High Disk I/O resulting in high CPU load

anerimasi
Participant I

Hello,

We have an18.0.1 Apigee Edge cluster where we are seeing high disk i/o which is resulting in high CPU loads. We are trying to investigate this, and are wondering if log level of INFO could be the issue.

So a couple questions

1) Has anyone else noticed this behavior? If yes, what did you do to resolve it?

2) We are trying to change the log levels for nginx access logs to WARN, but looks like there is no straightforward way to do it. I do see this in nginx documentation, but want to check here if there is anything simpler

https://docs.nginx.com/nginx/admin-guide/monitoring/logging/#enabling-conditional-logging

Thanks!

Kashmira

0 3 811
3 REPLIES 3

remeeshnair
Participant IV

@Kashmira Patel We are going to upgrade to 18.01 soon, did you get resolution for this problem? Do you suspect any particular component logs doing this?

Regards,

Remeesh

Hi Remeesh,

Turns out disk i/o was not the issue, it was a hardware problem on our end. Default log levels are ok. But I did successfully change access log level from default to warn using the instructions i mention above.

HTH,

Kashmira

Good to know it resolved for you !