search

varfish-org / varfish-installer-LEGACY

DEPRECATED, use https://github.com/bihealth/varfish-docker-compose instead
MIT License
1 stars 0 forks source link

Installation issue on Ubuntu server 20 #1

Open mlebeur opened 3 years ago

mlebeur commented 3 years ago

Hi, I was trying to install VarFish on Ubuntu server 20 but I encounter several issues :

I try to remove pretasks from plays/postgres.yml but the files are'nt find :

fatal: [postgres]: FAILED! => { "ansible_facts": {}, "ansible_included_var_files": [], "changed": false, "message": "Could not find file\nSearched in:\n\t/home//varfish-installer/.imported_roles/bihealth.postgres_server/vars/[]\n\t/home//varfish-installer/.imported_roles/bihealth.postgres_server/[]\n\t/home//varfish-installer/.imported_roles/bihealth.postgres_server/tasks/vars/[]\n\t/home//varfish-installer/.imported_roles/bihealth.postgres_server/tasks/[]\n\t/home//varfish-installer/plays/vars/[]\n\t/home//varfish-installer/plays/[] on the Ansible Controller.\nIf you are using a module and expect the file to exist on the remote, see the remote_src option" }

Have you try on Ubuntu server 20 OS ?

Another question where is the Web client ?

Thank's a lot for your wonderfull work very inspiring.

Best

Manuel

stolpeo commented 3 years ago

Hello Manuel,

thank you very much for your interest in VarFish and your kind words! I also want to apologize that you experience issues with the installation. Currently, we are testing and revising the installation routine, so it is still work-in-progress, as you might have noticed.

That being said, the installation via ansible only supports CentOS 7 at this point in time. Unfortunately the manual is not clear in this. Technically, VarFish can also run on Ubuntu, and it is on our list to support other distributions in ansible as well. Do you have the possibility to set up a clean CentOS 7 for now?

My colleague @holtgrewe who primarily worked on the ansible scripts might add his thoughts (about next week).

The github repository of VarFish Web is located here: https://github.com/bihealth/varfish-server

And the checkout on the server instance will be located here: /srv/varfish-web/master

Cheers, Oliver

mlebeur commented 3 years ago

Thanks a lor for your quick answer. I will try with a CentOS7 distribution for now. Thanks for the links. Cheers, Manuel

mlebeur commented 3 years ago

Hi, I am trying to install VarFish on CentOS7 but I have some issue. First the root must have permission to ssh in sshd_config (or sshkegen) for the make jannovar/postgres step. Then during make varFish step I have an issue with on the "TASK [bihealth.postgres_client : on postgres host -- create database] " postgres_client_host is not defined. So I just add the variable definition in varfish-installer/inventories/production/group_vars/all/servers.yml as follow : postgres_client_host: "127.0.0.1" Then the make varFish goes to the TASK [bihealth.sodar_core_app : create virtualenv environment] . Where i have this issue : ################################### fatal: [varfish]: FAILED! => { "changed": true, "cmd": "set -xeo pipefail\ntrap \"rm -rf /srv/varfish-web/master-venv\" ERR\n$(which virtualenv) -p /usr/bin/python3.6 /srv/varfish-web/master-venv\n", "delta": "0:00:00.003513", "end": "2020-10-22 13:23:40.805035", "rc": 127, "start": "2020-10-22 13:23:40.801522" }

STDERR:

MSG:

non-zero return code ###################################

I was thinking about a classical problem or name between python 2 vs 3 sor I create a symbolic link ln -s /bin/virtualenv-3 /bin/virtualenv. Then when I rerun the make varFish I have similar issue but with much more logs.

################################### fatal: [varfish]: FAILED! => { "changed": true, "cmd": "set -xeo pipefail\ntrap \"rm -rf /srv/varfish-web/master-venv\" ERR\nsource /srv/varfish-web/master-venv/bin/activate\npip install -r requirements/production.txt\n", "delta": "0:00:38.797057", "end": "2020-10-22 13:35:46.584571", "rc": 1, "start": "2020-10-22 13:35:07.787514" }

STDOUT:

Collecting python-dotenv==0.9.1 Downloading python_dotenv-0.9.1-py2.py3-none-any.whl (17 kB) Collecting wheel==0.34.2 Downloading wheel-0.34.2-py2.py3-none-any.whl (26 kB) ...

 building 'psycopg2._psycopg' extension
creating build/temp.linux-x86_64-3.6
creating build/temp.linux-x86_64-3.6/psycopg
gcc -pthread -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -m64 -mtune=generic -D_GNU_SOURCE -fPIC -fwrapv -fPIC -DPSYCOPG_VERSION=2.8.6 (dt dec pq3 ext) -DPG_VERSION_NUM=90224 -I/usr/include/python3.6m -I. -I/usr/include -I/usr/include/pgsql/server -c psycopg/psycopgmodule.c -o build/temp.linux-x86_64-3.6/psycopg/psycopgmodule.o -Wdeclaration-after-statement
In file included from psycopg/psycopgmodule.c:28:0:
./psycopg/psycopg.h:36:22: fatal error: libpq-fe.h: No such file or directory
 #include <libpq-fe.h>
                      ^
compilation terminated.

It appears you are missing some prerequisite to build the package from source.

You may install a binary package by installing 'psycopg2-binary' from PyPI.
If you want to install psycopg2 from source, please install the packages
required for the build and try again.

For further information please check the 'doc/src/install.rst' file (also at
<https://www.psycopg.org/docs/install.html>).

error: command 'gcc' failed with exit status 1
----------------------------------------

ERROR: Command errored out with exit status 1: /srv/varfish-web/master-venv/bin/python3.6 -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-k96uu23d/psycopg2/setup.py'"'"'; file='"'"'/tmp/pip-install-k96uu23d/psycopg2/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' install --record /tmp/pip-record-ede71h00/install-record.txt --single-version-externally-managed --compile --install-headers /srv/varfish-web/master-venv/include/site/python3.6/psycopg2 Check the logs for full command output.

MSG:

non-zero return code

###################################

I did not understand well what happen here. The problem seems linked to virtual env and psycog2. I tried to install it via root with yum install python-psycopg2 It seems to work as the new error is :

################################### TASK [bihealth.sodar_core_app : django -- prepare static files] fatal: [varfish]: FAILED! => { "changed": false, "cmd": "set -xeo pipefail\nsource /srv/varfish-web/master-venv/bin/activate\npython manage.py collectstatic --no-input\n", "delta": "0:00:00.295932", "end": "2020-10-22 13:52:44.474874", "rc": 1, "start": "2020-10-22 13:52:44.178942" }

STDERR:

MSG:

non-zero return code

################################### But this time even a yum install python-sqlalchemy does not work. This error persist. Sorry for the long post but I am not so far away from the end. Have you got any idea of what happened in there ? Does something I have done is wrong ?

Thanks a lot for your time,

Cheers,

Manuel