About Connection Issue on Clusters

[Onur Pulaşlı]

Dear HPC System Users

   Owing to the work on clusters that we did on the weekend, Yunus and Somon clusters are up and steady again. You my run your jobs on clusters. Right now there is a running process on Lufer which is about all the home folder's transmission to the new cluster. Because of the disk fault , there is major performance issue on Lufer. To reduce this, we switched off the accessibility on Lufer cluster until the data transmission is over. We will announce immediately when the cluster gets back.

   

   On the headnode of Slurm cluster (which is Biyofiz 4.0) , also there is disk fault on the one of the disk. When we seperate the disk and the node , we noticed that cluster is working properly. Right now we could be able to reach the disk and we are covering the data on the faulty disk by transmitting the data to the proper disk. We will open up the cluster when the transmission is done and virtual machines (spacewalk and prov) are up again. We assume that Slurm cluster will be up tonight. Also we started working on the faulty disk.

For your kind attention,

Onur PULAŞLI

Information Technologies