Posted on

NIRD and Service Platform downtime – 26.06.2019

  • 2019-06-26 14:05: Vendor is meticulously checking each NIRD storage component and decided to replace main controller chassis.
    In the mean time we are applying firmware updates on the Service Platform to improve stability and security.
  • 2019-06-26 08:15: Maintenance has started.

Dear NIRD and Service Platform User,

We have a planned downtime on the 26th of June, Wednesday next week, to replace some defective hardware. Systems will taken offline starting from 08:00AM.

Engineer from storage vendor will assist us from the very first hour.

We expect the maintenance to finish in one day.
Will keep you updated here.

Metacenter Operations

Posted on

Fram MDS patched

Dear Fram User,

This morning around 09:05, once again has the Fram metadata server crashed and likely had impact on running jobs.

A mitigating patch was delivered by the vendor yesterday and we used this opportunity to apply it on our metadata servers.

We will keep the system closely monitored and cooperate with the vendor on further stabilizing the system.

Apologies for any inconvenience this may have caused!

Posted on

Fram MDS crashed

Dear Fram User,

Once again has the Fram metadata server crashed and likely had impact on running jobs.
We are in contact with the storage vendor for patching the file system.

Apologies for the inconvenience!

Posted on

Clean-up of Fram filesystem needed

Dear Fram users,

The Fram filesystem, and most critically /cluster/work and /cluster/home, is running out of inodes, and there is only 8% left. If we run out of inodes it will not be possible to create new files. To avoid loss of data and job crashes we kindly ask all of you to if possible delete files that you no longer need.

Best regards,

Jon, on behalf of the Operations Team

Posted on

Missing mounts for Project NS2345K on NIRD

Missing mounts for Project NS2345K on NIRD
The NFS server which is exporting /tos-project1/NS2345K/FRAM to NIRD/FRAM has crashed yesterday around 16:00. We have recovered the NFS server and /tos-project1/NS2345K/FRAM is re-exported agian. We are currently working on mounting the filesystem in login containers and meanwhile investigating the cause. We are sorry for the inconveniences caused.

Posted on

Scheduled downtime – NIRD storage expansion – 2nd of April

Update:

  • 2019-04-03 13:25: NIRD and the service platform are back into production.
  • 2019-04-03 10:59: Maintenance work has finished. We are proceeding in starting back the filesystems and services.
  • 2019-04-03 08:22: Disk expansion and rebalancing is finished. HW checks are currently ongoing and shall finish in a couple of hours. Will keep you posted.
  • 2019-04-02 09:55: NIRD filesystems are unmounted from Fram and replicated data is available read-only trough login-trd.nird.sigma2.no
  • 2019-04-02 08:06: Maintenance work has started.

Dear NIRD User,

NIRD and the Service Platform will be under maintenance to expand the disk capacity in Tromsø.

The operations for storage expansion and disk pool rebalancing will start on the 2nd of April at 8:00 am CET and will last for maximum 2 days. During the maintenance, the services running on the NIRD Service Platform and on the NIRD Toolkit will not be available.

During the downtime we plan to make project data mirrored to Trondheim available in read-only mode trough a specially built login node. This solution will be first tested with real load during this downtime, thus we might encounter some technical difficulties.
That being said, to access the remote, mirrored data, please login to login-trd.nird.sigma2.no.

We apologise for the inconvenience.
Metacenter Operations

Posted on

MDS crash 13.03.2019

Main MDS server for lustre filesystem crashed between 14:00 and 14:30, And secondary MDS server took over and restored filesystem around 14:40 . Some of the jobs running on Fram might be affected. We are investigating the root cause of the incident.

Posted on

Short outage on Service Platform

Servers running the Service Platform and NIRD login nodes, have some issues with the remote filesystems.
The problem is already identified and being taken care of, but you might experience short hiccups until the problem is fixed on all the affected nodes.
You will have to re-login to NIRD login nodes.

We expect to be ready in maximum one hour.