[PA] Network Maintenance Advisory - 9/28/2008

[swadmin]

Network Engineering will be performing maintenance in our PA facilities during the hours of 6AM and 8AM EST on Sunday, September 28th, 2008.

Network Engineering will be performing a significant upgrade/replacement of our existing network infrastructure. This upgrade will involve removing/replacing our older/out-dated core Cisco™ GSR & Extreme Summit™ routing/switching equipment, with newer state-of-the-art Cisco™ 6500-series routing/switching equipment, and implementing a new network design---allowing for easier upgrade, higher fault-tolerance, additional capacity, and extra redundancy.

This maintenance will be service affecting, however, we expect the overall network downtime to be 15-30 minutes, during the 2-hour maintenance window...as equipment, cabling, and routes are changed. There will also be momentary periods of higher than normal latency while circuits are re-routed, BGP updated/reloaded, and cabling switched, during the 2-hour maintenance window. Proper planning, configuration, and testing has been completed in advance of this maintenance period, in order to minimize as much as possible the overall time that will be service affecting.

NOTE: Network Engineering will only be performing a small step in the overall network upgrade plans during this maintenance window. We will be moving BGP sessions with our carriers on to new core routers, and then linking them to the existing aggregation/distribution portion (which will not be modified during this maintenance window) of the network. This portion was completed successfully in the prior maintenance window on Sept 20th, prior to reverting back to the old configuration following an issue at the aggregation level (a step further down the upgrade path)--so this maintenance should go smoothly. Further maintenance will be required once the aggregation level equipment is upgraded/replaced, but that should be spread out over several weeks, with minimal downtime involved (...planned to be completed slowly on a distribution switch by switch basis, (...on to new aggregation equipment running simultaneously with the old)---not network-wide).

We thank you for your patience and understanding with this necessary maintenance, as well as for your continued business!

[swadmin]

This maintenance has been completed successfully.
The core routers have been upgraded to Cisco™ 6500-series equipment, and are fully operational as of earlier today.

There has been sporadic outages, due to our on-going maintenance/upgrade, longer than planned/expected, but this was an absolute necessity to upgrade/repair the network, to allow for future growth and stability/redundancy moving forward. We expect stability from here going forward, as the most technical/complicated portion of the upgrade has now been completed.

We will next be focusing on upgrades to the aggregation level switches, which will also be upgraded/replaced with Cisco™ 6500-series equipment. This should occur over the next 30-60 days, however, this should be minimally service affecting: We will be running the new aggregation switches simultaneously with the current/older switches. Each distribution switch will then be moved over to the new aggregation switches one at a time, several per day, with only a few moments of downtime per distribution switch as routes are updated and cables moved. The older aggregation switches will then be decommissioned once all distribution switches have been moved to the new aggregation switches.

We thank you for your patience and understanding with this necessary, but stressful, maintenance, as well as for your continued business!