[Berzelius-users] Unplanned login-node downtime

[Henrik Henriksson]

Dear Berzelius users,

Due to a currently undiagnosed issue, we lost the login node `berzelius001`
earlier this afternoon. All active sessions on that server are lost. Jobs and
job scheduling are unaffected.

While connections to `berzelius.nsc.liu.se` should work normally, users
connecting directly to `berzelius1.nsc.liu.se` will experience issues. Please
use "berzelius.nsc.liu.se" to connect to Berzelius via SSH/Thinlinc until the
issue can be diagnosed and resolved, or connect directly to
`berzelius2.nsc.liu.se`.

We do not expect to resolve the issue before the weekend.

-- 
Henrik Henriksson
Systems Administrator
National Supercomputer Centre