Nomad Exec get disconnected when running adhoc long running scripts


We are facing some issues when running nomad exec.
We find some disconnection when using nomad exec to docker containers after a few time.
For example we need to run some adhoc post deployment scripts, a few times the script takes 30-60 mins. But the exec disconnects. This is one case when we run exec from local laptop.
Another case is one nomad batch job(shell script) doing exec to postgres container running within same nomad cluster to run backup, also see their disconnection once in a 10-15 times.

Is nomad exec not reliable enough ? Or there is some setting which we must change.

We do exec using this command nomad alloc exec -e none $ALLOCATION_ID /bin/bash

Running nomad 0.12.7 version.

Hi @surajthakur,

Apologies this is causing problems. I believe what you are seeing is being tracked within this issue on the Nomad repository. It is currently being investigated by the team.

jrasell and the Nomad team

Thanks for your response @jrasell. I will monitor that issue.