Hi there,
This morning, I found several nodes of our cluster unreachable by the queueing system, i.e. their status became "au" or "E" (pls see the attached pic.). Indeed, I can access them by ssh, e.g., "ssh compute-0-10", but it seems that the queueing system cannot find them.
Any idea about why this happens and any possible solution?
Thanks a lot.
Mohamed Talaat Harb
au status usually means that the grid engine isn't running on the node as the "a" means alarm and the "u" means unreachable you will need to check for the sgeclient service on the nodes having the problem. also the cause for this may be a crash of the service on this nodes and restarting the service or the whole node may be the solution.
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Control Theory
- Control Systems
More Tongchuan Suo's questions See All
Similar questions and discussions