Unable to install fullstack on single Ubuntu server

[Sidd Maini]

Hello,

I was trying to follow instructions given at https://github.com/edx/configuration/wiki/edX-Ubuntu-12.04-64-bit-Installation to install OpenEdx on a single Ubuntu server. Everything worked fine except the last command:

sidd@openedx:/var/tmp/configuration/playbooks$ cd /var/tmp/configuration/playbooks && sudo ansible-playbook -c local ./edx_sandbox.yml -i "localhost,"

ERROR: debug is not a legal parameter in an Ansible task or handler

I tried searching for where the error lies but it seems like a version problem to me. Please help.

My server specs:

Ubuntu 15.04

I am attaching a screen shot that lists the ansible packages installed and the error I am getting.

Thanks,

Sidd

[Ishmael Cuyson]

Hi Sidd! Yes it is a version problem. It's recommended to use 12.04 64bit ubuntu server.

[Siddharth Maini]

Thanks Ishmael. I will give this a shot asap. 

Sidd

--
You received this message because you are subscribed to a topic in the Google Groups "General Open edX discussion" group.

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/5f210e71-0025-4cc2-99fc-6ccc5448ccdd%40googlegroups.com.

[Sidd Maini]

I think default OS to Ubuntu 12.04 helped. The installed progressed quite a bit but resulted in error below.

TASK: [xqwatcher | restart xqwatcher] *****************************************

failed: [localhost] => {"failed": true}

msg: xqwatcher: ERROR (not running)

xqwatcher: ERROR (abnormal termination)

FATAL: all hosts have already failed -- aborting

Any thoughts on why this is happening? 

[mino...@gmail.com]

I get the same error (msg: xqwatcher: ERROR (not running) xqwatcher: ERROR (abnormal termination))
and it's frustrating because I have tried some things that might have caused it with no success.
It seems that all the services are up, though when I run the following:

sudo /edx/bin/supervisorctl -c /edx/etc/supervisord.conf status

and also I can see the LMS and CMS pages at the browser but I worry that after this error some things were not installed properly and I will have trouble in the future!
I also would appreciate any help in this!
Dim

[alber...@gmail.com]

Hello all,

Could you fix this xqwatcher error? How was it going? I have had same error for two days and no clue where it comes from.

Thanks.

[Sidd Maini]

I haven't been able to fix it. But looking at https://github.com/edx/configuration/wiki/edX-Ubuntu-12.04-64-bit-Installation it seems like there might some PIP issues. I am not sure what the real cause. 

However, I have been able to install the Birch release using the Bitnami stack https://bitnami.com/stack/edx/installer on a local machine. It is working fine but has some minor issues though.

If you have EC2, you should be able to deploy it.

[Albert Brian]

Hello Sidd,

Thank you for your reply. Currently I keep tracking where its problem is. It could be 'enviroment' variable in xqwatcher.js2 file but I am not very sure right now. I do appreciate your suggestion. I dont have EC2 since I am in Europe. Does bitnami work as a server production or does it work as Virtual Machine for local uses?

Thanks again!

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/35f66821-902c-4b4f-be41-4b3351c8a1c2%40googlegroups.com.

[Siddharth Maini]

Albert, 

You might want to check again for EC2 availability in Europe. See this: http://aws.amazon.com/about-aws/global-infrastructure/

Bitnami stack can work as a virtual machine using VMware image they have and also on EC2. They have the birch release available for EC2.

I am not sure about it being a production ready system. They released it last week ;) and so it needs to be tested. I am testing it right now as we speak :)

Let me know if are able to figure out the basic installation issue.

Sidd

User Experience Engineer

School of Informatics and Computing

Indiana University - Bloomington

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/CAPA%2BH3qVdeAuQ6mMZeTfNw--K16UVRHtzxJBsjXq4kLH6t3UjA%40mail.gmail.com.

[Albert Brian]

Hi Sidd,

I download the installer and I am installing it. I am not sure if that is the production which you said it was released last week, but it looks promissing :)

Albert

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/CANk6dmVZjR6-kXUqKpjL%3Dp1OssNbP2jhfk_0LNys9pMTnS%2B6vA%40mail.gmail.com.

[Albert Brian]

There is a progressing bar 0% - 50% - 100%. It stopt at 100% for awhile, is it ok?

[Siddharth Maini]

Yes it is OK!

Just wait until it finishes.

Did you enable email? I tested both Gmail and Custom setting and it works fine.

If you are using email, please make sure when it asks for your username that you fill in your ip address:8080 so it can generate the correct URL otherwise you will have to manually insert :8080 in the activation link, without which it doesn't work.

Sidd

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/CAPA%2BH3qmLB_UPDE3dqTzFmXWGgK_KK1nkPueoSgxpa%2Bq8d-ohQ%40mail.gmail.com.

[Albert Brian]

I am running it on Ubuntu command line. Do you know if we can uninstall it so we can do fresh reinstall? It has reached and stopt at 100% for 20' without any comment lol.

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/CANk6dmWe7hLnKLeD8%2B_7rbDnj%2BMS_VVuC%3DDR1qgHtZZYBAVJVA%40mail.gmail.com.

[SIDDHARTH MAINI]

I would just wait. It does take a few minutes. There is also an uninstall script in the folder that gets created that you can use to uninstall.

---

From: Albert Brian
Sent: ‎6/‎17/‎2015 9:33 AM
To: edx-...@googlegroups.com
Subject: Re: [edx-code] Re: Unable to install fullstack on single Ubuntuserver

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/CAPA%2BH3oG%2BBpcATAi5rP6qb6anpissXc4XfQhkXLo2DqAiqpKOw%40mail.gmail.com.

[Albert Brian]

Great! We keep up-to-date in personal email.

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/558177cb.920d6b0a.3af4.7ce2%40mx.google.com.

[Albert Brian]

Go back to the topic, I believed that is the only error left which i could not fix, so my solution was hiding "restart xqwatcher" task. Then bravo, it's done! :-) Yes, I am going to fix and run it later.

[manz...@gmail.com]

Hi all,

We are experiencing the exact same error when installing edX Birch release on a fresh Ubuntu 12.04 64 bit AWS instance. I'm doing a  export OPENEDX_RELEASE=named-release/birch before running the vagrant.sh file (as instructed in: https://github.com/edx/configuration/wiki/edX-Ubuntu-12.04-64-bit-Installation)

I tried to increase the verbosity of the provisioner  i.e. ansible (by passing a -vvv argument to the ansible-playbook command) to see if it shows any error, and I was getting this then:

TASK: [xqwatcher | restart xqwatcher] *****************************************

<localhost> REMOTE_MODULE supervisorctl state=restarted supervisorctl_path=/edx/app/xqwatcher/venvs/supervisor/bin/supervisorctl config=/edx/app/xqwatcher/supervisor/supervisord.conf name=xqwatcher
<localhost> EXEC ['/bin/sh', '-c', 'mkdir -p /tmp/ansible-tmp-1434625127.41-134718173992986 && chmod a+rx /tmp/ansible-tmp-1434625127.41-134718173992986 && echo /tmp/ansible-tmp-1434625127.41-134718173992986']
<localhost> PUT /tmp/tmpKmYmq2 TO /tmp/ansible-tmp-1434625127.41-134718173992986/supervisorctl
<localhost> EXEC ['/bin/sh', '-c', u'chmod a+r /tmp/ansible-tmp-1434625127.41-134718173992986/supervisorctl']
<localhost> EXEC /bin/sh -c 'sudo -k && sudo -H -S -p "[sudo via ansible, key=cimvnlciorcrxrlrowjfwhhedxqeoesp] password: " -u xqwatcher /bin/sh -c '"'"'echo BECOME-SUCCESS-cimvnlciorcrxrlrowjfwhhedxqeoesp; LANG=en_US.UTF-8 LC_CTYPE=en_US.UTF-8 /usr/bin/python /tmp/ansible-tmp-1434625127.41-134718173992986/supervisorctl'"'"''
<localhost> EXEC ['/bin/sh', '-c', 'rm -rf /tmp/ansible-tmp-1434625127.41-134718173992986/ >/dev/null 2>&1']

failed: [localhost] => {"failed": true}

msg: xqwatcher: stopped

xqwatcher: ERROR (abnormal termination)


FATAL: all hosts have already failed -- aborting

PLAY RECAP ********************************************************************
           to retry, use: --limit @/home/ubuntu/edx_sandbox.retry

localhost                  : ok=564  changed=247  unreachable=0    failed=1  

Not helpful at all. Peeking at the log file for xqwatcher (located at /edx/var/log/xqwatcher/xqwatcher-stdout.log) reveals this:

No xqueue watchers configured

Then, I checked the source for xqueue-watcher/xqueue_watcher/manager.py and found that it checks for .json files in the directory /edx/app/xqwatcher/conf.d/, but there are none so the xqwatcher service terminates. Now the question is, do the files in conf.d/ need to be created/configured manually or by the ansible playbooks?

By the way, the issue was fixed (for me) by doing a git checkout named-release/birch in the edx-configuration repo and then running the edx_sandbox.yml playbook, although I already was setting the OPENEDX_RELEASE environment variable to named-release/birch. Before that, the edx-configuration repo was on the branch master. Anyone from the edX team, can they confirm if this was the issue?

[alber...@gmail.com]

Hello Mazoor,

File xqwatcher.json is found in /app/xqwatcher/xqwatcher.json. It is fine.

I will look into your error again to see if it is helpful. I think this is a new error or known error but experts don't mind.

I guess it could be that we didn't configure xqueue_watcher service, so xqwatcher couldn't work. However, I need to confirm this thought.

Are you sure that you fixed it just by change branch? From what i have known it only gives you same branch. Actually there is no branch called 'named-release/birch' in github anymore. There is only one birch called 'named-release/birch/rc'.

So how exactly have you fixed this error? This could be helpful. In my case, I fixed it by simply skipping it lol.

Albert

[alber...@gmail.com]

Hello again,

Please post your issue here to get attention from devs: https://github.com/edx/configuration/issues/2096

Albert

[Manzoor Ahmed]

Hi Albert,

I'm not sure about the git branch existing, but doing a `git status` on that same edx-configuration repo (where edX is installed) outputs this:

/var/tmp/configuration$ git status
# HEAD detached at named-release/birch
nothing to commit, working directory clean

And that's exactly how the installation got completed for me. I believe the reason it worked is because the old playbooks (after checking out named-release/birch) do not have the xqwatcher task which is failing.

[Albert Brian]

Hi Manzoor,

Yeah that is HEAD, not Branch :-)

However, I have checked and seen named-release/birch was behide master, it didn't have xqwatcher. That means it missing grader.

Albert

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/49c1ffd1-a04c-4bcd-b2dd-4868b0d6c207%40googlegroups.com.

[Sidd Maini]

Now I am getting an SSH error during installation. I used the master branch. I am not able to debug this. Please help!

TASK: [edx_service | tag instance] ********************************************

failed: [localhost] => (item=[{'REPO': 'edx-analytics-data-api.git', 'DOMAIN': u'github.com', 'VERSION': u'master', 'PROTOCOL': u'https', 'PATH': u'edx', 'DESTINATION': u'/edx/app/analytics_api/analytics_api', 'SSH_KEY': None}, {'invocation': {'module_name': u'git', 'module_args': u'repo=https://github.com/edx/edx-analytics-data-api.git dest=/edx/app/analytics_api/analytics_api version=master'}, 'item': {'REPO': 'edx-analytics-data-api.git', 'DOMAIN': u'github.com', 'VERSION': u'master', 'PROTOCOL': u'https', 'PATH': u'edx', 'DESTINATION': u'/edx/app/analytics_api/analytics_api', 'SSH_KEY': None}, 'changed': True, 'after': '99f9c79f153bf5bb04830ca95c678093b0b74439', 'before': None}]) => {"failed": true, "item": [{"DESTINATION": "/edx/app/analytics_api/analytics_api", "DOMAIN": "github.com", "PATH": "edx", "PROTOCOL": "https", "REPO": "edx-analytics-data-api.git", "SSH_KEY": null, "VERSION": "master"}, {"after": "99f9c79f153bf5bb04830ca95c678093b0b74439", "before": null, "changed": true, "invocation": {"module_args": "repo=https://github.com/edx/edx-analytics-data-api.git dest=/edx/app/analytics_api/analytics_api version=master", "module_name": "git"}, "item": {"DESTINATION": "/edx/app/analytics_api/analytics_api", "DOMAIN": "github.com", "PATH": "edx", "PROTOCOL": "https", "REPO": "edx-analytics-data-api.git", "SSH_KEY": null, "VERSION": "master"}}]}

msg: value of region must be one of: ap-northeast-1,ap-southeast-1,ap-southeast-2,cn-north-1,eu-central-1,eu-west-1,eu-central-1,sa-east-1,us-east-1,us-west-1,us-west-2,us-gov-west-1, got: nova

FATAL: all hosts have already failed -- aborting

[Sidd Maini]

Manzoor try git clone -b master https://github.com/edx/configuration

after in the tmp folder on a fresh ubuntu and follow all steps again.

That might help.

[Xavier Antoviaque]

Hi Sidd,

 

(You should be posting this on the openedx-ops mailing-list btw - edx-code is reserved to code-related discussions. If you reply, do it there, I'll be seeing it there too.)

 

It seems to be because you're using an OpenStack VM - this tagging task is AWS-specific, and doesn't recognize "nova" as a valid region. I use a workaround for this that you could cherry-pick: https://github.com/open-craft/configuration/commit/1197af261a16724821cd0a074417a67a71969073

 

I'd actually like to submit a proper patch for this, but I'm not sure what's the best approach. Is this tagging task necessary, or can it be safely skipped? Also, checking for "nova" is likely brittle, as this is configurable by the OpenStack cloud, and there can be arbitrary region names here I think, so a better solution would likely be needed.

 

--

Xavier

@OpenCraft

--

You received this message because you are subscribed to the Google Groups "General Open edX discussion" group.

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/9eeb54ff-4ef6-4a45-aa1c-be74b81f2edf%40googlegroups.com.

 

[Siddharth Maini]

Hi Xavier,

Thank you so much for recognizing nova. Yes I am using an openstack vm. Did you use the master branch config or the named-release/birch branch for this purpose?

How is your installation experience with it so far?

Thanks,

Sidd

--
You received this message because you are subscribed to a topic in the Google Groups "General Open edX discussion" group.
To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/1435128094.2266268.306245257.572CA4E4%40webmail.messagingengine.com.

[Xavier Antoviaque]

We mostly use it for sandboxes at this point, so it doesn't need the full production setup, but it works well for that use-case. I use the master branch of the configuration repo, and this is the only issue I know of for deploying the default fullstack setup.

 

--

Xavier

@OpenCraft

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/CANk6dmV-z-rUcZVwGW7Xo1N0K435pAdCnxmagLhq0dTzrrOasg%40mail.gmail.com.

 

[Traore Ousmane]

HI DEAR 

I WANT UNINSTALL BITNAMI EDX .

after this Warning: Problem running post-install step. Installation may not complete

correctly

 Error running curl -d "test" http://127.0.0.1:18080/: /bin/sh: 1: curl: not

found 

How i can resolve this probleme? 

thank you

[Sidd Maini]

you can stop all the services and then simply remove the directory to uninstall Bitnami. 

https://bitnami.com/stack/edx/README.txt

then you can try to install it again by downloading the latest release file at https://bitnami.com/stack/edx/installer

[Traore Ousmane]

THANK YOU 

--

You received this message because you are subscribed to a topic in the Google Groups "General Open edX discussion" group.

To view this discussion on the web visit https://groups.google.com/d/msgid/edx-code/a44a27e4-f7ff-409a-a91b-de9e78a9403e%40googlegroups.com.

--

 

BEST REGARDS

 

TRAORE Ousmane
Commercial Dpt Retail & Commercial Banking UBA CI

Network & System Engineer

Mobile : (+225) 07 482 303