Re: TNS Ping Response spike during high cpu spikes

From: Jared Still <jkstill_at_gmail.com>
Date: Tue, 22 Nov 2016 17:27:11 -0800
Message-ID: <CAORjz=OSyfPy1UpWD8MYkwQLhf32ODZqokWDfkxGNVG0wSf8HA_at_mail.gmail.com>



On Thu, Nov 17, 2016 at 1:21 PM, Jeff Chirco <backseatdba_at_gmail.com> wrote:

> Lately I have been getting critical TNS Ping Response Time alerts from EM
> 12.1.0.5. The threshold is set to default of 1000ms. I have been noticing
> that I usually get them when the CPU on the server spikes 100% which makes
> since, if the CPU is busy then it can't respond as fast. Other than
> getting additional cpu power is there anything else I can do?
>
>

Hi Jeff,

In all sincerity, if it is not possible to add CPU, and not possible to reduce resource consumption, it seems there is little left to do than raise the threshold.

That said, were it me, I would investigate what is causing these CPU spikes.

Usually such things can be corrected, but determining why they happen is at least half the battle.

EE 11.2.0.4 on Windows 2008 R2 64bit
> Listener serves up 5 databases.
> EM stats show listener is processing 142 connections per minute but when I
> query the listener log I am seeing 225 entries in one minute, or 178 with a
> command of "establish". Not sure what the difference is.
>

Is this related to the previous question on CPU spikes?

Please take a few moments to consider it from the perspective of someone that knows little about the system in question (that would be us, on this list) and try to explain what is happening.

Jared Still
Certifiable Oracle DBA and Part Time Perl Evangelist Principal Consultant at Pythian
Pythian Blog http://www.pythian.com/blog/author/still/

--
http://www.freelists.org/webpage/oracle-l
Received on Wed Nov 23 2016 - 02:27:11 CET

Original text of this message