Random Parsing Problems in EC2?

Archives Ansible Project
1

This isn't a problem in Ansible (I pretty sure) but I'm curious
whether other people have seen this. (This is with Ansible
1.7.2, but I don't think it matters).

I have a playbook containing a bunch of tasks. No big deal.
I run the playbook on my desktop machine to control a machine
running in AWS. What happens is that various tasks fail with
a message like:

failed: [ec2-XX-XXX-XX-XXX.compute-1.amazonaws.com] => {"failed": true, "parsed": false}

What's strange is that if I rerun the playbook right away, without
making *any* changes, I get the same error, except in a different
task. *Update* This morning when I came in, I again ran the same
playbook. It finished without any problems!

One such task is that failed is the following:

- name: Enable hstore

command: psql -d template1 -c 'create extension if not exists hstore;'

   sudo: yes
   sudo_user: postgres

It doesn't get much simpler than that.

I'm guessing that this is some kind of problem at AWS but
I don't know what it could be. Has anyone else seen this?

Cordially,
Jon Forrest

2

When you get a message like this there should be a “msg” below it that says what the error/response was.

Did you get this?

3

No. I checked all three examples of different tasks failing.
They all said the same thing, which was exactly (copying and
pasting, except the host name)

failed: [ec2-XX-XX-XX-XX.compute-1.amazonaws.com] => {"failed": true, "parsed": false}

FATAL: all hosts have already failed -- aborting

This is why I don't think it was Ansible that was failing.
I suspect something lower level.

Jon Forrest

4

Hmm, odd, at least in recent versions you should get some output back.

No output back would be very unusual.