Gradual Switch Migration 06-23-2003
Posted: Mon Jun 23, 2003 4:57 pm
We will slowly be migrating servers off the sw1, sw2, and sw3 switches to our new Cisco 5513 switch. The users on the afforementioned servers will also be routed through our new patching setup.
Users will experience 2-5 seconds of downtime per server during this migration, and we will be doing them at random, as we slowly move clients off the old 2900 series switches to our 5500 series core switch. Users who are moved, will retain their old bandwidth graphs, and will not loose any data in that respect, and will merely be informed VIA email of their new MRTG url.
Because the downtime per server is only the amount of time it takes to unplug one cable and plug in another, and connections (ssh, ftp, etc.) should not drop during this as both cables will already be in place, etc. we will not be providing additional warning of such activities. If 2-5 seconds of downtime is a problem for anyone, please let me know and we'll happilly schedule in a time to move your port specifically.
The new core switch has redundant supervisor engines, redundant power supplies, connected to redundant circuits, and is rated to have capacities for over 10,000,000 packets per second, and switching capabilities for over 50,000 mbps of traffic; In short, it can maintain 100mbps full duplex on every port without ever skipping a beat, and will not be outgrown for at least 5 years.
If anyone has any questions, feel free to ask!
.
Users will experience 2-5 seconds of downtime per server during this migration, and we will be doing them at random, as we slowly move clients off the old 2900 series switches to our 5500 series core switch. Users who are moved, will retain their old bandwidth graphs, and will not loose any data in that respect, and will merely be informed VIA email of their new MRTG url.
Because the downtime per server is only the amount of time it takes to unplug one cable and plug in another, and connections (ssh, ftp, etc.) should not drop during this as both cables will already be in place, etc. we will not be providing additional warning of such activities. If 2-5 seconds of downtime is a problem for anyone, please let me know and we'll happilly schedule in a time to move your port specifically.
The new core switch has redundant supervisor engines, redundant power supplies, connected to redundant circuits, and is rated to have capacities for over 10,000,000 packets per second, and switching capabilities for over 50,000 mbps of traffic; In short, it can maintain 100mbps full duplex on every port without ever skipping a beat, and will not be outgrown for at least 5 years.
If anyone has any questions, feel free to ask!
.