing and have ~/.local/share/juju/clouds.yaml: clouds: lxd: type: lxd config: apt-http-proxy: http://192.168.0.106:3142 apt-https-proxy: http://192.168.0.106:3142 enable-os-upgrade: false As long as you're

Re: Running KVM in addition to LXC on local LXD CLOUD

Generally you have to configure the profile of the containers (either by configuring the default profiles or by altering the configuration of a single container and then restarting that container). If there are particular modules that you know you will need then you can use "linux.kernel_modules"

Re: juju enable-ha with lxd servers

Are these 3 LXD containers on one machine, or 3 different host machines where you want to run LXD containers? At present we don't support scheduling across LXD hosts, so the easier way would probably be to treat 3 host machines as separate 'manually provisioned' machines, and then deploy to

Re: docker -> lxdbr0

2017, at 10:31 PM, John Meinel <j...@arbash-meinel.com> wrote: > > This definitely sounds interesting. > > You included the layer python code, but not the "daemon.json.j2" file. > Isn't that part of getting the networking config in place? > > John > =:->

Re: docker -> lxdbr0

2017, at 10:31 PM, John Meinel <j...@arbash-meinel.com> wrote: > > This definitely sounds interesting. > > You included the layer python code, but not the "daemon.json.j2" file. > Isn't that part of getting the networking config in place? > > John > =:->

Re: docker -> lxdbr0

This definitely sounds interesting. You included the layer python code, but not the "daemon.json.j2" file. Isn't that part of getting the networking config in place? John =:-> On Wed, Jun 21, 2017 at 6:16 AM, James Beedy wrote: > On integrating docker and lxd deployed

Re: multi model/cloud deploy (how to deploy kubernates-workers in multitenant openstack)

If you have started the machine yourself, you should be able to "juju add-machine ssh:IP_ADDRESS" and then use that as a "juju deploy --to X" target. However, you will still need to tear down the machine when you're done. We don't yet support multi-provider models. Likely we won't, but we will

Re: juju deploy with a series

"juju show-machine 10" is likely to tell you why we are failing to provision the machine. My guess is that we acctually need the alias to be "juju/centos7/amd64" for Juju to recognize that it is the container image we want to be starting. John =:-> On Thu, Jun 15, 2017 at 8:37 PM, Daniel

Re: Maximum AllWatcher frame size?

In 2.1 we did not chunk, in 2.2 we switch to gorilla/websocket which does support chunking into frames. I don't think we do any internal "well that would be too much information so we wont send it all". John =:-> On Wed, Jun 14, 2017 at 7:11 PM, Cory Johns wrote: >

Re: killing old dead controllers

If the machines are just gone (you manually destroyed them via 'lxc stop/delete'). You can just do: juju unregister lxd-test It will remove it from the local registry without trying to tear anything down. John =:-> On Tue, Jun 13, 2017 at 8:17 AM, Daniel Bidwell wrote:

Re: killing old dead controllers

If the machines are just gone (you manually destroyed them via 'lxc stop/delete'). You can just do: juju unregister lxd-test It will remove it from the local registry without trying to tear anything down. John =:-> On Tue, Jun 13, 2017 at 8:17 AM, Daniel Bidwell wrote:

Re: Is there a way to login to lxd containers of existing openstack?

'ssh ubuntu@IPADDRESS' generally works, as the 'ubuntu' user is the one that is configured with .ssh/authorized_keys that match the keys described by the model. I don't know what ssh keys you added to the model/what it was bootstrapped with, but as long as you still have access to one of those ssh

Re: OS X VMS on JAAS

... > >> Which is why using something like the "lxd provider" would be a more >> natural use case, but according to James the sticking point is having to >> set up a controller in the first place. So "--to lxd:0" is easier for them >> to think about than setting up a provider and letting it

Re: OS X VMS on JAAS

... > >> Which is why using something like the "lxd provider" would be a more >> natural use case, but according to James the sticking point is having to >> set up a controller in the first place. So "--to lxd:0" is easier for them >> to think about than setting up a provider and letting it

Re: Exiting an unconditional juju debug-hooks session

Doesn't the equivalent of ^A ^D (from screen) also work to just disconnect all sessions? (http://www.dayid.org/comp/tm.html says it would be ^B d). Or switching to session 0 and exiting that one first? I thought we had a quick way to disconnect, but its possible you have to exit 2x and that fast

Re: Exiting an unconditional juju debug-hooks session

Doesn't the equivalent of ^A ^D (from screen) also work to just disconnect all sessions? (http://www.dayid.org/comp/tm.html says it would be ^B d). Or switching to session 0 and exiting that one first? I thought we had a quick way to disconnect, but its possible you have to exit 2x and that fast

Re: OS X VMS on JAAS

gt;> they don't have the controller on them as well. @jameinel totally, possibly >> I'll add the bridge bits in place of the lxd-proxy in that write up, or >> possibly in another. >> >> ~James >> >> On Jun 2, 2017, at 12:56 AM, John Meinel <j...@arbash-me

Re: OS X VMS on JAAS

gt;> they don't have the controller on them as well. @jameinel totally, possibly >> I'll add the bridge bits in place of the lxd-proxy in that write up, or >> possibly in another. >> >> ~James >> >> On Jun 2, 2017, at 12:56 AM, John Meinel <j...@arbash-me

Re: OS X VMS on JAAS

Interesting. I wouldn't have thought to use a manually added machine to use JAAS to deploy applications to your local virtualbox. Is there a reason this is easier than just "juju bootstrap lxd" from inside the VM? I suppose our default lxd provider puts the new containers on a NAT bridge, though

Re: OS X VMS on JAAS

Interesting. I wouldn't have thought to use a manually added machine to use JAAS to deploy applications to your local virtualbox. Is there a reason this is easier than just "juju bootstrap lxd" from inside the VM? I suppose our default lxd provider puts the new containers on a NAT bridge, though

Re: juju test doesn't work

I'm pretty sure 'charm' tools have moved over to using 'snap install charm' as the preferred method for getting the charm tools. I'm not sure that there is a way to deprecate/remove the versions that are in the archives. For something like Zesty, it probably would have been possible to just remove

Re: database-relation-join not invoked

It depends if your hook goes into 'error' state or 'blocked'. Error should generally be avoided because it is a signal to Juju that you can no longer make forward progress (generally meant to mean there is a logic bug/typo/etc in your charm). With Error Juju may retry the hook that failed but it

Re: charm build -s centos7 does nothing special?

ipts. If so, my immediate thought is that charm build does nothing > special for series then, since I'm coding the "if series=='ubuntu'... > else:", the build script cannot touch my code, right? > > On 05/20/2017 03:17 AM, John Meinel wrote: > > I would guess the charm ca

Re: JUJU_UNIT_NAME no longer set in env

juju run --application runs as every unit of the application, thus we have individual units (if you have 2 units of an application on a machine it will run twice on that machine). 'juju run --unit" obviously runs as a unit. 'juju machine' can't, because we don't have any unit associated with it.

Re: JUJU_UNIT_NAME no longer set in env

So you're running the 'install' hook directly, are you currently in a 'juju debug-hooks' session, or are you just changing into that directory? Juju sets it during the run of a charm hook, but it is not set globally on the machine (we can't set UNIT globally anyway, because you can colocate many

Re: Scale testing analysis

> > ... > > We have most of the responsive nature of Juju is driven off the watchers. > These watchers watch the mongo oplog for document changes. What happened > was that there were so many mongo operations, the capped collection of the > oplog was completely replaced between our polled watcher

Re: django charm broken or am I just doing it wrong?

I believe Tim Penhey makes active use of the python-django charm, but it's possible he uses it in a different fashion. John =:-> On May 21, 2017 14:25, "Erik Lönroth" wrote: > Hello! > > I'm trying out the django charm to deploy a django website I was going to > try to

Re: charm build -s centos7 does nothing special?

I would guess the charm can define series specific build instructions, but that layer-basic doesn't, itself, have different instructions for centos7. John =:-> On Sat, May 20, 2017 at 10:07 AM, fengxia wrote: > Hi Juju, > > I made a hello world charm based on charm tutorial,

Re: PROPOSAL: stop recording 'executing update-status hook'

urprise in not-great situations where the deployment is wedged. > > We would not know that from the status right? Only from the debug-log. > > On May 19, 2017 3:46 AM, "John Meinel" <j...@arbash-meinel.com> wrote: > >> All agents start up in DEBUG until they can ta

Re: PROPOSAL: stop recording 'executing update-status hook'

urprise in not-great situations where the deployment is wedged. > > We would not know that from the status right? Only from the debug-log. > > On May 19, 2017 3:46 AM, "John Meinel" <j...@arbash-meinel.com> wrote: > >> All agents start up in DEBUG until they can ta

Re: PROPOSAL: stop recording 'executing update-status hook'

All agents start up in DEBUG until they can talk to the controller and read what the current logging config is set to. Otherwise you wouldn't be able to debug startup issues. That said, I think there was a request to cache the last-known value in agent.conf which would let restarts be less noisy.

Re: PROPOSAL: stop recording 'executing update-status hook'

All agents start up in DEBUG until they can talk to the controller and read what the current logging config is set to. Otherwise you wouldn't be able to debug startup issues. That said, I think there was a request to cache the last-known value in agent.conf which would let restarts be less noisy.

Re: How to build juju for centOS

Also, while agents can be built for CentOS we don't support Controllers on CentOS at this point. So bootstrap I believe only supports Ubuntu. John =:-> On May 10, 2017 11:44, "Andrew Wilkins" wrote: > On Wed, May 10, 2017 at 3:08 PM fengxia

Re: Juju deploy same application twice, failed "application already exist"

You can already do that without the extra flag: juju deploy app [name] You can name any app any name, we just default to the app name if it is not supplied. John =:-> On Apr 21, 2017 18:12, "Patrizio Bassi" wrote: > Dear All > > this conversation led me to an

Re: How to add openstack cloud to juju 2.1.2-xenial

12, 2017 at 9:56 AM, John Meinel <j...@arbash-meinel.com> wrote: > It sounds like the docs are out of date. The key should be > "agent-metadata-url" 'tools' was a much older name. > > John > =:-> > > > On Wed, Apr 12, 2017 at 7:32 AM, Daniel Bidwell <drb

Re: How to add openstack cloud to juju 2.1.2-xenial

It sounds like the docs are out of date. The key should be "agent-metadata-url" 'tools' was a much older name. John =:-> On Wed, Apr 12, 2017 at 7:32 AM, Daniel Bidwell wrote: > This gets me down to: > > juju bootstrap acauits acauits-controller --config

Fwd: Quick analysis of 'stale' transactions

We're trying to look at reasons why we end up with some stale transaction ids in the txn-queue. I did find a machine with 18 entries in its txn queue, all of which are '6: applied'. I ended up discovering that SetLinkLayerDevices and SetDeviceAddresses are both causes of a fair number of

Re: Network Space Binding Changes After Deployment or Upgrades

On Mon, Mar 27, 2017 at 3:31 PM, Dmitrii Shcherbakov < dmitrii.shcherba...@canonical.com> wrote: > Hi everybody, > > As far as I can see, there is no way to change a network space binding or > add a new one after a charm has been deployed. > The one caveat here is that in 2.1.2+ if you used a

Re: A new release of Juju, 2.2-beta1 and conjure-up are here!

Did you "juju expose mediawiki" ? That should have Juju update the security group on the mediawiki machines to expose port 80. John =:-> On Sun, Mar 26, 2017 at 10:58 PM, Giuseppe Attardi wrote: > I solved the problem by using Chrome instead of Safari. > > I was

Re: Can't juju deploy p2.xlarge in aws/us-east-1

Preferring a VPC if there is a single one that exists, even if it isn't flagged as "default" for a region would probably be reasonable, and probably not a lot of effort. If there are multiple, I wonder if we could refuse to bootstrap/add-model unless one is specified. John =:-> On Tue, Mar 21,

Re: Can't juju deploy p2.xlarge in aws/us-east-1

Preferring a VPC if there is a single one that exists, even if it isn't flagged as "default" for a region would probably be reasonable, and probably not a lot of effort. If there are multiple, I wonder if we could refuse to bootstrap/add-model unless one is specified. John =:-> On Tue, Mar 21,

Re: Constraints for Additional Units

Correct. Putting the constraints on the machines is equivalent to doing: juju add-machine --constraints mem=4gb juju deploy mysql --to $MACHINE Rather than doing: juju deploy mysql --constraints mem=4gb Constraints on an Application will affect all new units of an application, but

Re: maas how to add-cloud to a zone

Zones are generally meant to be used to provide fault domains such that you should be spreading your applications across zones. I can see how zones could be used differently but it does feel like it would be going against the grain. John =:-> On Mar 15, 2017 13:01, "Menno Smits"

Re: juju 2.0.2 picking a random ip address from maas

t; > Patrizio > > 2017-03-09 16:24 GMT+01:00 John Meinel <j...@arbash-meinel.com>: > >> I do have a fix up for review: >> https://github.com/juju/juju/pull/7081 >> And I have tested it live against a local maas, though my maas doesn't >> have multiple inter

Re: juju 2.0.2 picking a random ip address from maas

>>> directly: so in the case of mysql that would be: >>> juju deploy mysql --bind "db=db-space monitors=db-space ha=db-space ..." >>> >>> John >>> =:-> >>> >>> >>> On Thu, Mar 9, 2017 at 7:25 AM, John Meinel <

Re: juju 2.0.2 picking a random ip address from maas

In the meantime, you can work around it by specifying the bindings directly: so in the case of mysql that would be: juju deploy mysql --bind "db=db-space monitors=db-space ha=db-space ..." John =:-> On Thu, Mar 9, 2017 at 7:25 AM, John Meinel <j...@arbash-meinel.com> wro

Re: juju 2.0.2 picking a random ip address from maas

which space to use because ambiguous bindings will >>> result in a provisioning failure." >>> >>> This is exactly my case: a machine with 2 eth ports, two different >>> subnets in two different spaces. >>> >>> the doc says i may do (c/p): $ juju

Re: juju 2.0.2 picking a random ip address from maas

db-space and so bind a maas space for all the application. Unfortunately it seems not working (i get the "empty names" error). Patrizio 2017-03-08 20:40 GMT+01:00 John Meinel <j...@arbash-meinel.com>: > So without bindings, I would expect the behavior, the question is why you >

Re: juju 2.0.2 picking a random ip address from maas

> not working (i get the "empty names" error). > > Patrizio > > 2017-03-08 20:40 GMT+01:00 John Meinel <j...@arbash-meinel.com>: > >> So without bindings, I would expect the behavior, the question is why you >> would be seeing: >> "cannot ru

MySQL charm bugs

I just saw a couple of bugs targeted at Juju that were about the MySQL charm: https://bugs.launchpad.net/charms/+source/mysql/+bug/1670431 and https://bugs.launchpad.net/charms/+source/mysql/+bug/1670430 I retargeted them at the mysql charm. Is that the correct place to put them? Thanks, John

Re: juju 2.0.2 picking a random ip address from maas

. > > Patrizio > > > 2017-03-08 16:27 GMT+01:00 John Meinel <j...@arbash-meinel.com>: > >> Is it possible for you to test with Juju 2.1? I haven't seen that >> particular bug with binding, but I have done a lot more testing with 2.1. I >> didn't think w

Re: juju 2.0.2 picking a random ip address from maas

s: xenial > hardware: arch=amd64 cores=4 mem=8192M tags=virtual > availability-zone=primary > > it looks a bug, or better, the bug: dns-name is 192.x.x.x but it's not > true, 10.0.8.12 is the real hostname provided by external dns. > > Patrizio > > 2017-03-08 14:57 GMT+

Re: juju 2.0.2 picking a random ip address from maas

You should be using "juju deploy application --bind space" to declare which set of addresses you want the applications to use. Does that not work? John =:-> On Wed, Mar 8, 2017 at 4:44 AM, Patrizio Bassi wrote: > Hi All, > > I'm quite new the juju and it's charms.

Re: Issue with Amulet testing

Seems more like just a python thing, vs anything to do with Amulet. https://github.com/kennethreitz/requests/issues/2214 Says you can do some magic around disabled warnings A better goal might be to have the charm optionally take the config of a cert to use and configure your tests to include

  1   2   3   4   5   >