Juliet unavailable

Submitted by Allan Streib (… on Fri, 07/17/2020 - 00:18
Status
Resolved
Description

The juliet login nodes are offline. We are looking into the problem, however it is not able to be resolved remotely so the anticipated end of the outage will be no sooner than the morning of July 17.

As the juliet login nodes are also used by romeo and tango clusters, those clusters are unavailable as well.

Start of Outage
Anticipated End of Outage
Resolution

New kernel was hanging on boot-up. Seemed related to Infiniband (mlx5_core driver).

Booted with prior kernel and OK for now. Will need to test future updates to see if resolved.

Type
Outage
Impacted Systems
Juliet
Romeo
Tango