Did router reboot cause 5 hour outage to MPLS network
Client HQ lost contact to all branches. Note: Hub and Spoke network. Since this is a major problem for a financial services company I recommended they reboot the Level3 edge router. This is a 10 Meg MPLS connection to the MPLS cloud that connects to the 20 branches. In my limited and long time ago technical experience there should not have been any issue that would not be resolved after say 15 minutes.
Yes I have warned them that this is a single point of failure in the design.
Background - client is running BGP on a Cisco (1 year old) Core router behind the Level3 edgar router. The core router is running BGP and there are 2 other connections - both internet - 100 Mege TWT and 10 Meg Skyriver wireless.
It is interesting to me that Level3 would blame the router reboot as we were down before we rebooted the router. Client agrees with my logic and is not blaming me. But, what issues could I have caused.
Thanks
Yes I have warned them that this is a single point of failure in the design.
Background - client is running BGP on a Cisco (1 year old) Core router behind the Level3 edgar router. The core router is running BGP and there are 2 other connections - both internet - 100 Mege TWT and 10 Meg Skyriver wireless.
It is interesting to me that Level3 would blame the router reboot as we were down before we rebooted the router. Client agrees with my logic and is not blaming me. But, what issues could I have caused.
Thanks
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
There are two reasons that I can think of.
The first reason could be that running configuration was not saved to startup config. So, when router was reloaded it lost running configuration than later they configured router again (4 hours and 55 minutes later). :)
The second reason could be that they are running OSPF in their network and that, since it is hub and spoke topology (all areas must be attached to area 0), they have network design error where it is possible for one of spokes routers to become DR. When current DR (hub should be DR in hub and spoke OSPF topology, and there should be no BDR) was rebooted some spoke router become DR, and if DR is misplaced all communication was lost simply because there is no internal route updates.
The first reason could be that running configuration was not saved to startup config. So, when router was reloaded it lost running configuration than later they configured router again (4 hours and 55 minutes later). :)
The second reason could be that they are running OSPF in their network and that, since it is hub and spoke topology (all areas must be attached to area 0), they have network design error where it is possible for one of spokes routers to become DR. When current DR (hub should be DR in hub and spoke OSPF topology, and there should be no BDR) was rebooted some spoke router become DR, and if DR is misplaced all communication was lost simply because there is no internal route updates.
ASKER