[ansible-project] Using cobbler.py to discover hosts on multiple cobbler servers

I am curious whether anyone has used cobbler.py to discover inventory on multiple cobbler servers, without needing to change the host variable to a new cobbler server every time. -- You received this message because you are subscribed to the Google Groups "Ansible Project" group. To

[ansible-project] How to write cobbler.ini for multiple cobbler servers

I want to set up multiple cobbler servers' hosts in my cobbler.ini to use their inventory...how do I do that? -- You received this message because you are subscribed to the Google Groups "Ansible Project" group. To unsubscribe from this group and stop receiving emails from it, send an email

Re: [ansible-project] Re: Using with_subelements, cannot reference dict item's key (v 2.2.0)

sucks! > > Thanks! > > > On Wednesday, June 29, 2016 at 1:43:39 PM UTC-5, Joanna Delaporte wrote: >> >> Here's my version: ansible 2.2.0 (devel 87928ff56b) last updated >> 2016/06/22 13:32:23 (GMT -500) >> >> I'm not sure if this is a bug, or misplaced syn

Re: [ansible-project] ERROR! 'ignore_errrors' is not a valid attribute for a Task

Lol, sometimes it just takes another set of eyes. Thanks! On Friday, July 29, 2016 at 4:29:10 PM UTC-5, Kai Stian Olstad wrote: > > On 29. juli 2016 22:00, Joanna Delaporte wrote: > > Hello, > > > > I have a task that looks like this: > > ## We need to ignore

[ansible-project] Re: Issue with Ansible

"applications": [ *'*local$43d12cfd-f64a-4fa9-9367-3a0109c1721d" ]}' has non-matching quotes in your action: shell task. That one right before local needs to be a double quote, I think. On Friday, July 29, 2016 at 2:19:56 PM UTC-5, GBANE FETIGUE wrote: > > Hi guys, > I have some issue with

[ansible-project] Re: No epoch or seconds in windows facts ? How can i generate unique strings ?

Seconds are at the end of the iso8601 string. I use the iso8601 for one playbook where I need a unique string on volume snapshots. On Thursday, July 28, 2016 at 8:13:13 AM UTC-5, ishan jain wrote: > > Here is a small part of windows facts: > > "ansible_facts": { >

[ansible-project] Re: Blockinfile indentation

on and 2 > for the block itself. > > From this excellent answer: http://stackoverflow.com/a/21699210 > > > On Friday, 27 May 2016 08:33:15 UTC+12, Joanna Delaporte wrote: >> >> I am using blockinfile to write lines to a vars file that includes a few >> different

[ansible-project] ERROR! 'ignore_errrors' is not a valid attribute for a Task

Hello, I have a task that looks like this: ## We need to ignore errors because links won't all be there. - name: migrate group-owned staff files to new GIDs shell: "find /home/staff -gid {{item.0.gid}} -exec chgrp {{item.1.stdout}} \\{\\} +" ignore_errrors: yes with_together: -

[ansible-project] Re: Safe method to resize existing filesystem

dn't boot because they both had a filesystem where fsck failed because > the filesystem was larger than the underlying device. I'm quite sure the > problem was caused by some buggy Puppet manifest code which had left a data > corruption problem waiting to happen. > > On Thursday, 28 July 2016

[ansible-project] Safe method to resize existing filesystem

I was looking at the filesystem module today, hoping to use it to resize a filesystem that is living in a logical volume. However, I am not sure how to use it safely so it doesn't create a filesystem. Has anyone used the filesystem module to resize an LV? If so, how did you do it? Thanks!

[ansible-project] Re: Could a value refer another key’s value as variable

Does field2's value display correctly if you use the full path to the other variable? --- - hosts: localhost vars: record: field1: 1 field2: "{{*record.*field1}}" tasks: - name: This one does not work debug: msg="{{record.field2}}" On Wednesday, July 20, 2016 at

[ansible-project] Re: Using ec2_vol with dynamic inventory

Hi Max, Are you running this playbook with the ec2 instance as the remote target? You are missing the reference to the registered fact, I think: - name: gather ec2 facts action: ec2_facts register: ec2_facts - debug: var=ec2_facts - name: Add volume for cassandra nodes

[ansible-project] Re: lineinfile tasks adds line when exists

Hi Robert, To replace an existing line, you will want to use backrefs. Otherwise, I believe the default behavior is to insert the line after EOF or the last match of a specified regexp. Insertafter and backrefs are exclusive of each other, since the line will be added after, or replace an

[ansible-project] lineinfile with backrefs and create: yes is failing with path does not exist

Is this expected behaviour? I am trying to change a line in a file if it is already present, or create the file if it isn't: - name: set ypserver to localhost lineinfile: dest: /etc/yp.conf create: yes state: present backup: yes backrefs: yes regexp: "^ypserver .*"

[ansible-project] Cannot specify volume_size together with id or snapshot (ec2_vol) - How to increase size while creating volume from snapshot?

I have the use case where I want to create a snapshot, make a new volume from that snapshot, but larger, and then attach the new volume to an instance. So far, I have the snapshot creation working, but the ec2_vol module is now telling me that I cannot specify a volume size with the snapshot

Re: [ansible-project] ec2_snapshot module - how to get snapshot id of newly created snapshot?

_snapshot', not the keys underneath it. > > On Thu, Jul 14, 2016 at 12:53 PM Joanna Delaporte <joannad...@gmail.com > > wrote: > >> I am using the ec2_snapshot module to make a snapshot of a volume. In the >> next task, I want to use that new snapshot to create a new volu

[ansible-project] ec2_snapshot module - how to get snapshot id of newly created snapshot?

I am using the ec2_snapshot module to make a snapshot of a volume. In the next task, I want to use that new snapshot to create a new volume for a different server. How do I get the new snapshot volume id from the ec2_snapshot module? It errored (unsupported parameter for module: register) when

[ansible-project] Re: EC2 slow cloud-init, Ansible SSH connection fails due to race condition (wait_for is not good enough)

I just discovered this issue as well, with various random ssh connection or generic authentication/permission failure messages, some occurring after a play successfully passed a task or two. It occurred very consistently with many CentOS 7 t2.nano hosts. A 2-minute pause after waiting for ssh

[ansible-project] acl module - Does it work for nfs4 acls?

I am new to NFSv4, but I have it mostly working with my IPA (kerberized) domain. I just noticed that my nfs clients are not picking up the ACLs from the NFS server, that I set with the ACL Ansible module. So, I suspect the module is setting POSIX acls, not nfs4 acls. Is that correct? -- You

[ansible-project] Re: shell module find -exec chown - how to parse/escape {} \; ?

I figured it out. I also switched to batch mode instead of single mode: shell: "find /home/staff -gid {{item.0.stdout}} -exec chgrp {{item.1.stdout}} \\{\\} +" On Wednesday, July 6, 2016 at 11:49:19 AM UTC-5, Joanna Delaporte wrote: > > I want to find and chown about 4000 us

[ansible-project] shell module find -exec chown - how to parse/escape {} \; ?

I want to find and chown about 4000 users' files while migrating to a new authentication protocol and domain. I have a task that looks like this: - name: find and fix UID ownerships in all staff homedirs shell: "find {{'~'+item.name | expanduser }} -uid {{item.uid}} -exec chown

[ansible-project] Do lists and dicts keep the same order for every task?

I am curious whether lists and dicts maintain their order when used in different tasks. I want to make sure I don't accidentally scramble my user/group data as I migrate. -- You received this message because you are subscribed to the Google Groups "Ansible Project" group. To unsubscribe from

[ansible-project] Re: Synchronize command not working when upgraded to Ansible 2.1.0

Hi Mick, The docs say that using push mode, the delegate is the source of the sync. So, your task is trying to ssh from HostC (the delegate) to HostB. Can these two hosts talk to each other? Maybe you meant to delegate_to: HostA? Joanna On Tuesday, July 5, 2016 at 4:45:26 PM UTC-5, Michael

[ansible-project] Re: Using with_subelements, cannot reference item.0.key - why?

, Joanna Delaporte wrote: > > Here's my version: ansible 2.2.0 (devel 87928ff56b) last updated > 2016/06/22 13:32:23 (GMT -500) > > I'm not sure if this is a bug, or misplaced syntax. When I attempt to use > a dictionary, referencing the key of the dictionary and a subelement,

[ansible-project] Re: Using with_subelements, cannot reference dict item's key (v 2.2.0)

Apparently with_subelements needs a list, not a dict? On Wednesday, June 29, 2016 at 1:43:39 PM UTC-5, Joanna Delaporte wrote: > > Here's my version: ansible 2.2.0 (devel 87928ff56b) last updated > 2016/06/22 13:32:23 (GMT -500) > > I'm not sure if this is a bug, or misplaced

[ansible-project] Re: question on lineinfile module

Hi Werner, You would want to use the backrefs feature of the lineinfile module. Here is an example that I use for deleting root password hashes from the /etc/shadow file. I use regex numbered groups (with parens and \1, etc) to capture and replay existing parts of lines as needed: - name:

[ansible-project] Re: List all variables/facts that are used in a playbook?

Kevin, That sounds really useful. I am not aware of any simple way to do that, but I am following this conversation because I might learn something useful. Thanks for asking! Joanna On Wednesday, June 29, 2016 at 1:48:25 PM UTC-5, Kevin Schumacher wrote: > > Hi all, > > I am looking for a way

[ansible-project] Using with_subelements, cannot reference dict item's key (v 2.2.0)

Here's my version: ansible 2.2.0 (devel 87928ff56b) last updated 2016/06/22 13:32:23 (GMT -500) I'm not sure if this is a bug, or misplaced syntax. When I attempt to use a dictionary, referencing the key of the dictionary and a subelement, I get the error FAILED! => {"failed": true, "msg":

Re: [ansible-project] with_dict expects a dict...can a dict be too large

new users appearing on my IPA server with appropriate data. Thanks Dick and Matt! If you're ever in Champaign, IL USA, look me up. I'll buy you a coffee. Joanna On Wednesday, June 22, 2016 at 8:24:29 PM UTC-5, Joanna Delaporte wrote: > > Yup, a test with a short dict of only three records

Re: [ansible-project] with_dict expects a dict...can a dict be too large

ectly (no whitespace gotchas?). > > Also, are you sure the dict is in scope at this point? > > On 22 June 2016 at 20:55, Joanna Delaporte <joannad...@gmail.com > > wrote: > > I messed up and deleted my first post. Sorry! > > > > I think I need an

Re: [ansible-project] with_dict expects a dict...can a dict be too large

t; little sample set, just to verify the YAML > is formatted correctly (no whitespace gotchas?). > > Also, are you sure the dict is in scope at this point? > > On 22 June 2016 at 20:55, Joanna Delaporte <joannad...@gmail.com > > wrote: > > I messed up and deleted my fi

Re: [ansible-project] with_dict expects a dict...can a dict be too large

formatted correctly (no whitespace gotchas?). > > Also, are you sure the dict is in scope at this point? > > On 22 June 2016 at 20:55, Joanna Delaporte <joannad...@gmail.com > > wrote: > > I messed up and deleted my first post. Sorry! > > > > I think I

[ansible-project] with_dict expects a dict...can a dict be too large?

I think I need another pair of eyes here. I am attempting to run a dict of users through an ipa user-add command, but it fails with the error "with_dict expects a dict." I believe I have formatted my dict correctly (it's about 40,000 lines long, generated from a python3 script): --- students:

[ansible-project] Re: Syntax for skip_missing flag for with_subelements?

4 AM UTC-5, Joanna Delaporte wrote: > > What is the syntax for the third element that can be used with_subelements > to use to skip missing subelements? > > "Optionally, you can add a third element to the subelements list, that > holds a dictionary of flags. Currently you can add t

Re: [ansible-project] Ansible modify /etc/sysconfig/network file

name with FQDN" > lineinfile: > dest=/etc/sysconfig/network > line="HOSTNAME={{ ansible_fqdn }}" backrefs=yes > regexp="^HOSTNAME=.*" > state=present On Friday, June 3, 2016 at 11:20:11 AM UTC-5, Joanna Delaporte wrot

Re: [ansible-project] Ansible modify /etc/sysconfig/network file

In the setup module (get_facts), Ansible should be detecting an FQDN that you can access with '{{ ansible_fqdn }}' I would use the lineinfile module. A rough draft (untested) without > correct indentation: > > - name: "Ensure FQDN is present" > lineinfile: > dest=/etc/sysconfig/network

[ansible-project] Syntax for skip_missing flag for with_subelements?

What is the syntax for the third element that can be used with_subelements to use to skip missing subelements? "Optionally, you can add a third element to the subelements list, that holds a dictionary of flags. Currently you can add the ‘skip_missing’ flag. If set to True, the lookup plugin

[ansible-project] Can update_cache be used with package module?

I often have playbooks fail to install because my apt or yum cache is out of date. Can I use update_cache: yes in the package module, or do I need to go back to using apt module and yum module? Thanks! Joanna -- You received this message because you are subscribed to the Google Groups

[ansible-project] Can individual variables be set to no_log?

I thought I ran across a playbook output that indicated individual variables had been set with a no_log parameter. Unfortunately, I cannot find the website where I saw it yesterday. Is it actually possible to specify no_log by variable? Thanks! Joanna -- You received this message because you

[ansible-project] Re: How to loop through blockinfile?

Yup, that worked. However, the blocks don't apply/respect indentation, which is another problem. On Tuesday, May 31, 2016 at 12:54:59 PM UTC-5, Joanna Delaporte wrote: > > I just realized the last example on the blockinfile page will probably > help me. Each "ANSIBLE MANAGED BL

[ansible-project] Re: How to loop through blockinfile?

I just realized the last example on the blockinfile page will probably help me. Each "ANSIBLE MANAGED BLOCK" statement needs a unique name, apparently. On Tuesday, May 31, 2016 at 12:50:29 PM UTC-5, Joanna Delaporte wrote: > > I have the following task, which appears to exec

[ansible-project] How to loop through blockinfile?

I have the following task, which appears to execute successfully, but I end up with only one block (the last), while I actually want to have a block for each item entered into my file. How do I do that? task: - name: configure dns zone SRV records delegate_to: 127.0.0.1 blockinfile:

[ansible-project] Blockinfile indentation

I am using blockinfile to write lines to a vars file that includes a few different layers of indentation. How do I make sure the indentation levels are maintained correctly using the block | statement? Thanks! Joanna -- You received this message because you are subscribed to the Google Groups

[ansible-project] user management for IPA/IdM domains

and default assigned values. Is anyone on this list managing IPA/IdM user administration, group memberships, and hbac lists with Ansible? If so, do you use a lot of command/shell, or do you prefer a different method? Thanks! Joanna Delaporte -- You received this message because you

Re: [ansible-project] using wildcards to fetching src files

I am wondering if a remote fileglob mechanism has been added for fetch. If not, I'll use Daniel's solution from above. Thanks! Joanna On Monday, October 29, 2012 at 7:26:50 PM UTC-5, Michael DeHaan wrote: > > I prefer Daniel's approach in this case, but I think we can also make > syntax easier

[ansible-project] Re: packages = p['package'].split(',')\r\nAttributeError:

ckages/ubuntu-sso-client'] > > > Le mercredi 2 mars 2016 00:06:38 UTC+1, Joanna Delaporte a écrit : >> >> I just had a few weird attribute errors that were caused by having >> multiple installations of Ansible. I had a system install and I was trying >> to use a Githu

Re: [ansible-project] Re: Cloned Ansible from github, now ad-hoc fails with the error: 'module' object has no attribute 'COLOR_ERROR'

Hi Brian, Thanks! It turns out I had a system installation still hanging around in addition to the github clone. So, I pip uninstalled to remove the system installation of Ansible and now everything works again. On Wednesday, March 2, 2016 at 8:21:33 AM UTC-6, Brian Coca wrote: > > run

Re: [ansible-project] Re: Ansible pause module fails to abort when more than one host is in inventory list

I just upgraded my Ansible to follow development yesterday after seeing a bug report that had been closed with a patch. I'll test again when I'm on campus Thursday. Thanks! Joanna On Mar 2, 2016 7:39 AM, "Brian Coca" wrote: > I cannot reproduce the issue in 2.0.1, when I hit

[ansible-project] Re: packages = p['package'].split(',')\r\nAttributeError:

I just had a few weird attribute errors that were caused by having multiple installations of Ansible. I had a system install and I was trying to use a Github clone of Ansible. You can use the following command to see where your system Ansible locations might be (thanks to Matt Martz): $

Re: [ansible-project] Re: Updated ansible to track latest, now I have error 'Connection' object has no attribute '_split_ssh_args'

Okay, so apparently that output is okay. The problem is that my system had another installation of Ansible on it. I asked pip to uninstall it for me, and now Ansible playbooks and ad-hoc commands are working again. On Tuesday, March 1, 2016 at 4:20:53 PM UTC-6, Joanna Delaporte wrote

Re: [ansible-project] Cloned Ansible from github, now I have 'module' object has no attribute 'COLOR_ERROR' (too many Ansibles!)

Matt, I owe you a drink. sudo pip uninstall ansible removed the extra ansible installation. Now ansible functions again from the git clone. Thanks! Joanna On Tuesday, March 1, 2016 at 4:46:13 PM UTC-6, Joanna Delaporte wrote: > > It appears that I do have another installation as wel

Re: [ansible-project] Cloned Ansible from github, now I have 'module' object has no attribute 'COLOER_ERROR'

HONPATH from using env-setup at the > front after ''. If not, then something could be inserting additional paths > for python to search, and you could be grabbing files from incorrect > locations. > > On Tue, Mar 1, 2016 at 4:34 PM, Joanna Delaporte <joannad...@gmail.com >

[ansible-project] Re: Cloned Ansible from github, now ad-hoc fails with the error: 'module' object has no attribute 'COLOR_ERROR'

t; > > On Tuesday, March 1, 2016 at 4:14:03 PM UTC-6, Joanna Delaporte wrote: >> >> That sounds like a bug, then. Do you know where to report it? >> >> On Tuesday, March 1, 2016 at 4:10:25 PM UTC-6, Arthur Reyes wrote: >>> >>> To be clear, i

Re: [ansible-project] Cloned Ansible from github, now I have 'module' object has no attribute 'COLOER_ERROR'

d version of ansible. I > might recommend completely uninstalling and reinstalling ansible. > > On Tue, Mar 1, 2016 at 3:32 PM, Joanna Delaporte <joannad...@gmail.com > > wrote: > >> When running ad-hoc ansible commands, I get the following error: >> >> $ a

Re: [ansible-project] Re: Updated ansible to track latest, now I have error 'Connection' object has no attribute '_split_ssh_args'

sible-user/ansible/docs/man: Remember, you may wish to specify your host file with -i Done! On Tuesday, March 1, 2016 at 4:16:21 PM UTC-6, Joanna Delaporte wrote: > > Hi Matt, > > I did an rm -rf of my ansible installation and re-cloned it from github > right before reproduci

Re: [ansible-project] Re: Updated ansible to track latest, now I have error 'Connection' object has no attribute '_split_ssh_args'

e a partially updated version > of ansible. > > On Tue, Mar 1, 2016 at 3:18 PM, Joanna Delaporte <joannad...@gmail.com > > wrote: > >> Hi Jon, >> >> Thanks for the advice. I tried running that command, followed by the >> submodule commands, and still got t

[ansible-project] Re: Cloned Ansible from github, now ad-hoc fails with the error: 'module' object has no attribute 'COLOR_ERROR'

Fixed the typo in the subject. The ad-hoc commands are failing instantly after this error. On Tuesday, March 1, 2016 at 3:32:58 PM UTC-6, Joanna Delaporte wrote: > > When running ad-hoc ansible commands, I get the following error: > > $ ansible slackware-lab-64 -m setup > >

[ansible-project] Re: extra-vars behavior

I've used multiple vars similar to your second example for quite a while, and it's documented to be used like that , so I believe it is a feature, not a bug. I'm not sure I have ever used quotes

[ansible-project] Re: tarballs failing with ValueError: No JSON object could be decoded

What does the play look like that is running this task? On Tuesday, March 1, 2016 at 3:33:23 PM UTC-6, Niche Meister wrote: > > In the middle of multiple tarball installations, ansible started throwing > the below error... help! > TASK: [install-tarball | download the file if the tag is missing

[ansible-project] Cloned Ansible from github, now I have 'module' object has no attribute 'COLOER_ERROR'

When running ad-hoc ansible commands, I get the following error: $ ansible slackware-lab-64 -m setup [WARNING]: Error when using >: 'module' object has no attribute 'COLOR_ERROR' I am also getting the ssh_split error whenever I try to run a playbook, so my Ansible is completely broken right

[ansible-project] Re: Updated ansible to track latest, now I have error 'Connection' object has no attribute '_split_ssh_args'

t; > Doesn't sound like exactly the same issue but perhaps worth a look. > > Hope this helps, > > Jon > > On Thursday, 25 February 2016 22:19:36 UTC, Joanna Delaporte wrote: >> >> I just git pulled the latest build from GitHub, and then ran the >> following t

[ansible-project] Re: Ansible pause module fails to abort when more than one host is in inventory list

If we need to file a bug on this, please tell me where. It is important to me that this feature work. On Tuesday, March 1, 2016 at 9:22:17 AM UTC-6, binbash root wrote: > > Originally opened bug #14252 but this bug for Ansible 2.0.0.1, but this > still appears to be in Ansible 2.0.1.0. It

[ansible-project] Updated ansible to track latest, now I have error 'Connection' object has no attribute '_split_ssh_args'

I just git pulled the latest build from GitHub, and then ran the following two commands as recommended on the Installation page: $ git pull --rebase $ git submodule update --init --recursive However, I am now getting an exception that the 'Connection' object has no attribute '_split_ssh_args'

[ansible-project] Ansible Docs site search is not working

I am not able to search using the Google Custom Search field on docs.ansible.com in either Firefox (43) or Chrome (47). To be more specific, the URL loads with the search terms, but the search results do not appear on the page. The page remains the original Welcome to Ansible documentation

[ansible-project] Interrupting a playbook - interrupt failed to prevent a reboot for ~ 1/2 of hosts in a group

I have a playbook that has a reboot directive at the end, with a pause and prompt at the end to allow me to interrupt and abort the playbook without rebooting. Or at least, it used to. I pointed the playbook at ~28 machines. Twice yesterday, it rebooted half the machines even though the play

[ansible-project] Ansible 2.1.0, I used --limit to run a playbook and host file hostvars for the host didn't load

Last night, I ran a playbook that had a group specified for hosts. That group has a custom variable defined in my inventory. On the command line I used --limit=jupiter to test on a single host. The defined hostvar was not loaded, apparently, because my machine had patches installed from a

Re: [ansible-project] ansible.cfg file...where to find a new copy after in-place upgrade to 2.0?

Great, thanks! On Monday, December 7, 2015 at 2:00:46 PM UTC-6, Brian Coca wrote: > > In the repo its always been in examples/ansible.cfg > > -- > Brian Coca > -- You received this message because you are subscribed to the Google Groups "Ansible Project" group. To unsubscribe from this

Re: [ansible-project] How to make Ansible output look nice?

Hi, thanks! I read through the blog and attempted to set it up, but it's not working for me yet. On Dec 9, 2015 11:25 AM, "Paul Hardwick" wrote: > I hadn't gotten a chance to install yet, but I had found this when I was > looking for myself. > > >

[ansible-project] How to make Ansible output look nice?

I have been looking around and cannot find the answer to this...my ansible output looks terrible, and is difficult to read. What is the best way to make it more readable? (item={'value': {u'services': [u'dns'], u'sources': [u'0.0.0.0/0'], u'state': u'enabled', u'permanent': u'True'}, 'key':

[ansible-project] Re: Reusing a set of taks across many roles

You can include role dependencies in the *meta/main.yml* for a role. These are roles that will also be installed by the role, prior to the role installing itself. So, for common things you need on all machines, you can have a 'common' role. Then, in the meta/main.yml for more specific roles,

[ansible-project] Re: ansible galaxy question

Hi Serge, Sorry, tab works differently in chrome than vim. Maybe I can find a plugin to fix my browser. :). I was going to write this small example of a list of dependencies that could be listed in your meta/main.yml file: #cloudera-hadoop/meta/main.yml dependencies: - serge.common

[ansible-project] Re: ansible galaxy question

I believe Galaxy roles can have role dependencies stated in meta/main.yml, just like any other Ansible role on your system. So, individual roles could be written, and then encapsulated into a more elaborate role. For example, in the role you have provided a link, you could separate out hue and

Re: [ansible-project] yum module might be broken in RC1...

Hi Brian, Okay, but why does my play work with package, but not yum? On Nov 25, 2015 1:53 PM, "Brian Coca" wrote: > package will just call yum module, the main difference is in the > 'squashing', another check you can do to confirm this is set the env > var

Re: [ansible-project] yum module might be broken in RC1...

Alright, it worked with that variable you requested. I exported the env var (it didn't work just set in the current shell): export ANSIBLE_SQUASH_ACTIONS="apt,pkgng" And ran this play: - yum: name="{{ item }}" state=present with_items: - ypbind - rpcbind The play succeeded with this

Re: [ansible-project] yum module might be broken in RC1...

Yes, right. Sorry for my misunderstanding. I'll get back to you with a test run after I feed my kids dinner. On Nov 25, 2015 6:10 PM, "Brian Coca" wrote: > if you run the test i asked, i can confirm it will be because of > argument squashing, which is an optimization that will

Re: [ansible-project] yum module might be broken in RC1...

-packages/ansible/modules (Slackware 14.1) This will probably fix several errors I have been trying to work around, mostly related to lists breaking, in 2.0. Thanks! Joanna On Wednesday, November 25, 2015 at 8:55:43 PM UTC-6, tkuratomi wrote: > > On Wed, Nov 25, 2015 at 5:19 PM, Joanna Del

Re: [ansible-project] yum module might be broken in RC1...

the yum module. > Perhaps there's some old version of ansible present on the machine and > its yum module is overriding the one from rc1 or you have the > ANSIBLE_LIBRARY environment variable set to some path where an > alternative yum module is living. > > -Toshio > >

Re: [ansible-project] Attempting to yum install multiple packages using with_items results in Python AttributeError: 'list' object has no attribute 'split'

with only one item works (both of the following succeeded): - yum: name=ypbind state=present - yum: name: rpcbind state: present On Wednesday, November 25, 2015 at 1:00:42 PM UTC-6, Joanna Delaporte wrote: > > Hi David, > > Thanks. I also tried it with more indentatio

[ansible-project] yum module might be broken in RC1...

sible-tmp-1448479744.44-240902329915911/yum\", line 716, in ensure\r\nitems = pkgspec.split(',')\r\nAttributeError: 'list' object has no attribute 'split'\r\n", "parsed": false} On Wednesday, November 25, 2015 at 1:00:42 PM UTC-6, Joanna Delaporte wrote: > > Hi David

[ansible-project] Attempting to yum install multiple packages using with_items results in Python AttributeError: 'list' object has no attribute 'split'

Good afternoon, I am attempting to install multiple packages on a CentOS 7 server. I am using 2.0.0-rc1. I keep running into a python error. I tried using a yum task in a role, using with_items like so: - yum: name={{ item }} state=present with_items: - ypbind - rpcbind I think that is

Re: [ansible-project] Attempting to yum install multiple packages using with_items results in Python AttributeError: 'list' object has no attribute 'split'

indentation problem, > - yum: name={{ item }} state=present > with_items: > - ypbind > - rpcbind > > > David Karban > Linux server specialist/Specialista na správu linuxových serverů > www.karban.eu > > 2015-11-25 18:40 GMT+01:00 Joanna Delaporte <joan

[ansible-project] How to fix this nmcli error - Error: invalid . 'autoconnect'.\n"

fatal: [redacted]: FAILED! => {"changed": false, "failed": true, "invocation": {"module_args": {"autoconnect": "yes", "conn_name": "enp5s0", "dns4": "redacted", "gw4": "redacted", "ip4": "redacted", "state": "present", "type": "ethernet"}, "module_name": "nmcli"}, "msg": "Error: invalid .

[ansible-project] Re: Howto Install nmcli extras module?

23, 2015 at 11:51:01 AM UTC-6, Joanna Delaporte wrote: > > Hi everyone, > > I was trying to use the nmcli module this morning, and kept getting an > error (ERROR: nmcli is not a legal parameter in an Ansible task or > handler). Apparently, this module doesn't ship with

Re: [ansible-project] Howto Install nmcli extras module?

Hi Brian, I just noticed now it is in version 2.0. Now to find instructions for installing that version... Thanks! Joanna On Monday, November 23, 2015 at 12:06:07 PM UTC-6, Brian Coca wrote: > > The extras get installed with Ansible, you don't need to do anything. > > The most probable cause

[ansible-project] Re: Howto Install nmcli extras module?

inventory/ packaging/ > > $ ls /usr/share/ansible/network/ > get_url slurp uri > > On Monday, November 23, 2015 at 11:51:01 AM UTC-6, Joanna Delaporte wrote: >> >> Hi everyone, >> >> I was trying to use the nmcli module this morning, and kep

[ansible-project] Re: nmcli errors when adding more than one DNS server

y, November 23, 2015 at 9:35:25 PM UTC-6, Joanna Delaporte wrote: > > > The documentation states that multiple DNS entries should be formatted > like so: > > A list of upto 3 dns servers, ipv4 format e.g. To add two IPv4 DNS server > addresses: ["8.8.8.8 8.8.4.4"]

[ansible-project] Re: How to fix this nmcli error - Error: invalid . 'autoconnect'.\n"

The module code appears to break while parsing the dictionary into a string, at or near line 505. This happens with the DNS entry when I pass in more than one entry as well. 505 dstr+="%s: %s\n" % ( key, str_val) On Monday, November 23, 2015 at 8:40:20 PM UTC-6, Joanna Delap

[ansible-project] Re: How to fix this nmcli error - Error: invalid . 'autoconnect'.\n"

d returned the same "'autoconnect'.\n" in the debug output. Removing the autoconnect assignment made the play run smoothly for each different syntax attempt. On Monday, November 23, 2015 at 1:10:40 PM UTC-6, Joanna Delaporte wrote: > > fatal: [redacted]: FAILED! => {&qu

[ansible-project] nmcli errors when adding more than one DNS server

The documentation states that multiple DNS entries should be formatted like so: A list of upto 3 dns servers, ipv4 format e.g. To add two IPv4 DNS server addresses: ["8.8.8.8 8.8.4.4"] Following that example, I get the error: "Error: failed to modify ipv4.dns: invalid IPv4 address

[ansible-project] Re: Howto Install nmcli extras module?

be an issue. On Monday, November 23, 2015 at 12:43:20 PM UTC-6, Joanna Delaporte wrote: > > After installing 2.0.0 beta 3 from a tarball, do I need to do something to > reload modules or plugins? > > FAILED! => {"failed": true, "msg": "ERROR! Invalid sh

[ansible-project] Howto Install nmcli extras module?

Hi everyone, I was trying to use the nmcli module this morning, and kept getting an error (ERROR: nmcli is not a legal parameter in an Ansible task or handler). Apparently, this module doesn't ship with Ansible, as noted on the Extras Modules page

[ansible-project] Setup facts: ansible_default_ipv4 is empty, even when interface is active with an ip address

I am writing a playbook where I want to name a file including the ipaddress. However, my default ipv4 in the facts is empty, even though my interface is active and has a static ip. My target machine is CentOS 7.1, and my ansible version is 1.9.3. *"ansible_default_ipv4": {}, *

[ansible-project] Re: Setup facts: ansible_default_ipv4 is empty, even when interface is active with an ip address

y one key in it. I'm hoping to end up with something like ansible_52.x.x.x, ansible_192.x.x.x, ansible_176.x.x.x files instead, each with the ansible user's key on that host. On Wednesday, November 18, 2015 at 10:27:46 AM UTC-6, Joanna Delaporte wrote: > > I am writing a playbook w

Re: [ansible-project] Re: msg: hostname module cannot be used on platform Linux (Centos linux)

I did a fresh install on a new CentOS 7.1.1503. I got the same error response to the hostname module from my Slackware control machine. I installed ansible 1.9.3 and git cloned enough over to my Ubuntu 14.04 laptop to rerun the same playbook against the same CentOS 7.1 target. The error was

[ansible-project] Re: msg: hostname module cannot be used on platform Linux (Centos linux)

I updated to Ansible 1.9.4, and I still have the same issue on both target hosts. On Wednesday, November 11, 2015 at 2:44:02 PM UTC-6, Joanna Delaporte wrote: > > Hello everyone, > > I am getting the following error while attempting to run a Galaxy role > that was published for

[ansible-project] Re: msg: hostname module cannot be used on platform Linux (Centos linux)

aborting Reading through the module code, it seems I am getting dumped right into UnimplementedStrategy class, but I don't understand yet how it's deciding which OS and distribution version it's looking at, before getting to the UnimplementedStrategy. It's the first Ansible module that I've ev

[ansible-project] Re: Manage Unix Users Over Multiple Evironments with Ansible

Does the following work? when: "inventory_hostname in item.*value*.access_to" I have used dicts a little for users, and that is how I reference details for users. Joanna On Friday, November 13, 2015 at 12:02:25 PM UTC-6, Ralph Bolton wrote: > > Hi, > > I'm trying to manage a small number

Re: [ansible-project] Re: msg: hostname module cannot be used on platform Linux (Centos linux)

Yes, I can tell that's how it's supposed to work. However, it's like my hosts don't even get past the distribution match, wherever that happens in the code...they get dumped straight into the UnimplementedStrategy class for error handling. If the code were recognizing CentOS, it would be using

Re: [ansible-project] Re: msg: hostname module cannot be used on platform Linux (Centos linux)

Actually, it happens both on a local install of CentOS 7.0.1406 and AWS's CentOS7.0.1406. hostnamectl works fine on my AWS host. On Friday, November 13, 2015 at 1:08:20 PM UTC-6, Greg Swift wrote: > > Joanna found that it has something to do with the AWS CentOS7 install. I'm > assuming based

Re: [ansible-project] Re: msg: hostname module cannot be used on platform Linux (Centos linux)

platform.dist()) ('centos', '7.1.1503', 'Core') >>> print(platform.linux_distribution()) ('CentOS Linux', '7.1.1503', 'Core') On Friday, November 13, 2015 at 3:40:44 PM UTC-6, Joanna Delaporte wrote: > > I was wrong about CentOS 7.1 not being available...it is now the CentOS > version you can get

  1   2   >