I am using slurm version 24.05.1 , my slurmd has a single node where I connect multiple gres by enabling the overscribe feature.

I am able to use the advance reservation of gres only using gres name (tres=gres/gpu:SYSTEM12).

i.e while in reservation period , if other users submits job with SYSTEM12 , then slurm places this job in queue 

but when other users just submit a job without any system  name , slurm jobs goes through on that gres immediately even though it is reserved.

mylinux.wbi.com                                                                                         

Also I can see GresUsed in busy mode using "scontrol show node -d"   , this means the job is running on Gres/GPU and not on cpu etc.                                                                                                               

Same way , job submission based on Feature "rev1 in my case" is also going through even though it is reserved for other users in multiple partition slurm.

while execution , Slurm o/p's some extra prints in the srun output

user1@host$ srun --gres=gpu:1 hostname                                                                                                                                                                                      

srun: error: extract_net_cred: net_cred not provided                                               

srun: error: Malformed RPC of type RESPONSE_NODE_ALIAS_ADDRS(3017) received                                                                                                                  srun: error: slurm_unpack_received_msg: [mylinux.wbi.com]:41242] Header lengths are longer than data received 

Regards,

MS