Alamo nimbus partition outage
Submitted by David Gignac on Mon, 12 Mar 2012, 17:23:53 GMT
FutureGrid Hardware Outage Information
Alamo nimbus partition outage
- Status
- Resolved
- Type
- Software System, Network
- Impacted systems
- alamo
- Start of outage
- Mon, 12 Mar 2012, 12:00 EDT
- Anticipated end of outage
- Thu, 15 Mar 2012, 17:30 EDT
Description
nimbus jobs are causing a Gig E switch to hang. Disabling nimbus partition until I can get the Nimbus VM traffic off of the uplink.
Resolution
Initially we retired an older switch that was throwing errors. We also had an issue with nimbus VM's not shutting down correctly. The worksp were being re-used, so duplicate mac addresses were being used on the new switch. Nimbus partition was put back in production last night.