The Novell server “ST” was rebooted this morning due to a stuck process. This process caused the CPU to constantly stay at around 30%utilization.
The backup had hung because of this problem.
In trying to figure out what/why I noticed that a users connection could not be disconnected.
So are we chasing a server or client problem?
The main disk partition had run out of space earlier this week. So a reboot was needed to ensure drive integrity.
Reboot to load new drivers (ce1000).
The server will be rebooted and new drivers installed.
CE1000 for the nic (latest I could find)
pedge3.ham for the raid card (from sp6)
After the initial reboot. A second reboot was necessary.
This is a recurrence of the problem from Saturday night. We believe it is caused by old Novell clients. Information System will be pushing out an update.
About 9:00pm some students began to experience inconsistent responses from the Novell server (i.e. accessing the P: drive). Info Systems began investigating the problem about 9:40pm and by 10:40pm it was determined that some type of process was consuming excessive CPU cycles and that reboot would be required to stabilize the ST server.
The ST Novell server was rebooted with some unexplained error messages but the reboot seemed to be successful. There is the possibility that some files are in a corrupted state. If students receive errors while accessing any data on their P: drive indicating a corrupt file exists they should notify the Helpdesk, giving the particulars of their Novell userid, the text of the error message and the name of file and the directory path where it exists on the P: drive.
If corrupt data is found the ST server will need to be taken off-line on Monday, 12/8, to run a repair utility.
If students encounter Novell network file problems before 8:00am Monday, 12/8, and they are convinced that Info Systems should consider bringing a technician to campus to investigate during weekend hours, please call the Info Systems emergency number (from an on-campus phone, x4541).
Info Systems will take ALL Novell servers off-line early Saturday morning, 12/6, to make system images to enable recovery from any future catastrophic failures. Details:
DATE: Sat, December 6, 2003
BEG-TIME: About 7:00am
END-TIME: About 8:30am
IMPACT: All Novell services, including network files (P:, G:, Z: drives), network printing will NOT be available during the outage period.
Also, because the EMU email server and Oracle calendar server login authentications are provided by the Novell directory service, access to the EMU email services, including WebMail, will NOT be available during the outage period.
Users WILL be able to access the Internet, Blackboard, CampusWeb and the Sadie (library catalog).
All Novell servers, except FSAPPS, should be returned to service by 8:30am. FSAPPS will remain off-line while it hardware for it is upgraded. See FSAPPS details described in a separate outage report for 12/06/03.
————————————–
All Novell servers used by students, faculty and staff (except FSAPPS) were returned to service at 8:25am, 12/06/03. The system images are now complete and up-to-date.