Crunch2 installation » History » Revision 2
Revision 1 (Tom Clegg, 06/14/2016 06:01 PM) → Revision 2/18 (Tom Clegg, 06/15/2016 09:16 PM)
h1. Crunch2 installation (DRAFT -- when ready, this will move to doc.arvados.org→install) {{toc}} h2. Set up a Pieces: * crunch-dispatch service Currently, dispatching containers via SLURM is supported. Install crunch-dispatch-slurm on a node that can submit SLURM jobs. This can be the slurm controller node, a worker node, or any other node that has the appropriate SLURM/munge configuration. <pre><code class="shell"> sudo apt-get install crunch-dispatch-slurm * crunch-run </code></pre> Create a privileged token for use by the dispatcher. If you have multiple dispatch processes, you should give each one a different token. <pre><code class="shell"> * kernel config apiserver:~$ cd /var/www/arvados-api/current * docker config apiserver:/var/www/arvados-api/current$ sudo -u webserver-user RAILS_ENV=production bundle exec script/create_superuser_token.rb zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz </code></pre> Save the token on the dispatch node, in <code>/etc/sv/env/ARVADOS_API_TOKEN</code> Example runit script: <pre><code class="shell"> #!/bin/sh export ARVADOS_API_HOST=uuid_prefix.your.domain exec chpst -e ./env -u crunch crunch-dispatch-slurm </code></pre> Ensure the @crunch@ user exists -- and has the same UID, GID, and home directory -- on the dispatch node and all SLURM compute nodes. Ensure the @crunch@ user can run docker containers on SLURM compute nodes. h2. Install crunch-run on compute nodes h2. Configure kernel cgroup accounting on compute nodes h2. Configure docker h2. Test the dispatcher * sanity check