Closed kotfic closed 8 years ago
@aashish24 could you please review? Thanks!
Sure, I didn't realize that this was ready for review. I thought you were still looking into memory issues?
How do I test this?
cd deploy/aws2/
local_vars.yml
in `deploy/aws2/``
prefix: aashish_NEX
# Make sure to set private key so we can access from dynamic inventory
ansible_ssh_private_key_file: [[ PATH TO YOUR AWS PEM FILE ]]
# General ec2 variables
ec2_key_name: kotfic-NEX
ec2_instance_profile_name: NEX
ec2_region: us-west-2
# Master related variables
ec2_master_ami: ami-e0f7e081
ec2_master_instance_type: c4.2xlarge
ec2_master_volumes:
- device_name: /dev/sda1
volume_size: 1000
# Make sure we're consistent with gobig ec2-pod role
ec2_master_additional_tags:
ec2_pod: "{{ prefix }}"
ec2_pod_instance_name: master
# Node related variables
ec2_node_ami: ami-e0f7e081
ec2_node_instance_type: c4.2xlarge
ec2_node_count: 2
ec2_node_volumes:
- device_name: /dev/sda1
volume_size: 1000
# Make sure we're consistent with gobig ec2-pod role
ec2_node_additional_tags:
ec2_pod: "{{ prefix }}"
ec2_pod_instance_name: nodes
AWS_ACCESS_KEY_ID
and AWS_SECRET_ACCESS_KEY
in your terminal. Run `export CLUSTER_ID=aashish_NEX``ansible-playbook -i "localhost," -e @local_vars.yml launch.yml"
ansible-playbook -i cumulus.py -e @local_vars.yml provision.yml
ansible-playbook -i cumulus.py -e @local_vars.yml terminate.yml
I can open a separate PR for resolving memory issues - those will be system dependent anyways
@aashish24 Please review.
Step 2 ran fine.
Step 3, I ran into this issue. I believe it could be because the instances launched are not reporting that they are using my key?
[chaudhary@einstein aws2 (aws_deploy_v2)]$ ansible-playbook -i cumulus.py -e local_vars.yml provision.yml [WARNING]: provided hosts list is empty, only localhost is available
PLAY [Provision the master node] *** skipping: no hosts matched
PLAY [Provision the regular nodes] ***** skipping: no hosts matched
PLAY [Start services and daemons on master node] *** skipping: no hosts matched
PLAY RECAP *****
[chaudhary@einstein aws2 (aws_deploy_v2)]$
two things to check:
which of the following are you running: ansible-playbook -i cumulus.py -e local_vars.yml provision.yml ansible-playbook -i cumulus.py -e @local_vars.yml provision.yml
the -e option to ansible playbook must have an arobase (@) if pulling extra variable from a file.
If you are running with the arobase, could you show me the output of running ./cumulus.py --list?
thanks!
On Fri, Apr 22, 2016 at 6:26 PM, Aashish Chaudhary <notifications@github.com
wrote:
Step 2 ran fine.
Step 3, I ran into this issue. I believe it could be because the instances launched are not reporting that they are using my key?
[chaudhary@einstein aws2 (aws_deploy_v2)]$ ansible-playbook -i cumulus.py -e local_vars.yml provision.yml [WARNING]: provided hosts list is empty, only localhost is available
PLAY [Provision the master node]
skipping: no hosts matched
PLAY [Provision the regular nodes]
skipping: no hosts matched
PLAY [Start services and daemons on master node]
skipping: no hosts matched
PLAY RECAP
[chaudhary@einstein aws2 (aws_deploy_v2)]$
— You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub https://github.com/OpenGeoscience/nex/pull/34#issuecomment-213610514
Christopher Kotfila R&D Engineer Kitware Inc. http://www.kitware.com/company/team/kotfila.html
@kotfic I am running it again to send you more infomation.
This time, I see this error:
[chaudhary@einstein aws2 (aws_deploy_v2)]$ ansible-playbook -i "localhost," -e @local_vars.yml launch.yml
PLAY ***************************************************************************
TASK [setup] *******************************************************************
ok: [localhost]
TASK [Launch master instance] **************************************************
An exception occurred during task execution. To see the full traceback, use -vvv. The error was: TypeError: __init__() got an unexpected keyword argument 'encrypted'
fatal: [localhost]: FAILED! => {"changed": false, "failed": true, "parsed": false}
NO MORE HOSTS LEFT *************************************************************
to retry, use: --limit @launch.retry
PLAY RECAP *********************************************************************
localhost : ok=1 changed=0 unreachable=0 failed=1
@kotfic doing it from start /scratch just to make sure that I didn't have anything bad locally.
What version of ansible are you using?
On Wed, Apr 27, 2016 at 12:53 PM, Aashish Chaudhary < notifications@github.com> wrote:
@kotfic https://github.com/kotfic doing it from start /scratch just to make sure that I didn't have anything bad locally.
— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/OpenGeoscience/nex/pull/34#issuecomment-215146507
Christopher Kotfila R&D Engineer Kitware Inc. http://www.kitware.com/company/team/kotfila.html
[chaudhary@einstein aws2 (aws_deploy_v2)]$ ansible --version ansible 2.0.1.0 config file = /home/chaudhary/tools/nex/src.git/deploy/aws2/ansible.cfg configured module search path = Default w/o overrides
@kotfic same error as last time for me in clean state as well.
please post the contents of your local_vars.yml
make sure to remove any sensitive information.
On Wed, Apr 27, 2016 at 1:00 PM, Aashish Chaudhary <notifications@github.com
wrote:
@kotfic https://github.com/kotfic same error as last time for me in clean state as well.
— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/OpenGeoscience/nex/pull/34#issuecomment-215148398
Christopher Kotfila R&D Engineer Kitware Inc. http://www.kitware.com/company/team/kotfila.html
@kotfic sent over the email.
After I updated boto version as suggested by @kotfic everyhing worked like a charm :clap: :2 LGTM
+2
@kotfic I am assuming you saw my +2. Looks good on my testing so far.
@aashish24 nope missed it, merging!
This addresses #30 by creating roles that allow for launching, terminating and provisioning clusters with spark and hdfs