Path Selection between 10 gig fiber and microwave

[asus zowey]

Hello everyone,

my network is running OSPF as an IGP, i have a 10 gig Ethernet  fiber connected between two sites and a microwave link as a redundant connection.
since ospf metric is cost ( or bandwidth ), the 10 gig ethernet connection is always preferred. however, sometimes the 10 gig link is flapping or the bit error rate is bad, is there anyway to change the path selection to go through the microwave when the bit error rate in the 10 gig link is bad or the link flaps ?

basically can we make the path selection based on anything than the speed or cost ?

[Joseph W. Doherty]

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

If the link is flapping, you might try IP dampening.

For dynamic path selection, look into OER/PfR.  (OSPF only has one cost metric.  EIGRP, I recall[?] might include link relalbility, but it might not be dynamic.)

[asus zowey]

i can't really change the routing protocol but i will look into oer/pfr.

[asus zowey]

if i start using oer/pfr, does that mean the b.w won't be the metric for path selection.

i am only worried that it starts choosing the microwave with less b.w over the 10 gig link because the microwave link has less delay for example. can i also include the b.w as a factor in oer/pfr??

[Joseph W. Doherty]

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

Bandwidth can be a metric to OER/PfR.  Much else can be used by OER/PfR.

The intent of this technology is sort of described by the names, Optimized Edge Routing (v1) and Performance Routing (v2).

Both can account for path bandwidth and/or analyze performance.

Understand typical dynamic routing protocols keep track of paths between source and destination and some have a way to "weight" paths  (for example, OSPF link cost [which by RFC, hasn't nothing to do with bandwidth, but is often based on that]).

OER/PfR, for example, can run their own SLA tests.

Years ago, I set up OER in large dual MPLS/VPN environment.  Our initial "problem", after activation, our WAN performance monitoring tools (and our users!) no longer "saw" any WAN performance issues.  They were still happening, but OER "saw" them first, and worked around them before the monitoring tools saw them.