Requested interactive jobs not appearing in queue and not being fulfilled

Discovery Cluster Slurm
1

I have not been getting any of my interactive job requests fulfilled, and they don’t even show up in the queue. I submit a job as so:

(base) [sagendor@discovery2 runs]$ salloc --ntasks=1 --mem=0 --time=00:29:59 --partition=debug
salloc: Pending job allocation 694445
salloc: job 694445 queued and waiting for resources

and it just hangs, as if I were waiting in the queue. However, I go to different window and check my queue

(base) [sagendor@discovery2 ~]$ squeue -u sagendor
             JOBID PARTITION     NAME     USER ST       TIME  NODES NODELIST(REASON) 
(base) [sagendor@discovery2 ~]$ sinfo -p debug
PARTITION AVAIL  TIMELIMIT  NODES  STATE NODELIST 
debug        up      30:00      1  down* d10-02 
debug        up      30:00      5   idle e01-60,e05-[42,76,78,80] 
(base) [sagendor@discovery2 ~]$

and no job is there, and I’m not being granted a node. I used to request jobs like this all the time before and no issues, so I don’t know what has changed. This happening to me on both debug and main partitions.

2

I think this may have resulted from network issues we were experiencing yesterday. Could you try again and see if it works for you now?

3

I just tried and I’m seeing the exact same behavior. I was also having trouble getting nodes as early as Tuesday.

Running batch scripts seems to be OK though…

As an added peice of info, which shouldn’t matter, but I am running tmux/3.0a module

4

Looks like there is an issue with salloc on the discovery2 login node. Try logging in to discovery1 and using salloc there.

5

salloc is now fixed on discovery2.

6

Sorry for the delay in response, but yes it’s working now. Thanks