Monthly Archives: April 2009

Our east coast data center, Equinix, has a planned maintenance for a redundant backbone link on 05/02/2009 at 01:00 to 05:30 (GMT -5). All customers homed to this location will not experience any downtime. Brief period of latency is possible due to routing re-convergance.

On Friday, May 1st, at 23:59 Pacific Time, we will be rebooting EY02. We expect the cluster to be fully rebooted after no more then 4 hours, by Saturday, May 2nd at 03:59 Pacific Time. Sites will be unavailable during the event. If you have a maintenance page installed into the Load Balancer, that will be displayed for the duration.

We will be installing the following upgrades:

1) PXE Booting: This will allow us to more rapidly add, remove and swap nodes with less failure. We consider this a medium priority upgrade. Testing has shown that the PXE boot system is far more robust then the current system.

2) Updated CoRAID drivers: This is the first step in upgrading all of the shelves on the Cluster. We consider this a high priority upgrade, and is the first step in our plan to refurbish all CoRAID shelves.

This is a low risk, scripted maintenance. We expect all sites to come back with out issue. Please do not hesitate to fully check your site after the maintenance window, and if you detect any problems, open a ticket highlighting the issue. We will have extra staff online to handle any issues which may arise. We also will be monitoring sites via SiteUpTime and Nagios, and will respond to any issues our monitoring detects.

All was well and quiet. The clusters were in good form and we experienced no failures. Thanks!

The push was done with no incident.

We are pushing changes to the ey00 and ey04 gateways right now.

The IPMI devices were successfully rebooted. No Engine Yard services or customers were impacted.

We are proceeding with rebooting IPMI devices for one load balancer and nodes 0, 6 and 8 on EY02.

The EY03 SNAT update has been completed without service interruption.

We are proceeding with the scheduled SNAT update on EY03 at this time.

All was well and quiet. The clusters were in good form and we experienced no failures. Thanks!