Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Apr 28 18:13
    hexylena commented #128
  • Apr 28 18:12
    hexylena commented #128
  • Apr 28 18:11
    hexylena edited #128
  • Apr 27 10:44
    hexylena edited #128
  • Apr 27 07:54
    hexylena edited #128
  • Apr 26 14:32
    hexylena commented #128
  • Apr 26 14:18
    hexylena edited #128
  • Apr 26 14:18
    hexylena edited #128
  • Apr 26 14:18
    hexylena edited #128
  • Apr 26 14:18
    hexylena edited #128
  • Apr 26 14:18
    hexylena edited #128
  • Apr 26 14:18
    hexylena edited #128
  • Apr 20 18:33
    nekrut commented #113
  • Apr 20 18:27
    nekrut commented #110
  • Apr 20 16:57
    nekrut assigned #113
  • Apr 20 16:27
    hexylena commented #128
  • Apr 20 16:21
    hexylena edited #128
  • Apr 20 16:21
    hexylena edited #128
  • Apr 20 16:21
    hexylena edited #128
  • Apr 20 16:18
    hexylena commented #128
Lucille Delisle
@lldelisle
Can one of the instructor try to come and debug me?
Nate Coraor
@natefoo
Someone @kpoterlowicz just ran in to: If you try to run the full playbook we have developed on another host, the database grants for tiaas on specific Galaxy tables fails because this happens before Galaxy's database schema is created.
It worked during the training because our playbook developed organically - by the time we added the tiaas grants to the playbook, the Galaxy database was already created on a much earlier playbook run.
Here is a post I wrote about this very sort of problem during last year's admin training: https://nate.coraor.org/2019/01/29/galaxy-admin-training-ansible.html
Helena
@hexylena
someone -> something
Nate Coraor
@natefoo
miniconda_manage_dependencies: false
Helena
@hexylena
thx
Tom
@tomraulet
Hi there, I had to leave at lunch time, (train to catch) but I wanted to say thank you to our teachers ;) great team, great materials! I learned a lot of useful things.
I wish you (all of you, students and teachers) the best ;) bye!
Nate Coraor
@natefoo
:beers:
Helena
@hexylena
We're glad to hear it!
Nicola Soranzo
@nsoranzo
Thanks @tomraulet , safe travels!
Helena
@hexylena

If anyone wants to share their galaxy, you can register it as a public galaxy in the hub, here is an example. Just write this index and everyone will hear about it :)

https://github.com/galaxyproject/galaxy-hub/blob/master/src/use/argalaxy/index.md

Nate Coraor
@natefoo
For anyone who stuck around to the end that wants my noodly slide: https://docs.google.com/presentation/d/1ei3DjVuZDrKEZ_7xOfZTFMkIxKmG4TE6cl-sEIFG6xE/edit#slide=id.g3c75d95126_0_0 (also linked from the Q&A doc)
jacorvar
@jacorvar
for those planning to deploying PostgreSQL on a different host than Galaxy, this is what worked for me (it should be pasted on the group_vars/galaxypostgresql.yml file):
---
# Python 3 support
pip_virtualenv_command: /usr/bin/python3 -m virtualenv # usegalaxy_eu.certbot, usegalaxy_eu.tiaas2, galaxyproject.galaxy
certbot_virtualenv_package_name: python3-virtualenv    # usegalaxy_eu.certbot
pip_package: python3-pip                               # geerlingguy.pip

# PostgreSQL
postgresql_objects_users:
  - name: galaxy
    password: galaxy
postgresql_objects_databases:
  - name: galaxy
    owner: galaxy
postgresql_conf:
  - listen_addresses: "'*'"
postgresql_pg_hba_conf:
  - "host all galaxy Y.Y.Y.Y/32 password"
  - "host all galaxy X.X.X.X/32 password"
Gonzalo Vera
@gonzalovera
---
# Python 3 support
pip_virtualenv_command: /usr/bin/python3 -m virtualenv # usegalaxy_eu.certbot, usegalaxy_eu.tiaas2, galaxyproject.galaxy
certbot_virtualenv_package_name: python3-virtualenv    # usegalaxy_eu.certbot
pip_package: python3-pip                               # geerlingguy.pip

# PostgreSQL
postgresql_objects_users:
  - name: galaxy
    password: galaxy
postgresql_objects_databases:
  - name: galaxy
    owner: galaxy
postgresql_conf:
  - listen_addresses: "'*'"
postgresql_pg_hba_conf:
  - "host all galaxy Y.Y.Y.Y/32 password"
  - "host all galaxy X.X.X.X/32 password"
Thank you! That will be useful!
jacorvar
@jacorvar
@jacorvar You only need the IP for the Galaxy Server indeed. You'll probably have to restart postgresql specifying the version on the postgresql host.
Nate Coraor
@natefoo
For anyone who was trying things on their own hosts and had issues where PostgreSQL did not seem to recognize changes made by the role, see galaxyproject/ansible-postgresql#23 - for the training instances, for example, you have to restart PostgreSQL with sudo systemctl restart postgresql@10-main
Gonzalo Vera
@gonzalovera
As a follow-up of our conversation about having fun in BCN, for tonight, 3 suggestions
This message was deleted
Nate Coraor
@natefoo
@gonzalovera thanks!
Lucille Delisle
@lldelisle
Hi, it looks like on the VM of the training my mules are down (the worker "execute" but the mules do not react then all the jobs are grey 'new'). How can I restart them?
Nicola Soranzo
@nsoranzo
@lldelisle Did you try with sudo systemctl restart galaxy.service ?
Lucille Delisle
@lldelisle
yes of course...
at least 2-3 times...
Nate Coraor
@natefoo

Hi @lldelisle. I added two entries to your playbook to disable the suppression of high rate systemd-journald messages. One in pre_tasks:

    - name: Disable systemd-journald message suppression
      ini_file:
        path: /etc/systemd/journald.conf
        section: Journal
        option: RateLimitBurst
        value: 0
      notify: Restart systemd-journald

and one in handlers:

    - name: Restart systemd-journald
      service:
        name: systemd-journald
        state: restarted
After restarting Galaxy this revealed that the job handlers are failing to fully start because they try to read the embedded pulsar configuration, which is missing:
Mar 09 12:47:32 gat-1 uwsgi[28171]: Traceback (most recent call last):
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "lib/galaxy/main.py", line 299, in <module>
Mar 09 12:47:32 gat-1 uwsgi[28171]:     main()
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "lib/galaxy/main.py", line 295, in main
Mar 09 12:47:32 gat-1 uwsgi[28171]:     app_loop(args, log)
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "lib/galaxy/main.py", line 142, in app_loop
Mar 09 12:47:32 gat-1 uwsgi[28171]:     attach_to_pools=args.attach_to_pool,
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "lib/galaxy/main.py", line 108, in load_galaxy_app
Mar 09 12:47:32 gat-1 uwsgi[28171]:     **kwds
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "/srv/galaxy/server/lib/galaxy/app.py", line 211, in __init__
Mar 09 12:47:32 gat-1 uwsgi[28171]:     self.job_manager = manager.JobManager(self)
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "/srv/galaxy/server/lib/galaxy/jobs/manager.py", line 26, in __init__
Mar 09 12:47:32 gat-1 uwsgi[28171]:     self.job_handler = handler.JobHandler(app)
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "/srv/galaxy/server/lib/galaxy/jobs/handler.py", line 51, in __init__
Mar 09 12:47:32 gat-1 uwsgi[28171]:     self.dispatcher = DefaultJobDispatcher(app)
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "/srv/galaxy/server/lib/galaxy/jobs/handler.py", line 952, in __init__
Mar 09 12:47:32 gat-1 uwsgi[28171]:     self.job_runners = self.app.job_config.get_job_runner_plugins(self.app.config.server_name)
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "/srv/galaxy/server/lib/galaxy/jobs/__init__.py", line 784, in get_job_runner_plugins
Mar 09 12:47:32 gat-1 uwsgi[28171]:     rval[id] = runner_class(self.app, runner.get('workers', JobConfiguration.DEFAULT_NWORKERS), **runner.get('kwds', {}))
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "/srv/galaxy/server/lib/galaxy/jobs/runners/pulsar.py", line 201, in __init__
Mar 09 12:47:32 gat-1 uwsgi[28171]:     self.__init_client_manager()
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "/srv/galaxy/server/lib/galaxy/jobs/runners/pulsar.py", line 213, in __init_client_manager
Mar 09 12:47:32 gat-1 uwsgi[28171]:     self.__init_pulsar_app(pulsar_conf, pulsar_conf_file)
Mar 09 12:47:32 gat-1 uwsgi[28171]:   File "/srv/galaxy/server/lib/galaxy/jobs/runners/pulsar.py", line 239, in __init_pulsar_app
Mar 09 12:47:32 gat-1 uwsgi[28171]:     with open(pulsar_conf_path, "r") as f:
Mar 09 12:47:32 gat-1 uwsgi[28171]: FileNotFoundError: [Errno 2] No such file or directory: '/srv/galaxy/config/pulsar_app.yml'
Lucille Delisle
@lldelisle
Ah, thank you very much
Is there a way to send journalctl messages to a file so you cannot miss one.
Nate Coraor
@natefoo
You can use the uWSGI logto option to have it log to a file instead, or see the logging config docs for fancier options (such as splitting mule logging in to separate files)
Lucille Delisle
@lldelisle
thanks
Ondrej Melichar
@ondrejme
Hi, is it possible to enable/disable specific tools/sections for specific users/groups?
Ondrej Melichar
@ondrejme
Nicola Soranzo
@nsoranzo
:+1:
Helena
@hexylena
@ondrejme that just hides, doesn't disable. https://galaxyproject.org/admin/config/access-control/ is a bit more comprehensive on the topic
kpbioteam
@kpbioteam
Hi all, I'm looking for the best way to download multiple data libraries from local instance, any recommendations?
Helena
@hexylena
if it's a local instance you could poke through the database for the dataset IDs and resolve them to their directories. Or use the api with bioblend/parsec.
but nothing ready-to-go, I think you might have to write some code yourself :/
kpbioteam
@kpbioteam
Thank you! I will try with bioblend
Helena
@hexylena
This message was deleted
Tom
@tomraulet
Hi there, I hope you're fine. I'm playing around with Ansible last days, and I found geerlingguy wrote a book. It's free through April. More info here : https://www.ansiblefordevops.com/
Slugger70
@Slugger70
Cheers!