cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
539
Views
0
Helpful
4
Replies

ACNS Kernel Crashes (and other ACNS Troubleshooting)

barry-peters
Level 1
Level 1

I have a CDN network that does primarily Live Web Broadcasting with pre-positioning of (large filesize) slides for training and product updates, etc.

I am receiving regular "high CPU / high Memory utilization" messages, as well as "kernel crash files amd / or user files detected" error messages. on 7325's and 510's (Root-CE's and remote Content Engines). When I dig further into the errors; I get "no WMT l4-switch enable" and other error messages. But I'm dead-ended as to where to find solutions to these problems.

Other than the times when we pre-position slides and run a live broadcast (once per week at most), the CE's should be dormant. Yet the high mem & CPU util messages constantly pop-up. We're using Accordant Presenter-Pro to create the broadcasts, and Jubilant to admin the CE's for pre-positioning and channel manipulation.

I can't find any troubleshooting documentation anywhere that tells me anything about these errors. Where do I look to get any clue as to why I would have high CPU / mem errors on devices that should be sitting idle?

4 Replies 4

ssoberlik
Level 4
Level 4

Since when are you getting these high memory utilization/ High CPU messages? Was this working fine previously. Was any changes done to the CDN network due to which you are seeing this messages .. like firmware upgrade or adding new slots

to the device ? understanding this will enable us get to the root cause of this issue.

I think I figured it out. We've been serving "live content",then archiving the events on the origin server. The more events we archive, the larger the content we need to replicate. Plus, the original CDN admin left the crawl depth to 10-levels. The Web server has been also used for additional static content as well.I think all this additional content on the origin server is causing the problem.

I did upgrade from ACNS 5.2 to 5.3; but the problemcropped-up before the upgrade.

Unfortunately, that was not the solution. I modified the acquisition rules to only include "static content" (.jpg, .gif, .js, .css, .htm, .xlm) The channel isn't failing anymore, however, I'm still getting high-CPU / Mem utils.

Any ideas?

Where are you getting the messages, on the console, the CDM GUI or in the Syslog?

I have a couple of thoughts you might want to consider...

1 - What is your warning level set to? You may be seeing informational messages. For instance, "No WMT L4 Switch enable" should be an "Informational or Notice" level message as it's just a message about your config. The Default for the system is "Warning" level.

2 - If you continue to have kernel crash files, I would open a TAC case to have them looked at by the experts.

Hope that helps...

Review Cisco Networking for a $25 gift card