On 04/09/2014 05:34 AM, Ramesh Raithatha wrote:
Hello,
I updated to ansible 1.5.3 and now i am not able to run my ruby
scripts using "local_action" and "command" module.
It was running perfectly fine with ansible 1.4.3.
TASK: [Checking elb stat]
On 04/10/2014 09:38 AM, Fred Badel wrote:
Hello,
I am trying to set ansible-pull to provision ec2 instances ... with
not much success so far ...
I tried using the ec2.py dynamic inventory:
ansible-pull -C feature/ci -d /home/ec2-user/ansible -i ansible/ec2.py
tag_Name_qatest23 -U ssh://git
Sorry for necro-ing this.
We solved this by adding the whole pair to the tag key itself, so it
creates groups like tag_class_database_ (mapped from "class_database": "").
Now you can add machines to multiple groups, filter machines that are in
more than one class, etc.
On Friday, J
Is there a way to define an ec2_group with port ranges, or maybe allow
all ports for the security group? There is one example with a protocol of
"all" which the code seems to map to -1, but if I use it, I get the
following:
...
boto.exception.EC2ResponseError: EC2ResponseError: 400 Bad Reques
On 04/10/2014 09:24 AM, seanosh wrote:
Not sure if this is a bug or not so I didn't want to file in GitHub
just yet. A rough timeline:
* Support added between 1.5.2 and 1.5.3 for checking hashed entries in
SSH known_hosts files
* 1.5.3 released
* https://github.com/ansible/ansible/pull/6639
*
Hi Drew,
You're right, the URI module is available. I didn't mention it because
Salman mention he was going to be coordinating the automation for an entire
cluster. The URI module is useful, yes, but interacting with a large API
for complex tasks? That's what I meant when I said that I don't wa
On 04/09/2014 12:23 PM, Stefan Eriksson wrote:
hi I have this exact issue but dont understand the anwer "Creates is
the way to go here!" could anyone explain in little more detail? thanks.
Den tisdagen den 6:e augusti 2013 kl. 03:09:21 UTC+2 skrev Jonathan
Rowlands:
Cool, thanks!
On
On Thursday, April 10, 2014 12:14:24 PM UTC-4, Paul Durivage wrote:
>
> The type of work you're looking to do is best handled by writing a
> Cloudera module. I don't want to imagine how one might wrap a REST API in
> a playbook -- that does not sound fun or reasonable. :)
>
Paul,
It has alread
I don't have any examples for you, but Ansible is definitely the right tool
for this kind of orchestration. I'm sure someone here can provide some
samples.
You'll want to write a playbook that enumerates these steps. Use the
"serial" directive in your play so that only one server is touched at o
You're looking for parameterized roles. Put the tasks in a role, call the
role conditionally when the criteria is met.
http://docs.ansible.com/playbooks_conditionals.html#applying-when-to-roles-and-includes
On Thu, Apr 10, 2014 at 4:38 AM, Stephan S. wrote:
> Hello.
>
> I would like to know i
The type of work you're looking to do is best handled by writing a Cloudera
module. I don't want to imagine how one might wrap a REST API in a
playbook -- that does not sound fun or reasonable. :)
On Wed, Apr 9, 2014 at 2:11 PM, Salman Haq wrote:
> I am exploring automation of a Hadoop cluster
Using a role is the way to go here. Yes, you'll need to physically crate
the role and use 3 files, but conditionally running the role with a "when"
statement is the only way, I think, that you can avoid writing "when: a ==
b" 10 times.
Why do you need 3 files for your role? You said it was simpl
Hello,
I have 2 Ansible groups:
- nginx_servers.
- api_servers.
I am trying to automate following:
1. Select 1 API server.
2. Mark selected API server as down in the all nginx configs. Reload nginx.
3. Restart selected API server.
4. Mark selected API server as up in the all nginx configs.
5. Got
Hi all,
I have a very simple pull request
at https://github.com/CaptTofu/ansible/compare/nova_compute. This fix makes
it so the ansible module, nova_compute, works with HP Cloud and any cloud
provider that has more than one region. It simply does this by providing
the region name which is used
Hello,
I am trying to set ansible-pull to provision ec2 instances ... with not
much success so far ...
I tried using the ec2.py dynamic inventory:
ansible-pull -C feature/ci -d /home/ec2-user/ansible -i ansible/ec2.py
tag_Name_qatest23 -U ssh://git@repo/infra/ansible-play.git local.yml
and I
Not sure if this is a bug or not so I didn't want to file in GitHub just
yet. A rough timeline:
* Support added between 1.5.2 and 1.5.3 for checking hashed entries in SSH
known_hosts files
* 1.5.3 released
* https://github.com/ansible/ansible/pull/6639
* 1.5.4 released
Looks like PR #6639 didn'
Hello.
I would like to know if it is possible to enclose multiple tasks with one
when clause.
I am trying to avoid to use the same when statement for all of my tasks in
the same playbook.
Furthermore, as a requirement is that the above should be implemented while
using a single playbook.
So a
Hi,
Suppose I have a role called Ruby. I will
1. install build tools
2. download ruby source
3. compile
4. install
Because it takes long time to do all these tings and I don't want to it it
the expected Ruby already installed.
How could I do it?
I could do it using the "when" condition on eve
I am exploring automation of a Hadoop cluster.
The vendor (Cloudera) requires that all configuration happen through a REST
API. See [1] for Cloudera Python API client.
My question is: Is there a prior example of an Ansible playbook that wraps
an HTTP api?
Thanks!
Salman
[1] http://cloudera.gi
hi I have this exact issue but dont understand the anwer "Creates is the
way to go here!" could anyone explain in little more detail? thanks.
Den tisdagen den 6:e augusti 2013 kl. 03:09:21 UTC+2 skrev Jonathan
Rowlands:
>
> Cool, thanks!
>
> On Monday, August 5, 2013 9:07:36 PM UTC-4, Michael De
Hello,
i have a playbook which contains 10 tasks. I want to run those tasks only
if a when clause if filled.
My first approach was to put on every task the when. E.g.:
*- name: My command command: sudo install when: a == b*
The above example worked fine but was not ideal. Th
Hello,
I updated to ansible 1.5.3 and now i am not able to run my ruby scripts
using "local_action" and "command" module.
It was running perfectly fine with ansible 1.4.3.
TASK: [Checking elb stat]
*
<127.0.0.1> REMOTE_MODULE command elbinfo
Hi,
I have some issues with for loop in jinja2 template. In
hosts_vars/name_of_host i have defined:
add_php_packages:
- php5-xcache
apps:
- {
...
postgres: {
db_host: 'xxx',
db_user: 'xxx',
db_name: 'some_name',
db_pass: 'xxx',
db_port: 5432
This is difficult because the underlying library used to talk to AWS, boto,
doesn't often expose this either. This is really an AWS question rather
than an Ansible question.
If you run your playbook with -vvv you might get more of the error messages
- or at least it should be obvious where it'
Petr, Adam, you are both right.
I't amazing how you don't see the trees because of the forest sometimes.
Thanks both.
Very much solved.
On Wednesday, 9 April 2014 19:57:35 UTC+1, Adam Morris wrote:
>
>
>
> On Wednesday, April 9, 2014 9:30:38 AM UTC-7, Petr Sukharev wrote:
>>
>>
>>
>>> And in theo
Hi all.
I have a master deployment playbook that more or less includes all of my
servers, which I use with tags to deploy only the apps/configs I want at
the time. Hosts are grouped into hostgroups by roles and each hostgroup has
it's own pre/post tasks for handling load-balancers, etc. It look
as stated in the subject: It is ansible 1.5.4
cheers
jc
Am Mittwoch, 9. April 2014 15:53:17 UTC+2 schrieb Michael DeHaan:
>
> For starters, what version of Ansible is this?
>
> Thanks!
>
>
>
>
> On Tue, Apr 8, 2014 at 4:53 AM, Jens-Christian Fischer
>
> > wrote:
>
>> I am trying to upgrade all
27 matches
Mail list logo
