Management of spike rate value ​​on the same proxy deployed on different orgs

Hello,

We have several Apigee Hybrid clusters (two orgs: one for testing, one for production).
Obviously the spikes in production have higher thresholds.
How is it better, also in terms of reading performance, to parameterize this rate value in the SpikeArrest Policy? Property Sets? Are these values ​​cached?

For example

 

test_rate=10pm
prod_rate=500pm

 

If I had to make some considerations, neither way convinces me:

  • Proxy Scoped Property Set:
    • Usually traffic peaks can vary in periods of business initiatives and redeploying a proxy to change only the proxy property set seems excessive to me.
    • This mode forces you to have logics to understand if you want to take the test value or the production value from the property set
    • However, I think it's convenient to manage attaching configurations directly inside the proxy, especially when you have many proxies.
  • Env Scoped Property Set:
    • you can modify them at runtime without having to modify the proxies, however as proxies grow they become difficult to maintain and with modifications you risk introducing regressions on the behavior of other proxies.

In my opinion it would not be bad to have these values ​​directly in the environment configurations (as for the target servers), populating the property sets every time is inconvenient and can introduce regressions in the modification.

What do you usually recommend to deal with this issue?

Thanks

0 0 64
0 REPLIES 0