Getting Started

From Grid5000
Revision as of 21:22, 22 October 2012 by Lnussbaum (talk | contribs) (New page: {{Warning|text=Work in progress}} == Firsts steps == # G5K general goals and admin context + orga + get support # ssh access.grid5000.fr ## you are now connected to Grid'5000 from here yo...)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
Warning.png Warning

Work in progress

Firsts steps

  1. G5K general goals and admin context + orga + get support
  2. ssh access.grid5000.fr
    1. you are now connected to Grid'5000 from here you can
      1. see the contents of your homedir on each site (these are different, and not backed-up)
      2. connect to a site
  3. ssh-keygen + installer les clés sur les sites. See the FAQ on how to allow direct access from your machine to a site.
  4. connect to a site
  5. run oarsub -I -> you are now connected to a node, reserved for your usage for an hour
    1. exit -> you are now disconnected
    2. oarsub 'sleep 3600'  , oarsub -C <jobid>
  6. visualize resources: MOTD, Gantt, Monika, API selector
    1. complex oarsub (-p cluster, -l nodes/walltime, -r)
    2. list resources, connect to them using oarsh, or use -t classic-ssh if you want to keep to pure ssh (pointeur vers advanced)
  7. your software environment on nodes is the production environment, you can use for MPI, JAVA applications or to start VMs using KVM. You might want more control (to be root)
  8. oarsub -t deploy
  9. kadeploy3 
  10. installer un paquet
    1. http_proxy (whitelisting)
    2. save for furture usage or script installation
  11. Kaconsole et Kareboot ou kapower3
  12. parler de g5k-checks, du focus XP de G5K, expliquer que g5k-checks compare ses données avec l'API de réference, et donc introduire l'API