While we (SRE) made some strides improving stat host experience in T373446 , we have not addressed resource contention. As a result, a runaway user-created process can still hang the machine.
Creating this ticket to link individual tasks which advance the goal of stabilizing the stat hosts.
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Open | None | T393180 Stabilize stat hosts umbrella task | |||
| Open | None | T393177 Ensure stat hosts' SSH is responsive regardless of user-generated load | |||
| Resolved | BTullis | T393181 Ensure --master=yarn is default in stat host users' spark configs | |||
| Declined | None | T410769 Stat hosts: evaluate effectiveness of oomd |