Skip to content

News from the staff team

  • 1 min read

Unhappy Web Server

As many of you have noticed, our web server has been experiencing lots of uptime issues. We're not exactly sure what's causing the problem, but we think it may be related to NFS and some prior issues we've had with the Solaris install on the machine. For now, it seems that rebooting the web server fixes the problem for a few hours, so, if you notice any issues with our web services, contact staff and we'll do something.

In regards to a long term solution to these problems, we are planning an overhaul of the Solaris install on the machine (read: re-install of the operating system). Unfortunately, this process is complicated by the fact that we have many other services hosted on the machine (in-lab printing, databases, etc), so check here for status updates regarding our repair plans.

  • 1 min read

New Linux Login Server

The OCF has a new Linux login server thanks to a recent hardware donation. This server is configured with the same software as our AMD64 lab machines but is secured in our server room, so it should be more reliable and less-prone to random reboots.

Just log into tsunami.ocf.berkeley.edu with your OCF account to use this new server.

  • 1 min read

Home Directories Restored

All home directories that were missing after the disk array failure have been restored.

  • 1 min read

PHP5 Beta Test

PHP5 has been setup on our web server. Any files with a '.php5' extension should be run with the PHP5 interpreter.

If you notice any bugs, please contact staff or post a comment.

  • 1 min read

Happy Thanksgiving!

We're hoping you're having a nice breather from classes and spending time with your loved ones.

  • 1 min read

Temporary Array Fixed

The lockup issues we were having with our temporary disk array have been resolved after a BIOS and firmware update were applied to our SAS HBA.

Also, please do not reboot our Linux computers, even if they are appear to be hung-up. Numerous reboots damage the hardware (especially the hard drive and power supply), and since those items are costly to replace, we end up having to take the machines offline for awhile.

  • 1 min read

Accounts Being Created

Account creation has been restarted. If you requested an account within the past month, it should have been created last night.

  • 1 min read

Bumpy Services

I apologize for the lack of status updates in the past few days. A lot of things have been occurring inside and outside of the OCF, so I haven't been able to dedicate a large quantity of my time towards updating this blog. Here's a quick summary of our disk array migration:

  • The downtime was much longer than we had anticipated. An OCF staffer was injured during an OCF outing over the extended weekend that the migration was supposed to occur, so our staff was divided and coordination was difficult.
  • Access control lists have not been migrated. Our new disk array uses NFSv4 ACLs, and our old disk array uses NFSv3 ACLs. We're working on coming up with a way to migrate them over.
  • Mail services a little bumpy. We're getting NFS issues, and we're working on debugging them. Mail spam filtering is still in a state of limbo.
  • There continues to be intermittent issues with our Linux lab machines (ex., random hangs). We presume these issues are due to the same NFS problems we're having on our mail server.

Now that we've transferred our services off of our primary disk array, I'm going to be working on restoring an old image of our disk array so that we can restore files that were overlooked during our disk array failure last month.

Thanks for your patience.

  • 1 min read

Disk Array Migration Ongoing

We're currently migrating the OCF off our current disk array onto a temporary disk array to allow our regular disk array to be serviced. According to the OCF announcement that was sent out yesterday, we expect downtime to end at approximately 8 AM Saturday morning (November 11).

  • 1 min read

Status Update

Files are still being copied, it's raining in Berkeley, and we might slip past our target deadline of 8 AM by a few hours.

I'll post more information as it comes.