https://github.com/natefoo/usegalaxy-clone
Deploy a clone of usegalaxy.org
https://github.com/natefoo/usegalaxy-clone
Last synced: 3 months ago
JSON representation
Deploy a clone of usegalaxy.org
- Host: GitHub
- URL: https://github.com/natefoo/usegalaxy-clone
- Owner: natefoo
- Created: 2019-10-04T04:25:10.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2021-02-14T19:56:04.000Z (over 4 years ago)
- Last Synced: 2025-02-15T19:21:05.537Z (4 months ago)
- Language: HTML
- Size: 527 KB
- Stars: 0
- Watchers: 4
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
**For a new clone:**
You probably should change the Galaxy UID and GID - I discovered when trying to launch more nodes that (due to having
developed the playbook orginically for the first node) the Galaxy GID was already taken by Docker.
**Manual steps for each new instance:**
1. Set instance IP in inventory after spawn. Could use openstack inventory plugin but need to filter out other hosts.
2. Add nodes to slurm_nodes and partition in group_vars/all.yaml
**Manual steps for Galaxy server only:**
1. Set server private IP in galaxy_data_addr. Again, could use openstack inventory for this.
2. Copy munge key to files/slurm/munge.key and vault once created on controller, set slurm_munge_key. Could precreate this.
3. Set slurm_create_user and slurm_user once created on controller. Could also precreate.
4. Set galaxy_group and galaxy_user once created on Galaxy server. Could also precreate.
**Scaling up/down**
Server becomes an exec host when not being used for workshops. When transitioning between workshop (scale up) and
non-workshop (scale down), adjust slurm_roles in group_vars/galaxyservers/vars.yaml accordingly.
**TODO**
- Pull sanitize_whitelist.txt and integrated_tool_panel.xml from Github so they don't get out of date.
- PostgreSQL tuning was applied by hand from my phone ;P
- Node addition in the slurm config vars is tedious, can this be improved somehow?