Cluster Controller at p55cc1

Titan & Pleione reboots are more or less completed now.

However, not all nodes were upgraded successfully, so there will be another reboot in near future. Maybe some time next week.

At that moment all clusters (titan, pleione and dione) will be rebooted.

Titan will be rebooted as soon as the queues are drained.

Pleione is being rebooted at the moment.

Titan is now ready to use.

Note that partition 'all' has vanished. Use partition 'normal' instead.

Note also that partition 'big' is a separate node and is not part of partition 'normal'.

Hence, in partition 'normal' there are nodes ti2 - ti17 and in partition 'big' there is only one node ti1.

Node ti2 is still in draining state and it will be rebooted as soon as the grid job completes. Likewise, with node ti99.

Titan reboot finished.

Pleione reboot finished.

Pleione is being rebooted now due to kernel upgrade. Titan will be rebooted in few days.

Pleione compute node 5 is fixed now and all compute nodes are online.

The OpemMPI implementation in Pleione is not standard so let me know if you encounter any problems.

Titan login problem. The /home mount has disappeared.

MBO failure on Pleione compute node 5. Fixing takes few days.