@Jan Rolny wrote:
Hi,
I think it's still the same old well-known bug related to that "Hulc LED Process". There are tens of discussions on cisco forum about same problem for this models of switches 3560X/3750X. That time when I was working with 3750X I had exactly same issue.
https://bst.cisco.com/bugsearch/bug/CSCtn42790?rfs=qvlogin
Best regards,
Jan
Yes @Jan Rolny, the "Hulc LED Process" has been considered, by many, on-going "issue" for years, but even your reference describes it as NOT a bug, and as "expected behavior".
For @DarrelR it's not just the under 20% that process is consuming, but what else is driving his overall CPU to 90%+.
BTW, the "Hulc LED Process" issue is also explicitly mentioned (see table 3), as up 30% utilization, is considered normal in the Cisco troubleshooting article I referenced. (You reference has its CPU listed between 15-30%.)
Also BTW, something I "discovered" long ago, Cisco switches seem to have a relatively weak CPU, compared to similar, or even more expensive software based routers. At first, I was surprised because such switches have multi-gig capacity while software based routers might struggle at well under a gig of throughput. But, if you think about it, all "heavy lifting" of data forwarding, on a switch, is handled by dedicated hardware so their CPU, normally, doesn't have to be a real powerhouse.
Reason I mention the forgoing, so if Cisco decides to add some system process on a switch, like "Hulc LED Process", it might consume a fair amount of CPU possibly because the original design didn't expect and allow for the extra CPU usage demand. If this is a background level priority, again, even if such a process consumed all available CPU cycles, it would have very minimal impact to normal switch operations.
