I have a small network of 4 x 3750s running routing (OSPF) which contain all company firewalls. It bridges internal trusted corporate networks (depots etc) to external untrusted networks (Internet/Datacenter networks). It sees a fairly high level of traffic running through it (it gets forwarded to firewalls at gig speed).
OSPF is costed to remove asymetric routing so one 3750 is acting as primary and does most of the work. The 3750s are in area 0, and all other areas are summarised to some degree. However, CPU is running high on the 3750 when looking at the history at times:
PPFX_X37_274#sh ip ro summ
IP routing table name is Default-IP-Routing-Table(0)
My question is - although the average is around 26% the 72hour picture would suggest that CPU is running at 100% at times. Is this an imminent concern that needs re-designing, or is this normal behaviour (SPF running for instance) that is manageable?
Can you also have a look at show processes cpu sorted ? Try out also 1min and 5min keywords to see what processes tend to create a long-standing load on your CPU. This command should place the most CPU-intensive tasks at the top of the listing so you can see what process is consuming most of your CPU time.
Having a load in the range of 26% is, in my personal opinion, nothing to be especially concerned about, but I do not quite like the spikes up to 100%. They need more investigation indeed.
This may look like trying to solve your issue quickly but I really recommend going through the following documents:
We are pleased to announce availability of Beta software for 16.6.3. 16.6.3 will be the second rebuild on the 16.6 release train targeted towards Catalyst 9500/9400/9300/3850/3650 switching platforms. We are looking for early feedback from custome...