D4Science Infrastructure

The catalog is going to serve a VRE that is still to be deployed ... the scope is not TextCrowd

I'm going to create the VRE right now and will let you know as soon as the VRE is there.

Andrea Dell'Amico wrote:

I guess that also the styles and plugins customizations are also missing?

I guess so. I think the ones available in the other instances would be fine, wouldn't they?

@leonardo.candela@isti.cnr.it please update this ticket as soon as the real context is available

Re plugins it is not planned that this instance of the catalogue harvest content from others (eg GN) or publishes its content (eg OAI-PMH). Is it possible to have a list of plug-ins we can configure per instance?

Re styles dunno what they are and what are the possible customizations. I would suggest going for an existing one (e.g. that of D4Science).

This catalogue will be used for data entry, we are going to define a specific resource typology.

We urgently need this machine, is there any planned date?

The ckan-eosc.d4science.org is close to be ready.
The Database al Solr pieces are already there.

There is a template needed by the playbook to complete:

http://svn.research-infrastructures.eu/public/d4science/gcube/branches/data-catalogue/ckan-d4science-extension/2.6/promoted/ckan-eosc.d4science.org-promoted.html

Could you prepare it, load on public svn and run the ansible playbook again?

There was problems during playbook exectuions, in particular the TASK [../../library/roles/ckan/ckan : Initialize the CKAN databases] has failed several times:

1. the table 'user' did not exist.. so I performed db init manually:

'SELECT \"user\".password AS user_password, \"user\".id AS user_id, \"user\".name AS user_name, \"user\".openid AS user_openid, \"user\".fullname AS user_fullname, \"user\".email AS user_email, \"user\".apikey AS user_apikey, \"user\".created AS user_created, \"user\".reset_key AS user_reset_key, \"user\".about AS user_about, \"user\".activity_streams_email_notifications AS user_activity_streams_email_notifications, \"user\".sysadmin AS user_sysadmin, \"user\".state AS user_state \\nFROM \"user\" \\nWHERE \"user\".name = %(name_1)s ORDER BY \"user\".name \\n LIMIT %(param_1)s' {'name_1': u'ckan_connector', 'param_1': 1}", "psql:/tmp/QknYwn.sql:1: ERROR:  relation \"user\" does not exist", "LINE 1: update \"user\" set state = 'active' where name = 'ckan_connec...", "               ^", "psql:/tmp/QknYwn.sql:2: ERROR:  relation \"user\" does not exist", "LINE 1: select apikey from \"user\" where name = 'ckan_connector' and ...", "                           ^"], "stdout": "", "stdout_lines": []}

1. the user 'ckan_connector' did not exist.. so I created it manually:

"stdout": "User \"ckan_connector\" not found\nCreate new user: ckan_connector? [y/n]Exiting ...", "stdout_lines": ["User \"ckan_connector\" not found", "Create new user: ckan_connector? [y/n]Exiting ..."]}

After fixes, now the playbook RECAP is ok:

PLAY RECAP *******************************************************************************************************************************************************************************************************
ckan-eosc.d4science.org : ok=227 changed=28 unreachable=0 failed=0

The Ckan Eosc is available at: https://ckan-eosc.d4science.org/ (now I'm going to update configurations for ckan template)

@costantino.perciante@isti.cnr.it, please, add the configurations (GR/RR) needed to run the catalogue in the VRE https://services.d4science.org/group/eosc_services asap, thanks.