The internet connection was down from 15:30 to 17:45 due to a routing loop created by our service provider, ELI. Apparently, some old routes were inadvertantly loaded at the Tukwilla office during a routine upgrade.
Category Archives: Failures
All services down for power improvements
All computer services were down from 7:00 to 18:00 for electrical improvements in the data center, and for a campuswide electrical maintenance event.
CPU replacement on Baker
To address more frequent crashes on Baker, HP support replaced its CPU.
Continue reading
Baker Hung
Baker hung, but apparently did not crash. System was rebooted.
Continue reading
Baker crashed
Baker crashed, with an error in PALCode, indicating either a memory or CPU hardware error.
Best Access System Hung
The Best Access license server failed today. No Best application could communicate with it.
KETRON, the computer that hosts Best Access services, also could not contact www.ups.edu or microsoft.com via web browser.
KETRON was rebooted
Webboard Failed, was rebooted
Webboard failed to respond. The server was rebooted.
Trimble Fiber Fails
Security reported Saturday night that the connectivity to Trimble Hall had failed. Several of the residents also reported that the Internet connection had been lost.
Trouble shooting revealed that one of the two fiber strands between McIntyre and Trimble had failed. The service was moved to another pair of fibers and service was restored about 9:30 PM Saturday, April 26th.
Account Management in ACADEMIA Fails IV – Problem Fixed
SCYLD was promoted to domain controller in ACADEMIA. The procedure went very smoothly. About one hour after promotion, the RID attributes on the primary controller were updated, and the problem was cleared.
Continue reading
Account Management in ACADEMIA Fails III
We attempted to reset RID attributes in Active Direcory for the ACADEMIA domain with the Java-based LDAP browser. Unfortunately, these attributes are read-only and can only be modified by the SYSTEM account.
We decided that we should bring another domain controller online in order to transfer or sieze RID master operations. It was my hope that the presence of another domain controller would force the RID master to issue another pool of RIDs, and clear the problem.
The only real candidate for promotion to domain controller was SCYLD. I obtained permission from Math/CS to promote SCYLD.
Account Management in ACADEMIA Fails II
Josh made me aware of the problem. Jeremy found that the problem is due to the RID Operations Master reaching the end of its allocation pool. We tried deleting over 100 accounts. This did not fix the problem.
Continue reading
Account Management in ACADEMIA Fails I
Account creations in the ACADEMIA domain began failing. Attempts to create accounts or add computers to the domain hung the operations master for the domain. This caused slow roaming profile loading across the domain. It became necessary to reboot the operation masters.
FENWICK Down
Fenwick began refusing telnet sessions. When it was rebooted, it failed to boot properly. Upon booting into minimal mode, it was determined that the system disk was full. The OPERATOR.LOG file was deleted, and Fenwick was rebooted, the system booted normally, and telnet sessions were no loger refused.
Webboard Crashed
Webboard crashed yet again. System was restarted at 7:30 3/25/03.
Merlin2 IO System Problem
On Friday, March 14th the “warning” indicator on Merlin2 was flashing, and the third drive was also flashing yellow. As we were in the middle of the data move from Merlin to Merlin2, we decided to push onward and deal with the problem later.
Today, the OS went BSOD. The server was shut down, then powered back up. The server came back up without any data loss and automatically started rebuilding the array onto the previously failed disk.
Continue reading