CCMSPING: Availability <100% in RZ20

[Former Member]

Hi veryone!

I know this could seem a silly question, but I can´t find a good explanation for that.

We have correctly installed the CCMSPING in our SOLMAN, so we collect centrally in the RZ20 of our SOLMAN availability data from all our Satellite Systems.

The question is that sometimes the Availability MTE which should be 100% for all the Satellite System in which there has been no interruption, I encounter values lower than that. For example 73%. 65%,etc...

I could understand that if the system has recently been down, this alert would show an average value, so it would not be nor 100% nor 0%. But the case is that the system are running OK for weeks, and the only explanation for not having a 100% of availability is that some network problems prevent the CCMSPING from collecting some values, and that could lower my percentage.But a network problem would affect not only the CCMSPING, but also the users, and nothing like this has been detected.

Has anybody else detected this?

regards

Antonio

[Former Member]

Hi Antonio,

did u have an clustered (Load Balancing) Landscape?

It could be possible if check the only hardware availability that the agent recognized that are some Load Balancing instance which were down ... inthat case the application runs well, only one instance for example (or Dialog Work processes) were down ... but this is also taking part in getting the availability of an server. If like to have only the availability of an application so choose Application availability.

This is an quite interesting issue, so please let me know if u can identify the root cause for that.

best regards

Lars

[Former Member]

Hi Lars!

I have this issue both for systems with load balancing and systems with onoly one instance.

Your guidelines have helped me a lot to better understand this question. The Availability entry in RZ20 for a Satellite System depends on the values related to its instance(s)

In one of my cases (development), I have a general entry for Availability for the System DHR, and another entry for each of its instances. I have only one instance

In the Open alerts, I have in the general availability for the system (it´s a pity I can´t upload a screenshot):

Availability , 73 % < 75 %

And under the Availability of its instance:

RFC_PING Availability , 67 % > 70 % Current value over threshold value

Instance RFC_PING Check , RFC_PING to instance SAX00010161_DHR_00 was not successfully performe

RFC error message

Dialog Response Time , 387 msec > 300 msec Dialog response time exceeds threshold (3 dialog steps)

So, I can see these values (RFC_PING, and even the Dialog Response Time (i have a low threshold in order to force the alert), have its effect under the general availability. So if the RFC_PINBG fails to connect (whatever the reason is, although the System is not down), this will affect the main value of the availabilty for all the system).

Thanks a lot for your guidelines

Bye!

[former_member215992]

Dear Lars,

  If you did the CCMSPING availability in your production system please let me know how you did? or please send the document to me.

Thanks,

Rao.ippili,

SAP Basis Team,