Version: [find the version by …]
I am curious if there is documentation I am missing that talks about the mechanics of how the compute nodes join the cluster.
We had to make some edits to the cloudformation so that our login nodes would join active directory on boot. All seems to work fine until I run Qrsh. That’s when I’m met with the "not able to run in any ques’ error. The command nodeattr -n nodes shows the compute nodes, I can SSH into compute nodes from master nodes, and qhost -q shows available ques. qhost -F shoes the hosts as well. It seems as though something is failing when the compute nodes join the cluster. Almost like they are joining the cluster but the login node isn’t aware of the ques available once they do. If I could learn more about the steps that are taken when the compute nodes join a cluster as they launch I’m sure I could solve it. Any info would be appreciated.