Reported Outages

OpenStack KVM unavailable during data migration

Resolved Posted by Pierre Riteau on September 23, 2015
Outage start Monday, September 28, 2015 2 a.m.
Expected end Monday, September 28, 2015 6 p.m.

On Monday, 28 September 2015, we will migrate all images, snapshots, volumes, and volume snapshots from the current OpenStack KVM deployment to the new Chameleon hardware.

Note that all instances will be stopped and that OpenStack KVM will be unavailable during this maintenance.

Network issues on CHI at TACC

Resolved Posted by Pierre Riteau on September 16, 2015
Outage start Tuesday, September 15, 2015 7 p.m.
Expected end Wednesday, September 16, 2015 5 p.m.

Update: The network issues at TACC have been resolved.

We are currently investingating network issues on the CHI cloud at TACC.

UC Chameleon down for electrical maintenance

Resolved Posted by Pierre Riteau on September 15, 2015
Outage start Tuesday, September 15, 2015 8:50 a.m.
Expected end Wednesday, September 16, 2015 3 p.m.

Update 2015-09-17 17:00 UTC: All storage nodes are now available, except one having hardware issues.

The UC Chameleon system is down for electrical work that will last approxiately 3 to 4 hours.

Scheduled network maintenance at UC

Resolved Posted by Pierre Riteau on July 17, 2015
Outage start Monday, July 20, 2015 6:30 a.m.
Expected end Monday, July 20, 2015 7:30 a.m.

A network maintenance is scheduled at the datacenter hosting the Chameleon UC resources. Connectivity will be disrupted during this maintenance planned for 6:30 AM to 7:30 AM on Monday, July 20, 2015.

Maintenance for UC network configuration changes

Resolved Posted by Pierre Riteau on July 14, 2015
Outage start Wednesday, July 15, 2015 2 p.m.
Expected end Friday, July 17, 2015 2:30 p.m.

A maintenance session is scheduled on Wednesday, July 15 at 2 PM Central Time to perform changes of the network configuration at the University of Chicago site.

Unreachable instances on FutureGrid@Chameleon

Resolved Posted by Pierre Riteau on July 01, 2015
Outage start Wednesday, July 01, 2015 10 a.m.
Expected end Wednesday, July 01, 2015 3 p.m.

Update at 21:20 UTC: All nodes are now back online.

Update at 20:40 UTC: The impacted nodes have been brought back online, except for one presenting a hardware failure. We are leaving the affected instances shut down: please start them again via the Horizon dashboard to regain access. If you don't need these instances anymore, please terminate them to release resources for other users.