Hello All. We're using Spectrum Control to monitor our Brocade SAN. We already have started getting tons of alerts right away on various metrics, and some seems scary. And, even when we have customized the thresholds to very very high values, still we keep receiving those alerts all the time.
Kind of 3000+ emails in a day, and all the time. So we realized that is not good for us. So, we have created a 'Custom Switch Policy', but still we're getting lots of alerts.
Current Alerting Settings (apart from few more) -
a) Port Congestion Index for ISLs >=200 counts and such condition remains for 5 minutes.
b) Port Congestion Index for Ports >=200 counts and such condition remains for 5 minutes.
c) Port Congestion Index for Whole Switch >=500 counts and such condition remains for 10 minutes.
d) Zero Buffer Credit Rate for ISLs - 2 counts per second
e) Zero Buffer Credit Rate for Switch - 6000 counts per second
f) Credit Recovery Link Reset Rate - How many counts per second and such condition remains for 10 minutes.
g) Port Receive Bandwidth Percentage >= 75 % & < 85 %
h) Port Send Bandwidth Percentage >= 95 % and such condition remains for 5 minutes.
Just for an example, this is a Director Port, and Zero Buffer Credit Rate seen is 65000 counts per second, which is unbelievable that nothing went down even then. Same we see for 'Congestion' and luckily it is not leading to 'bottleneck' or slow drain etc. But the alerts seen seems dangerous and can cause issues soon 
Anyone using Spectrum Control for SAN Monitoring and how they are doing it and what kind of thresholds they have setup in their environment. Please advise.
Does, IBM has any document on Brocade SAN on recommended values to be set for Switches/Directors. Although that may vary from environment to environ. But there may be some baseline values or optimum values for these thresholds etc.
Thank You.