[Sigma-users] Sigma compute nodes down

[Mats Kronberg]

Dear Sigma users,

Due to a partial power failure (one of three power subcentrals powering
Tetralith and Sigma), most compute nodes in Sigma (all 104 "thin" nodes)
went offline at 05:10 today.

All running jobs on these nodes failed.

Until power is restored to these compute nodes, only the four "fat"
compute nodes and the Sigma login node are available.

Estimated downtime: 4h (best case - after a quick inspection we press
the reset switch on the subcentral and boot the nodes) to 1-2 days
(something is damaged and we need to make repairs or move some compute
nodes from Tetralith to Sigma).

--
Mats Kronberg, NSC Support <support at nsc.liu.se>