2-Slurm/2.1-Batch-job-submission.md

@@ -155,4 +155,15 @@ In the above example, the used cumulated CPU time of the *job_id* job is about 1
 Display the real resources used by the terminated jobs of *username* starting and ending between February 1, 2021 and March 1, 2021 on the bigmem partition:
 ```
 sacct -u username --partition=bigmem --starttime=2021-02-01 --endtime=2021-03-01 --format=JobID,State,TRESUsageInMax%100 | grep batch
+```
+## Connect on nodes during batch jobs
+
+Direct connection to compute nodes is forbidden in the Pyrene cluster. The only way to access resources allocated to a job is to submit another job. To do this, use the `srun` command to connect inside the targeted job. Assuming you have submitted a _job_id_ batch job:
+```
+$ sbatch script.sh
+Submitted batch job job_id
+```
+simply use the _job_id_ allocated by Slurm to connect to the corresponding node:
+```
+$ srun --pty --jobid job_id /bin/bash -i
 ```
\ No newline at end of file