Add check_and_failed option to pacemaker_cluster module

[matbu]

In order to not set a state to a cluster while the
cluster in not in the expected state we need to check
it and fail, or poll until the cluster is in a required
state, example:

• name: Check pacemaker cluster running before upgrade
  pacemaker_cluster: state=online check_and_fail=true
  async: 30
  poll: 4

SUMMARY

ISSUE TYPE

• Feature Pull Request

COMPONENT NAME

ADDITIONAL INFORMATION

[felixfontein]

Please add a changelog fragment in changelogs/fragments/.

(You probably have to rebase with current master to see that directory. You can also simply create it yourself without rebasing; git does not care about directories, only about files.)

[matbu]

@felixfontein hm I rebased my fork, but I probably did mistake, I'm getting all commits from my last push :/

[felixfontein]

You need to update your local master first, probably, and then rebase against master. If you do git rebase -i master, remove all commits from the list which don't belong to your PR.

[gundalow]

bot_skip

[gundalow]

@matbu Git can be a pain sometimes, it's easy to get it into a confused state.

You should be able to git pull --rebase upstream master.

We have this guide for ansible/ansible https://docs.ansible.com/ansible/devel/dev_guide/developing_rebasing.html though remember that

1. You'll need to refer to community.general instead
2. This is using the master branch, not devel

[Andersson007]

@matbu i'll close the PR. Feel free to create a new one instead (or claim to reopen this if you want to rebase), thanks!