Project Storage

Department Project storage

EECS Full Network and Service Outage, Thurs. 12/22/16 7:30AM

December 22, 2016 by Derek Calderon

We will have a full system shutdown on Thursday, December 22nd at 7:30 AM to perform necessary fire and emergency power testing in our server rooms. The entire EECS network and all hosted resources will be offline during this maintenance.

This will affect networking in the following locations:

Wired and wireless networks offline:
* Soda Hall
* Cory Hall
* Sutardja Dai Hall
* Jacobs Hall
* BWRC

Wireless networks offline, wired remains online:
* Blum Hall
* Calvin Lab

Hosted resources such as the EECS website, file servers, etc. will be offline and globally inaccessible.

The facilities maintenance window is scheduled from 7:30 AM to 1 PM, but we anticipate the network and critical services should be back by 11AM if all goes well.

The intent is to perform these tests annually going forward; however we will learn from this year’s procedure and aim to keep more critical services online in the future.

If you have any questions about this maintenance, please contact help@eecs.berkeley.edu.
[Read more…] about EECS Full Network and Service Outage, Thurs. 12/22/16 7:30AM

Fileserver maintenance, Nov 24 10am

November 24, 2016 by Pathma Venasithamby

On the morning of Thursday, November 24 (Thanksgiving Day), we will be performing some disruptive maintenance on the department fileserver:
* Recable to new network datacenter switches.
* Upgrade cluster switches firmware.
* Upgrade NetApp OS.

Department fileservice will be unavailable during the maintenance window, approximately 10am until 1pm. Many services that rely on access to department fileservice (home directories, webservice, repo service, etc), will either be offline or will fail to work during this window. All NFS and CIFS mounts will be affected.

We expect most, if not all, clients to recover automatically once the fileservers are back online. This news item will be updated to indicate when the maintenance is completed.

Systems still experiencing mount issues after the maintenance window, would need to remount or reboot.

Storage maintenance upgrades 22aug16

August 26, 2016 by Pathma Venasithamby

Moving network cables to new network fabric, and maintenance upgrades of firmware. Expected outage of an hour.

Postponed to Thanksgiving timeframe 2016
[Read more…] about Storage maintenance upgrades 22aug16

Scheduled Storage Outage, Fri. 3/11/16 10PM

March 11, 2016 by Derek Calderon

This Friday at 10PM we will be doing a software upgrade on the switch which provides connectivity for the EECS storage infrastructure. This upgrade is to address an issue with in-band management of the switch.

All storage services (home directories, project storage, etc.) will be briefly unavailable during the upgrade. Past experience suggests that this downtime will only last approximately five minutes. We strongly recommend that you save and close all open files in home and project spaces before 10PM on Friday. Machines with permanent NFS mounts may need to be rebooted after the maintenance is complete.

Please contact help@eecs.berkeley.edu with any questions or concerns.
[Read more…] about Scheduled Storage Outage, Fri. 3/11/16 10PM

Project storage migration

February 15, 2016 by Lars Rohrbach

As
previously announced, we are in process of migrating our central file storage
system to a new cluster of Network Appliance hardware.

Beginning February 15 (President’s Day), we will be migrating project and cluster
storage spaces to the new hardware. Initial synchronizations will be done over the
next two weeks, so that the final sync and cutover time will be reduced for the larger project spaces.
We are initially proposing the following schedule:

Time                    Project            New mount
Feb 15 6:00-6:30pm      /cproj/nlp         proj40.eecs:/nlp
Feb 15 6:15-7:00pm      /proj/ptolemy      proj40.eecs:/ptolemy
Feb 15 6:45-7:30pm      /proj/parlab       proj40.eecs:/parlab
Feb 15 6:45-7:30pm      /proj/radlab       proj30.eecs:/proj30/radlab
Feb 15 7:15-8:00pm      /proj/efros        proj10.eecs:/efros
Feb 15 9:00pm-7:00am    /cproj/(others)    proj40.eecs:/cproj/(sharename)



Feb 16 8pm-7am          /proj/[a-c]*       proj10.eecs:/proj10/(sharename)



Feb 17 8pm-7am          /proj/[d-q]*       proj20.eecs:/proj20/(sharename)



Feb 18 8pm-7am          /proj/[r-z]*       proj30.eecs:/proj30/(sharename)



Feb 19 8pm-7am          Followup/postponed and missed migrations

If migration of your project space needs to be postponed or scheduled for
a specific time, please
let us know at help@eecs. We know that downtime for
some project spaces needs to be minimized, and
we’ll do our best to coordinate another time with you.

As with the home directories, NFS exports will change. Automount maps will be updated at the designated migration start time. Re-mounts of project spaces will be necessary.

CIFS mounts should use the new node hostnames for best performance, with the
usual sharename, e.g. \\\\proj20.eecs\macsww

NetApp filer upgrade on December 1

December 1, 2015 by Lars Rohrbach

As part of our preparation for migration to new NetApp file storage hardware, we will be upgrading the OS on our existing Network Appliance file servers at 5pm Tuesday, December 1.

The OS will be upgraded from Data ONTAP version 8.1.2 to 8.2.4. Upgrade will be performed for project.eecs and home.eecs first, and then for cluster.eecs and inst-fs.eecs.

No general service interruption is expected, as our servers are configured in High-Availability pairs. Windows/SMB/CIFS users may experience a hiccup, if they have an active file transfer at the time of the upgrade; such a file transfer would terminate and need to be restarted.

Work is expected to be complete by 6:30pm.
[Read more…] about NetApp filer upgrade on December 1

Potential loss of data on Archive project space

November 3, 2015 by Lars Rohrbach

We are investigating potential loss of data on archival project space, and examining options for restoration.

Archival project spaces include the following:

/arch/carmena
/arch/visa
/arch/simons-video
/arch/yosef-archive
/arch/secml
/arch/inst-archives

Fileservice from cluster.eecs malfunctioning

January 3, 2015 by Lars Rohrbach

After restoring other services following the network upgrade, file service from cluster.eecs and inst-fs.eecs to Soda and Cory clients continues to be severely degraded. This affects anyone who has project shares hosted on cluster.eecs, and servers that mount project shares from cluster.eecs.

Staff have been working with assistance from NetApp support to restore service for cluster.eecs.

We are currently working on various alternatives to make the filesystems accessible again. More updates will be provided on Monday.

cluster.eecs fileservice unresponsive

November 11, 2014 by Lars Rohrbach

We are seeing a high load on cluster.eecs, resulting in NFS fileservice requests being delayed or failing. This affects all project shares hosted on cluster.eecs, and any services dependent on files there. Notably, this includes many research group websites.

Staff are investigating.
[Read more…] about cluster.eecs fileservice unresponsive

UPS Maintenance in SDH 158

August 6, 2014 by Derek Calderon

At 11 AM today we will be replacing batteries and performing maintenance on a faulty UPS located in our primary server and network room for Sutardja Dai Hall. No service interruptions are expected. However, in the event of a failure, there may be a ~5 minute loss of wired and wireless networking in Sutardja Dai Hall, as well as a loss of the NetApp network file storage for the EECS Instructional Labs. The maintenance is expected to last for approximately one hour, but two have been allotted in case of complication.
[Read more…] about UPS Maintenance in SDH 158