IRIS

The inst.eecs.berkeley.edu web server is down with an apparent hardware failure. It is expected to be working again later today. Updates will follow as repairs progress.

Update:

The web server had a disk failure. We are re-installing it with a new disk. I expect it will be back up by 5pm.

There is an emergency patch released today from Microsoft. HERMES will be patched today, I will reboot at 4:00pm. Please be prepared.

IDSG recommends all Windows host apply this patch ASAP. Patches are available for Windows 2000, Server 2000, XP, Server 2003, Vista and Longhorn. x64 bit versions included.

Details, security bulletin and patch are available at: http://www.microsoft.com/technet/security/Bulletin/MS07-017.mspx

The patch is also available via Microsoft Update.

Mailing lists and the Jabber service will be briefly unavailable sometime between 5:00p.m. and 5:30p.m. on Wednesday, April 11, 2007.

The machine hosting these services will be having its operating system patched and will need to be rebooted. Unavailability shouldn’t last more than a few minutes.

Microsoft will release their patches for the month on Tuesday, April 10th.

HERMES, PRINT, RIS and WINSWW will be down for patching Wednesday, April 11th, from 3:00p–5:00p.

macserver.eecs.berkeley.edu, which hosts the Mac software warehouse, will be down for maintenance at this time as well.

Visual Studio 2005 is now available on winsww.eecs.berkeley.edu. To install you must have administrative privileges on your host. Go to \winsww\sww\public\visual_studio\VS2005

To install Visual Studio run the VS.setup.exe alias.
To install MSDN libraries run the MSDN.setup.exe alias.

The IRIS website has been redesigned!

The new, standards-based design brings with it a clean, usable layout, updated branding, and a nice foundation on which we plan to build some new features.

We hope you like it.

If you have any feedback you’d care to share with us, please e-mail img@eecs.

There are still some wireless problems in Soda 5th 6th and 7th floors. We are working on the solution.
[Read more…] about Network problems in Soda

We were hit today with a massive failure of events that caused a major network outage this afternoon. I sincerely regret the inconvenience that I know you all felt. Most services were back up by about 7PM this evening.

Here are the details for those who want specifics:

It appears that our core router had a serious failure. We attempted to switch to the back-up management module. However, that too failed. No amount of poking and prodding would bring either of them back. We then attempted to put in our hot spare. However, that unit began smoking! At this point we did not dare install anything in the chassis and we decided that we needed to replace the entire unit. This is no easy task as it is large, very heavy and has many fiber and copper cables connected to it. Once we swapped the spare chassis and all the cards and cables, a seemingly new problem arose, our firewall cluster started misbehaving immediately followed by a failure of one of the Gigabit Fiber I/O cards(the last one is what I now believe led to the whole business). We managed to stabilize the firewall and swap the last blade and slowly, networks services have come back. Many thanks to all of those who helped! It truly was a team effort. Whew, this has been a day that I will not wish to live over again.

Thank you for your patience,
Fred

In order to diagnose some network problems in Soda, the main router may need to be rebooted. This will cause a few minutes of downtime.

Sorry for the inconvenience

argus.EECS.Berkeley.EDU the department’s general purpose UNIX
login server was down for several minutes during the morning of
Tuesday March 20, 2007 for some emergency hardware maintenance.
We apologize for the inconvenience. If argus has any further
problems, troubleshooting will be a high priority.
[Read more…] about argus.eecs.berkeley.edu down for maintenance