Problem on remote server setup - ssmtp FAILED

bedrock
1

Hi,

I have a problem on setting up a production remote server on a Digital Ocean droplet, running an Ubuntu 14.04 x64 using the bedrock-ansible stack (https://github.com/roots/bedrock-ansible)

on a MacOs under Yosemite running: $ ansible-playbook -i hosts/production server.yml --ask-pass -vvvv I get the following error message after a while:

 TASK: [mariadb | Remove the test database] ************************************ 
<45.55.217.234> ESTABLISH CONNECTION FOR USER: root on PORT 22 TO 45.55.217.234
<45.55.217.234> REMOTE_MODULE mysql_db name=test state=absent login_user="root" login_password=VALUE_HIDDEN
<45.55.217.234> EXEC /bin/sh -c 'mkdir -p $HOME/.ansible/tmp/ansible-tmp-1430673939.38-121352451011028 && echo $HOME/.ansible/tmp/ansible-tmp-1430673939.38-121352451011028'
<45.55.217.234> PUT /var/folders/bz/13h9tsr523sdbr303t55q4500000gn/T/tmpvXgBXc TO /root/.ansible/tmp/ansible-tmp-1430673939.38-121352451011028/mysql_db
<45.55.217.234> EXEC /bin/sh -c 'LANG=en_US.UTF-8 LC_CTYPE=en_US.UTF-8 /usr/bin/python /root/.ansible/tmp/ansible-tmp-1430673939.38-121352451011028/mysql_db; rm -rf /root/.ansible/tmp/ansible-tmp-1430673939.38-121352451011028/ >/dev/null 2>&1'
ok: [45.55.217.234] => {"changed": false, "db": "test"}

TASK: [ssmtp | Install ssmtp] ************************************************* 
<45.55.217.234> ESTABLISH CONNECTION FOR USER: root on PORT 22 TO 45.55.217.234
fatal: [45.55.217.234] => {'msg': 'FAILED: timed out', 'failed': True}

FATAL: all hosts have already failed -- aborting

PLAY RECAP ******************************************************************** 
           to retry, use: --limit @/Users/vicensfayos/server.retry

45.55.217.234              : ok=28   changed=1    unreachable=1    failed=0

I configured groups_var/production and groups_var/all as described in manual (I hope) and created an account on mandrillapp.com.

using ansible 1.9.0.1

noticed that after this error the digital ocean droplet is unreachable and it has to be restarted.

Any guidance will be very appreciated.

Thanks!

2

I just changed the order on server.yml roles (so ssmtp task is rolled out before) and then it works BUT unfortunately it crash for the same reasons later on when PHP role is rolled out

TASK: [php | Add PHP 5.6 PPA] ************************************************* 
<45.55.217.234> ESTABLISH CONNECTION FOR USER: root on PORT 22 TO 45.55.217.234
fatal: [45.55.217.234] => {'msg': 'FAILED: timed out', 'failed': True}

FATAL: all hosts have already failed -- aborting

Last try… I enabled the flag HHVM to true and then it fails on the HHVM role execution.

3

It seems like you’re having some connection problems unfortunately. I haven’t heard of anyone else having multiple problems with time outs like this.

Can’t be a coincidence that the three problems you have are all timeouts. It can be pretty hard to debug connection problems though and the weird part is this is happening on a DO droplet. I know lots of people have gone through the same process without these timeouts.

4

Hi,

Finally I moved from DO to AWS and it worked well. I did not really know why I was experiencing this connections problems. DO worked well in other projects. Food for thoughts.

Thanks for the reply!

Vicens