C3850 Problems with Rapidly Increasing "Out Of Discard"

[JUNHOLEE95720]

hi

This is our problem.

White paper - "C3850-12S" - "2960X"

However, the 2960X is in a situation where it is hanging and we found a large number of "out of discards" per port in the C3850

1. How can I do this?

2. Is this the easiest way or is there a command?

3. Are there any caveats when "2" applying?

I attached the log, so I can refer to it.

Thanks

[balaji.bandi]

you need to check what kind of version running both the side to look for bugs.

i see some 3850 having bugs similar kind of issue

https://bst.cloudapps.cisco.com/bugsearch/bug/CSCvb65304/?reffering_site=dumpcr

[Georg Pauwen · Accepted solution]

Hello,

post the output of:

show platform port-asic stat drop GigabitEthernet1/0/1

You could have oversubscribed interfaces, so some sort of buffer tuning might help...

[JUNHOLEE95720]

Hi

What commands are available for buffer tuning?

Thanks

[Leo Laohoo]

Try this:  

qos queue-softmax-multiplier 1200